U.S. patent application number 16/132195 was filed with the patent office on 2019-03-14 for microorganism-related significance index metrics.
The applicant listed for this patent is uBiome, Inc.. Invention is credited to Daniel Almonacid, Zachary Apte, Ricardo Castro, Patricio Lagos, Jessica Richman, Paz Tapia.
Application Number | 20190080046 16/132195 |
Document ID | / |
Family ID | 63714165 |
Filed Date | 2019-03-14 |
![](/patent/app/20190080046/US20190080046A1-20190314-D00000.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00001.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00002.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00003.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00004.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00005.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00006.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00007.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00008.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00009.png)
![](/patent/app/20190080046/US20190080046A1-20190314-D00010.png)
View All Diagrams
United States Patent
Application |
20190080046 |
Kind Code |
A1 |
Apte; Zachary ; et
al. |
March 14, 2019 |
MICROORGANISM-RELATED SIGNIFICANCE INDEX METRICS
Abstract
Embodiments of a method and/or system, such as for
characterizing at least one microorganism-related condition, can
include: determining a set of associations (e.g., positive
associations such as positive correlations, negative associations
such as negative correlations, non-associations such as no
correlation or minimal correlation, etc.) between a set of
microorganism taxa and at least one microorganism-related
condition; determining a set of reference features (e.g., reference
abundance ranges, etc.) for the set of microorganism taxa; and
determining one or more significance index metrics based on the set
of associations and the set of reference features.
Inventors: |
Apte; Zachary; (San
Francisco, CA) ; Richman; Jessica; (San Francisco,
CA) ; Almonacid; Daniel; (San Francisco, CA) ;
Lagos; Patricio; (Santiago, CL) ; Castro;
Ricardo; (Santiago, CL) ; Tapia; Paz;
(Santiago, CL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
uBiome, Inc. |
San Francisco |
CA |
US |
|
|
Family ID: |
63714165 |
Appl. No.: |
16/132195 |
Filed: |
September 14, 2018 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62558489 |
Sep 14, 2017 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G01N 33/98 20130101;
G16H 50/30 20180101; C12Q 1/689 20130101; G16B 20/00 20190201; G16B
40/00 20190201; C12Q 2600/112 20130101; G16H 50/20 20180101 |
International
Class: |
G06F 19/18 20060101
G06F019/18; G06F 19/24 20060101 G06F019/24; G16H 50/20 20060101
G16H050/20; G16H 50/30 20060101 G16H050/30 |
Claims
1. A method for characterizing at least one microorganism-related
condition, the method comprising: determining a set of associations
between a set of microorganism taxa and the at least one
microorganism-related condition, wherein the set of associations
comprises at least one of positive associations, negative
associations, and non-associations; determining a set of reference
abundance ranges for the set of microorganism taxa, wherein the
reference abundance ranges are associated with the at least one
microorganism-related condition; and determining a significance
index metric associated with characterization of the set of
associations between the set of microorganism taxa and the at least
one microorganism-related condition, based on the set of
associations and the reference abundance ranges for the set of
microorganism taxa.
2. The method of claim 1, wherein determining the significance
index metric comprises: determining effect size metrics for the set
of associations between the set of microorganism taxa and the at
least one microorganism-related condition, based on the set of
reference abundance ranges for the set of microorganism taxa; and
determining the significance index metric based on the effect size
metrics.
3. The method of claim 2, wherein determining the effect size
metrics comprises determining a set of coefficient of correlations
for the set of associations between the set of microorganism taxa
and the at least one microorganism-related condition, based on a
meta-analysis, and wherein determining the significance index
metric based on the effect size metrics comprises determining the
significance index metric based on the set of coefficient of
correlations.
4. The method of claim 2, wherein determining the effect size
metrics comprises determining a set of z-scores for the set of
associations between the set of microorganism taxa and the at least
one microorganism-related condition; and modifying the set of
z-scores based on the reference abundance ranges for the set of
microorganism taxa and at least one of the positive associations
and the negative associations between the set of microorganism taxa
and the at least one microorganism-related condition; and wherein
determining the significance index metric based on the effect size
metrics comprises determining the significance index metric based
on the modified set of z-scores.
5. The method of claim 2, wherein determining the effect size
metrics comprises performing an interpolation process based on the
reference abundance ranges and a calibration curve derived from a
random set of abundances for the set of microorganism taxa, and
wherein determining the significance index metric based on the
effect size metrics comprises determining the significance index
metric based on the interpolation process.
6. The method of claim 2, wherein determining the significance
index metric comprises determining a propensity score for a user
describing an association between a user microbiome and the at
least one microorganism-related condition, based on the effect size
metrics and user abundances for the set of microorganism taxa.
7. The method of claim 6, wherein determining the significance
index metric comprises normalizing the propensity score based on a
set of empirical abundance ranges for the set of microorganism
taxa.
8. The method of claim 1, wherein determining the significance
index metric comprises: determining a set of labels for a user
sample, wherein determining the set of labels comprises determining
a label of the set of labels for a taxon of the set of
microorganism taxa based on satisfaction of an abundance condition
by a user abundance for the taxon in relation to a reference
abundance range for the taxon, and satisfaction of an association
type condition by an association, of the set of associations,
between the taxon and the at least one microorganism-related
condition, and determining the significance index metric for a user
associated with the user sample, based on the set of labels.
9. The method of claim 1, wherein determining the significance
index metric comprises determining a microorganism-related
condition classification associated with a health state of a user
for the at least one microorganism-related condition, based on user
microbiome composition features and a machine learning model
derived from the set of associations and the set of reference
abundance ranges.
10. The method of claim 1, further comprising facilitating
diagnosis of the user for the at least one microorganism-related
condition based on the significance index metric and a user sample
comprising microorganisms associated with the set of microorganism
taxa.
11. The method of claim 1, further comprising facilitating
therapeutic intervention for the user for the at least one
microorganism-related condition based on the significance index
metric and a user sample comprising microorganisms associated with
the set of microorganism taxa.
12. A method for characterizing at least one microorganism-related
condition in relation to a user, the method comprising: collecting
a sample from a user, wherein the sample comprises microorganisms
associated with the at least one microorganism-related condition;
determining user microbiome composition features associated with
the microorganisms, based on the sample; and determining, for the
user, a significance index metric characterizing an association
between a user microbiome and the at least one
microorganism-related condition, based on the user microbiome
composition features, reference microbiome composition features
associated with a set of microorganism taxa, and a set of
associations between the set of microorganism taxa and the at least
one microorganism-related condition.
13. The method of claim 12, wherein determining the significance
index metric for the user comprises determining the significance
index metric based on the user microbiome composition features and
a set of coefficient of correlations for the set of associations
between the set of microorganism taxa and the at least one
microorganism-related condition.
14. The method of claim 12, wherein determining the significance
index metric for the user comprises determining the significance
index metric based on the user microbiome composition features and
a set of modified z-scores determined based on the reference
microbiome composition features and a set of z-scores for the set
of associations between the set of microorganism taxa and the at
least one microorganism-related condition.
15. The method of claim 12, wherein determining the significance
index metric for the user comprises determining the significance
index metric based on the user microbiome composition features and
an interpolation process with the reference microbiome composition
features and a calibration curve derived from a random set of
abundances for the set of microorganism taxa.
16. The method of claim 12, wherein the user microbiome composition
features comprise user abundances for the set of microorganism
taxa, wherein the reference microbiome composition features
comprise reference abundance ranges for the set of microorganism
taxa, and wherein determining the significance index metric
comprises determining a propensity score for the user
characterizing the association between the user microbiome and the
at least one microorganism-related condition, based on the user
abundances and effect size metrics determined based on the
reference abundance ranges and the set of associations between the
set of microorganism taxa and the at least one
microorganism-related condition.
17. The method of claim 16, wherein determining the propensity
score comprises determining the propensity score based on the user
abundances, the effect size metrics, and significance metrics for
the effect sizes.
18. The method of claim 16, wherein determining the significance
index metric comprises normalizing the propensity score based on a
set of empirical abundance ranges for the set of microorganism
taxa.
19. The method of claim 12, wherein determining the significance
index metric comprises: determining a set of labels for the sample,
wherein determining the set of labels comprises determining a label
of the set of labels for a taxon of the set of microorganism taxa
based on satisfaction of an abundance condition by a user abundance
for the taxon in relation to a reference abundance range for the
taxon, and satisfaction of an association type condition by an
association, of the set of associations, between the taxon and the
at least one microorganism-related condition, and determining the
significance index metric for the user based on the set of
labels.
20. The method of claim 12, wherein determining the significance
index metric comprises determining a microorganism-related
condition classification associated with a health state of the user
for the at least one microorganism-related condition, based on the
user microbiome composition features and a machine learning model
derived from the set of associations and the set of reference
microbiome composition features.
21. The method of claim 20, wherein determining the
microorganism-related condition classification comprises
determining a diet-related condition classification, associated
with a diet-related condition, for the user based on the machine
learning model and the user microbiome composition features
associated with the set of microorganism taxa.
22. The method of claim 21, wherein the diet-related condition
comprises at least one of caffeine consumption, alcohol
consumption, artificial sweetener consumption, and sugar
consumption; wherein determining the microorganism-related
condition classification comprises determining at least one of a
caffeine consumption classification, an alcohol consumption
classification, an artificial sweetener consumption classification,
and a sugar consumption classification, for the user based on the
machine learning model and the user microbiome composition features
associated with the set of microorganism taxa, wherein the set of
microorganism taxa comprises at least one of: Alistipes;
Anaerotruncus; Bacteroides; Bifidobacterium; Bilophila; Blautia;
Butyricimonas; Clostridium; Collinsella; Erysipelatoclostridium;
Faecalibacterium; Flavobacterium; Flavonifractor; Granulicatella;
Hespellia; Intestinimonas; Kluyvera; Lachnospira; Marvinbryantia;
Odoribacter; Oscillibacter; Parabacteroides; Phascolarctobacterium;
Pseudobutyrivibrio; Roseburia; Streptococcus; Subdoligranulum;
Sutterella; and Terrisporobacter.
23. The method of claim 12, further comprising facilitating
diagnosis of the user for the at least one microorganism-related
condition based on the significance index metric.
24. The method of claim 12, further comprising facilitating
therapeutic intervention for the user for the at least one
microorganism-related condition based on the significance index
metric.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional
Application Ser. No. 62/558,489 filed 14 Sep. 2017, which is
incorporated in its entirety herein by this reference.
TECHNICAL FIELD
[0002] The disclosure generally relates to genomics and
microbiology.
BACKGROUND
[0003] A microbiome can include an ecological community of
commensal, symbiotic, and pathogenic microorganisms that are
associated with an organism. Characterization of the human
microbiome is a complex process. The human microbiome includes over
10 times more microbial cells than human cells, but
characterization of the human microbiome is still in nascent stages
such as due to limitations in sample processing techniques, genetic
analysis techniques, and resources for processing large amounts of
data. Present knowledge has clearly established the role of
microbiome associations with multiple health conditions, and has
become an increasingly appreciated mediator of host genetic and
environmental factors on human disease development. The microbiome
is suspected to play at least a partial role in a number of
health/disease-related states. Further, the microbiome may mediate
effects of environmental factors on human, plant, and/or animal
health. Given the profound implications of the microbiome in
affecting a user's health, efforts related to the characterization
of the microbiome should be pursued. However, conventional
approaches for analyzing microbiomes, such as in relation to one or
more microbiome-related conditions have left many questions
unanswered.
BRIEF DESCRIPTION OF THE FIGURES
[0004] FIGS. 1-2 include flowchart representations of variations of
an embodiment of a method;
[0005] FIG. 3 includes variations of embodiments of a system;
[0006] FIG. 4 includes flowchart representations of variations of
an embodiment of a method;
[0007] FIG. 5 includes graph representations of significance index
metric frequencies in variations of an embodiment of a method;
[0008] FIG. 6 includes a graph representation of metrics for a
Crohn's disease prediction machine learning model in a variation of
an embodiment of a method;
[0009] FIGS. 7A-7E include graph representations of metrics for
caffeine consumer prediction machine learning models in a variation
of an embodiment of a method;
[0010] FIGS. 8A-8D include specific examples of notifications
including significance index metrics;
[0011] FIG. 9 includes a schematic representation of variations of
an embodiment of the method;
[0012] FIG. 10 includes variations of determining significance
index metrics with one or more models;
[0013] FIG. 11 includes facilitating therapeutic intervention in a
variation of an embodiment of a method;
[0014] FIG. 12 includes a schematic representation of variations of
an embodiment of the method;
[0015] FIG. 13 includes a schematic representation of variations of
an embodiment of the method;
[0016] FIG. 14 includes a schematic representation of variations of
an embodiment of the method;
[0017] FIG. 15 includes a schematic representation of variations of
an embodiment of the method;
[0018] FIG. 16 includes a schematic representation of variations of
an embodiment of the method.
DESCRIPTION OF THE EMBODIMENTS
[0019] The following description of the embodiments is not intended
to limit the embodiments, but rather to enable any person skilled
in the art to make and use.
1. OVERVIEW
[0020] As shown in FIGS. 1-2, 4, and 9, embodiments of a method 100
(e.g., for characterizing at least one microorganism-related
condition, etc.) can include: determining a set of associations
(e.g., positive associations such as positive correlations,
negative associations such as negative correlations,
non-associations such as no correlation or minimal correlation,
etc.) between a set of microorganism taxa and at least one
microorganism-related condition S110; determining a set of
reference features (e.g., reference abundance ranges, etc.) for the
set of microorganism taxa S120; and determining one or more
significance index metrics based on the set of associations and the
set of reference features S130.
[0021] In a specific example, the method 100 (e.g., for
characterizing at least one microorganism-related condition, etc.)
can include: determining a set of associations between a set of
microorganism taxa and the at least one microorganism-related
condition, where the set of associations includes at least one of
positive associations, negative associations, and non-associations;
associated with the at least one microorganism-related condition;
determining a set of reference abundance ranges for the set of
microorganism taxa, where the reference abundance ranges are
associated with the at least one microorganism-related condition;
and/or determining a significance index metric associated with
characterization of the set of associations between the set of
microorganism taxa and the at least one microorganism-related
condition, based on the set of associations and the reference
abundance ranges for the set of microorganism taxa.
[0022] Embodiments of the method 100 can additionally or
alternatively include one or more of: facilitating diagnosis of one
or more microorganism-related conditions based on the one or more
significance index metrics S140; facilitating therapeutic
intervention for the one or more microorganism-related conditions
based on the one or more significance index metrics S150; and/or
any other suitable processes.
[0023] Embodiments of the method 100 and/or system 200 can function
to determine one or more metrics (e.g., significance index metrics)
characterizing associations between one or more taxa and one or
more microorganism-related condition (e.g., where the metrics can
provide an objective measurement of the association between a
combination of microorganisms and one or more microorganism-related
conditions; etc.). In specific examples, significance index metrics
can be used for characterizing one or more users (e.g., based on
novel user samples from the users; etc.); facilitating diagnosis;
facilitating therapeutic intervention; uncovering insights
regarding the relationship between one or more taxa and one or more
microorganism-related conditions; and/or confer any other suitable
benefits.
[0024] In a specific example, the method 100 (e.g., for
characterizing at least one microorganism-related condition in
relation to a user, etc.) can include: collecting a sample from a
user, where the sample includes microorganisms associated with the
at least one microorganism-related condition; determining user
microbiome composition features associated with the microorganisms,
based on the sample; and/or determining, for the user, a
significance index metric characterizing an association between a
user microbiome and the at least one microorganism-related
condition, based on the user microbiome composition features (e.g.,
user abundances for the set of microorganism taxa; etc.), reference
microbiome composition features (e.g., reference abundances and/or
abundance ranges for the set of microorganism taxa; etc.)
associated with a set of microorganism taxa, and a set of
associations between the set of microorganism taxa and the at least
one microorganism-related condition. In variations, collecting a
sample, determining microbiome features (e.g., microbiome
composition features), and/or suitable processes of embodiments of
the method 100 can be performed in any suitable manner described in
and/or analogous to U.S. application Ser. No. 16/115,542 filed 28
Aug. 2018 and/or U.S. application Ser. No. 16/047,840 filed 27 Jul.
2018, which are herein incorporated in their entireties by this
reference.
[0025] Additionally or alternatively, embodiments of the method 100
and/or system 200 can function to identify microbiome features,
supplemental features (e.g., derived from supplemental data, etc.),
and/or other suitable data associated with (e.g., positively
correlated with, negatively correlated with, etc.) one or more
microorganism-related conditions, such as for use in determining
significance indexes, for use as biomarkers (e.g., for diagnostic
processes, for treatment processes, etc.), for use in diagnostics
and/or therapeutics, and/or for other suitable purposes. In
examples, microorganism-related conditions (and/or significance
indexes and/or other suitable aspects) can be associated with at
least one or more of microbiome composition (e.g., microbiome
composition diversity, etc.), microbiome function (e.g., microbiome
functional diversity, etc.), and/or other suitable
microbiome-related aspects.
[0026] Additionally or alternatively, embodiments of the method 100
and/or system 200 can function to determine one or more metrics
(e.g., significance index metrics) for a panel of
microorganism-related conditions (e.g., a panel categorized by
condition type; etc.), such as in relation to characterizing a
plurality of associations between a plurality of taxa and a
plurality of microorganism-related conditions (e.g., where any
number of taxa can be associated with any number of
microorganism-related conditions in any suitable numerical
relationship; etc.). Additionally or alternatively, embodiments can
perform any suitable functionality described herein.
[0027] In variations, data from populations of users (e.g.,
populations of subjects associated with one or more
microorganism-related conditions; positively correlated, negatively
correlated, not correlated, with one or more microorganism-related
conditions; data derived from information sources such as
scientific peer-reviewed articles; etc.) can be used to determine
significance index metrics, such as for characterizing subsequent
users, such as for indicating microorganism-related states of
health and/or areas of improvement (e.g., for diagnostic purposes,
etc.), and/or to facilitate therapeutic intervention (e.g.,
promoting one or more therapies; facilitating modulation of the
composition and/or functional diversity of a user's microbiome
toward one or more of a set of desired equilibrium states, such as
states correlated with improved health states associated with one
or more microorganism-related conditions; etc.), such as in
relation to one or more microorganism-related conditions.
Variations of the method 100 can further facilitate selection,
monitoring (e.g., efficacy monitoring, etc.) and/or adjusting of
therapies provided to a user, such as through collection and
analysis (e.g., with significance index models) of additional
samples from a user over time (e.g., throughout the course of a
therapy regimen, through the extent of a user's experiences with
microorganism-related conditions; as shown in FIG. 13; etc.),
across body sites (e.g., across sample collection sites of a user,
such as collection sites corresponding to a particular body site
type such as a nose site, gut site, mouth site, skin site, genital
site; etc.), in addition or alternative to processing supplementary
data over time, such as for one or more microorganism-related
conditions. However, data from populations, subgroups, individuals,
and/or other suitable entities can be used by any suitable portions
of embodiments of the method 100 and/or system 200 for any suitable
purpose.
[0028] In variations, embodiments of the method 100 and/or system
200 can determine significance index metrics for determining one or
more microorganism-related characterizations and/or therapies
associated with one or more microorganism-related conditions, such
as characterizations and/or therapies described in U.S. application
Ser. No. 16/047,840 filed 27 Jul. 2018, which is herein
incorporated in its entirety by this reference.
[0029] Embodiments of the method 100 and/or system 200 can
additionally or alternatively generate and/or promote (e.g.,
provide; present; notify regarding; etc.) characterizations (e.g.,
diagnoses, etc.) and/or therapies for one or more
microorganism-related conditions.
[0030] Microorganism-related conditions can include one or more of:
diseases, symptoms, causes (e.g., triggers, etc.), disorders,
associated risk (e.g., propensity scores, etc.), associated
severity, behaviors (e.g., caffeine consumption, alcohol
consumption, sugar consumption, habits, diets, etc.), and/or any
other suitable aspects associated with microorganism-related
conditions. Microorganism-related conditions can include one or
more disease-related conditions, which can include any one or more
of: gastrointestinal-related conditions (e.g., irritable bowel
syndrome, inflammatory bowel disease, ulcerative colitis, celiac
disease, Crohn's disease, bloating, hemorrhoidal disease,
constipation, reflux, bloody stool, diarrhea, etc.);
allergy-related conditions (e.g., allergies and/or intolerance
associated with wheat, gluten, dairy, soy, peanut, shellfish, tree
nut, egg, etc.); locomotor-related conditions (e.g., gout,
rheumatoid arthritis, osteoarthritis, reactive arthritis, multiple
sclerosis, Parkinson's disease, etc.); cancer-related conditions
(e.g., lymphoma; leukemia; blastoma; germ cell tumor; carcinoma;
sarcoma; breast cancer; prostate cancer; basal cell cancer; skin
cancer; colon cancer; lung cancer; cancer conditions associated
with any suitable physiological region; etc.);
cardiovascular-related conditions (e.g., coronary heart disease,
inflammatory heart disease, valvular heart disease, obesity,
stroke, etc.); anemia conditions (e.g., thalassemia; sickle cell;
pernicious; fanconi; haemolyitic; aplastic; iron deficiency; etc.);
neurological-related conditions (e.g., ADHD, ADD, anxiety,
Asperger's syndrome, autism, chronic fatigue syndrome, depression,
etc.); autoimmune-related conditions (e.g., Sprue, AIDS, Sjogren's,
Lupus, etc.); endocrine-related conditions (e.g., obesity, Graves'
disease, Hashimoto's thyroiditis, metabolic disease, Type I
diabetes, Type II diabetes, etc.); skin-related conditions (e.g.,
acne, dermatomyositis, eczema, rosacea, dry skin, psoriasis,
dandruff, photosensitivity, rough skin, itching, flaking, scaling,
peeling, fine lines or cracks, gray skin in individuals with dark
skin, redness, deep cracks such as cracks that can bleed and lead
to infections, itching and scaling of the skin in the scalp, oily
skin such as irritated oily skin, skin sensitivity to products such
as hair care products, imbalance in scalp microbiome, etc.); Lyme
disease conditions; communication-related conditions; sleep-related
conditions; metabolic-related conditions; weight-related
conditions; pain-related conditions; genetic-related conditions;
chronic disease; and/or any other suitable type of disease-related
conditions.
[0031] In variations, microorganism-related conditions can include
one or more women's health-related conditions (e.g., reproductive
system-related conditions; etc.) described in U.S. application Ser.
No. 16/115,542 filed 28 Aug. 2018, which is herein incorporated in
its entirety by this references, such as where significance index
metrics can be determined and/or used for one or more women's
health-related conditions and/or other suitable
microorganism-related conditions.
[0032] Additionally or alternatively, microorganism-related
conditions can include one or more human behavior conditions which
can include any one or more of: diet-related conditions (e.g.,
caffeine consumption, alcohol consumption, sugar consumption,
artificial sweetener consumption, omnivorous, vegetarian, vegan,
sugar consumption, acid consumption other food item consumption,
dietary supplement consumption, dietary behaviors, etc.),
probiotic-related behaviors (e.g., consumption, avoidance, etc.),
habituary behaviors (e.g., smoking; exercise conditions such as
low, moderate, and/or extreme exercise conditions; etc.),
menopause, other biological processes, social behavior, other
behaviors, and/or any other suitable human behavior conditions.
Conditions can be associated with any suitable phenotypes (e.g.,
phenotypes measurable for a human, animal, plant, fungi body,
etc.). In variations, portions of embodiments of the method 100
and/or system 200 can be used for facilitating promoting (e.g.,
providing; recommending; etc.) of one or more targeted therapies to
users suffering from one or more microorganism-related conditions
(e.g., skin-related conditions, etc.), such as based on one or more
significance index metrics.
[0033] In variations, samples (e.g., described herein) can
correspond to a one or more collection sites including at least one
of a gut collection site (e.g., corresponding to a body site type
of a gut site), a skin collection site (e.g., corresponding to a
body site type of a skin site), a nose collection site (e.g.,
corresponding to a body site type of a nose site), a mouth
collection site (e.g., corresponding to a body site type of a mouth
site), and a genitals collection site (e.g., corresponding to a
body site type of a genital site).
[0034] Embodiments of the method 100 and/or system 200 can be
implemented for a single user, such as in relation to applying one
or more sample handling processes and/or significance index
determination processes for processing one or more biological
samples (e.g., collected across one or more collection sites, etc.)
from the user for determining a significance index metric for the
user, for microbiome-related characterization, facilitating
therapeutic intervention, and/or for any other suitable purpose.
Additionally or alternatively, embodiments can be implemented for a
population of subjects (e.g., including the user, excluding the
user), where the population of subjects can include subjects
similar to and/or dissimilar to any other subjects for any suitable
type of characteristics (e.g., in relation to microorganism-related
conditions, demographic characteristics, behaviors, microbiome
composition and/or function, etc.); implemented for a subgroup of
users (e.g., sharing characteristics, such as characteristics
affecting microorganism-related characterization and/or therapy
determination; etc.); implemented for plants, animals,
microorganisms, and/or any other suitable entities. Thus,
information derived from a set of subjects (e.g., population of
subjects, set of subjects, subgroup of users, etc.) can be used to
provide additional insight for subsequent users. In a variation, an
aggregate set of biological samples is associated with and
processed for a wide variety of subjects, such as including
subjects of one or more of: different demographic characteristics
(e.g., genders, ages, marital statuses, ethnicities, nationalities,
socioeconomic statuses, sexual orientations, etc.), different
microorganism-related conditions (e.g., health and disease states;
different genetic dispositions; etc.), different living situations
(e.g., living alone, living with pets, living with a significant
other, living with children, etc.), different dietary habits (e.g.,
omnivorous, vegetarian, vegan, sugar consumption, acid consumption,
caffeine consumption, etc.), different behavioral tendencies (e.g.,
levels of physical activity, drug use, alcohol use, etc.),
different levels of mobility (e.g., related to distance traveled
within a given time period), and/or any other suitable
characteristic (e.g., characteristics influencing, correlated with,
and/or otherwise associated with microbiome composition and/or
function, etc.). In examples, as the number of subjects increases,
the predictive power of processes implemented in portions of
embodiments of the method 100 and/or system 200 can increase, such
as in relation to characterizing subsequent users (e.g., with
varying characteristics, etc.) based upon their microbiomes (e.g.,
in relation to different collection sites for samples for the
users, etc.). However, portions of embodiments of the method 100
and/or system 200 can be performed and/or configured in any
suitable manner for any suitable entity or entities.
[0035] In variations, portions of embodiments of the method 100 can
be repeatedly performed in any suitable order and/or any suitable
components of embodiments of the system 200 can be repeatedly
applied, such as to improve any suitable portions of embodiments of
the method 100 and/or any suitable components of embodiments of the
system 200. In variations, the method 100 can be repeatedly
performed to enable refining of one or more microorganism-related
databases (e.g., including associations between microorganism taxa
and microorganism-related conditions; including effect size
metrics; including reference microbiome features such as reference
abundance ranges; etc.), such as by collecting and analyzing
additional information sources, samples (e.g., such as samples
collected from subjects over time, the course of one or more
microorganism-related conditions, and/or therapeutic interventions;
etc.), and/or other suitable components. In variations, the method
100 can include refining processes for determining significance
index metrics, such as for improving accuracy and/or other suitable
aspects associated with significance index metrics.
[0036] Data described herein (e.g., significance index metrics,
effect size metrics, taxa identifiers, associations, microbiome
features, user features, reference features, microorganism
datasets, models, microorganism-related characterizations,
supplementary data, notifications, etc.) can be associated with any
suitable temporal indicators (e.g., seconds, minutes, hours, days,
weeks, months, years, etc.) including one or more: temporal
indicators indicating when the data was collected (e.g., temporal
indicators indicating when a sample was collected; sampling time;
etc.), determined, transmitted, received, and/or otherwise
processed; temporal indicators providing context to content
described by the data (e.g., temporal indicators associated with
significance index metrics, etc.); changes in temporal indicators
(e.g., changes in microbiome over time; such as in response to
receiving a therapy; latency between sample collection, sample
analysis, provision of a microorganism-related characterization or
therapy to a user, and/or other suitable portions of embodiments of
the method 100; etc.); and/or any other suitable indicators related
to time.
[0037] Additionally or alternatively, parameters, metrics, inputs,
outputs, and/or other suitable data can be associated with value
types including: scores (e.g., propensity scores; feature relevance
scores; correlation scores; covariance scores; microbiome diversity
scores; severity scores; etc.); individual values (e.g., individual
microorganism-related condition scores, such as condition
propensity scores for different conditions, for different
collection sites, etc.), aggregate values, (e.g., overall scores
based on individual microorganism-related scores for different
conditions, collection sites, taxa; etc.), binary values (e.g.,
classifications of a health sample or a sample presenting a
microorganism-related condition; etc.), relative values (e.g.,
relative taxonomic group abundance, relative microbiome function
abundance, relative feature abundance, etc.), classifications
(e.g., microorganism-related condition classifications and/or
diagnoses for users; feature classifications; behavior
classifications; demographic characteristic classifications; etc.),
confidence levels (e.g., associated with significant index metrics
and/or other suitable data; etc.), identifiers, values along a
spectrum, and/or any other suitable types of values. Any suitable
types of data described herein can be used as inputs (e.g., for
different analytical techniques, models, and/or other suitable
components described herein), generated as outputs (e.g., of
different analytical techniques, models, etc.), and/or manipulated
in any suitable manner for any suitable components associated with
the method 100 and/or system 200.
[0038] One or more instances and/or portions of embodiments of the
method 100 and/or processes described herein can be performed
asynchronously (e.g., sequentially), concurrently (e.g., parallel
data processing; concurrent cross-condition analysis; multiplex
sample processing; performing sample processing and analysis for
substantially concurrently evaluating a panel of
microorganism-related conditions; computationally determining
significance index metrics, microorganism datasets, microbiome
features, and/or characterizing microorganism-related conditions in
parallel for a plurality of users; such as concurrently on
different threads for parallel computing to improve system
processing ability; etc.), in temporal relation (e.g.,
substantially concurrently with, in response to, serially, prior
to, subsequent to, etc.) to a trigger event (e.g., performance of a
portion of the method 100), and/or in any other suitable order at
any suitable time and frequency by and/or using one or more
instances of the system 200, components, and/or entities described
herein.
[0039] Additionally or alternatively, embodiments of the method 100
and/or system 200 can perform any suitable sample processing
operations described in U.S. application Ser. No. 16/115,542 filed
28 Aug. 2018, such as for determining microorganism datasets and/or
microbiome features usable in determining one or more significance
index metrics. For example, embodiments of the method 100 and/or
system 200 can generate microorganism sequence datasets and/or
other suitable microorganism data based on applying one or more
sequencing systems 215 (e.g., next-generation sequencing systems,
sequencing systems for targeted amplicon sequencing,
sequencing-by-synthesis techniques, capillary sequencing technique,
Sanger sequencing, pyrosequencing techniques, nanopore sequencing
techniques, etc.) for sequencing one or more biological samples
(e.g., sequencing microorganism nucleic acids from the biological
samples, etc.). Next-generation sequencing systems (e.g.,
next-generation sequencing platforms, etc.) can include any
suitable sequencing systems (e.g., sequencing platforms, etc.) for
one or more of high-throughput sequencing (e.g., facilitated
through high-throughput sequencing technologies; massively parallel
signature sequencing, Polony sequencing, 454 pyrosequencing,
Illumina sequencing, SOLiD sequencing, Ion Torrent semiconductor
sequencing, DNA nanoball sequencing, Heliscope single molecule
sequencing, Single molecule real time (SMRT) sequencing, Nanopore
DNA sequencing, etc.), any generation number of sequencing
technologies (e.g., second-generation sequencing technologies,
third-generation sequencing technologies, fourth-generation
sequencing technologies, etc.), amplicon-associated sequencing
(e.g., targeted amplicon sequencing), sequencing-by-synthesis,
tunnelling currents sequencing, sequencing by hybridization, mass
spectrometry sequencing, microscopy-based techniques, and/or any
suitable next-generation sequencing technologies.
[0040] As shown in FIG. 3, embodiments of the system 200 (e.g., for
characterizing a microorganism-related condition) can include any
one or more of: a handling system (e.g., a sample handling system;
including a sequencing system 215; etc.) 210 operable to collect
and/or process biological samples (e.g., collected by users and
included in containers including pre-processing reagents; etc.)
from one or more users (e.g., a human subject, patient, animal
subject, environmental ecosystem, care provider, etc.) for
facilitating determination of a microorganism dataset (e.g.,
microorganism genetic sequences; microorganism sequence dataset;
etc.); a microorganism-related characterization system 220 operable
to determine significance index metrics, features (e.g., microbiome
composition features; microbiome functional features; diversity
features; relative abundance ranges; such as based on a
microorganism dataset and/or other suitable data; etc.), and/or
microorganism-related characterizations (e.g.,
microorganism-related condition characterizations, therapy-related
characterizations, characterizations for users, etc.); a therapy
facilitation system 230 operable to facilitate therapeutic
intervention (e.g., promote a therapy, etc.) for one or more
microorganism-related conditions (e.g., based on one or more
significance index metrics; for improving one or more
microorganism-related conditions; etc.); an interface 240 for
presenting significance index metrics and/or other suitable data;
sample kits 250 for collecting one or more samples; and/or other
suitable components. While the components of embodiments of the
system 200 are generally described as distinct components, they can
be physically and/or logically integrated in any manner. In an
example, embodiments of the system 200 can omit a therapy
facilitation system 230. Additionally or alternatively, the
functionality of embodiments of the system 200 can be distributed
in any suitable manner amongst any suitable system components.
However, the components of embodiments of the system 200 can be
configured in any suitable manner.
[0041] However, the method 100 and/or system 200 can be configured
in any suitable manner.
2.1 Determining a Set of Associations.
[0042] Embodiments of the method 100 can include determining a set
of associations between one or more microorganism taxa and one or
more microorganism-related conditions S110, which can function to
determine associations for use in determination of significance
index metrics.
[0043] Associations can include any one or more of positive
associations (e.g. positive correlations; causative associations;
etc.); negative associations (e.g., negative correlations;
causative associations; etc.); non-associations (e.g., no
correlation; etc.); and/or any other types of associations (e.g.,
relationships, connections between, etc.) between one or more taxa
and one or more microorganism-related conditions.
[0044] Any suitable number of taxa can be associated with any
suitable number of microorganism-related conditions, in any
suitable numerical relationship (e.g., 1 to many; many to 1,
etc.).
[0045] Determining a set of associations can include processing
condition-related information sources (e.g., third-party
information sources such as scientific literature, clinical tests,
etc.; sources including information regarding conditions,
associated microorganism taxa, associated markers; proprietary
sources; first-party sources; etc.). In a variation, Block S110 can
include manually processing condition-related information sources
(e.g., with human curation of markers, associations, effect sizes,
data usable for calculating effect sizes, and/or associated
information, etc.) to determine the set of associations and/or
other suitable parameters. In another variation, Block S110 can
include automatically processing condition-related information
sources. For example, Block S110 can include: generating a list of
online information sources; obtaining the online information
sources based on the list; processing the online information
sources to extract a set of taxa, associated conditions, and/or
other associated data (e.g., through applying natural language
processing techniques, etc.) for generating the set of associations
and/or other suitable data. In another example, automatically
processing information sources can include applying natural
language processing approaches and/or other suitable approaches for
analysis of the information sources, such as for extracting types
of taxa associated with one or more microorganism-related
conditions.
[0046] In variations, determining a set of associations can be
based on one or more conditions (e.g., using the association, such
as for downstream processing in determining one or more
significance index metrics, if the conditions are met; updating a
microorganism-related database with the association; etc.).
Conditions can include any one or more of: subject conditions such
as in relation to subject type such as human or animal,
characteristics regarding the human or animal, etc.; sample
conditions such as in relation to the sampling site for the samples
used in identification of the one or more associations; analytical
technique conditions such as in relation to types of analytical
techniques used in identification of the associations; metric
conditions such as in relation to the types of metrics provided by
information sources and/or used in characterizing the associations;
and/or any other suitable types of conditions.
[0047] In a specific example, determining associations can be based
on two conditions (e.g., using the association, such as for
downstream processing in determining one or more significance index
metrics, if the two conditions are met; updating a
microorganism-related database with the association if the two
conditions are met; etc.) including: (1) samples used in the
information sources were from adult humans and collected from an
appropriate sampling site; and (2) appropriate metrics (e.g.,
statistics) are available (e.g., provided by the information
source; etc.), such as including metrics accounting for the
direction of the association (e.g., whether a positive association
between a taxon and a microorganism-related condition, a negative
association between a taxon and a microorganism-related condition,
or a non-association between a taxon and a microorganism-related
condition; etc.), effect size metrics (e.g., coefficients of
correlation, such as between abundances of one or more
microorganism taxa and one or more microorganism-related
conditions; z-scores, etc.) and/or data enabling calculation of
effect size metrics (e.g., where such metrics can be transformed
into effect size metrics such as coefficients of correlation and/or
z-scores, etc.), such as one or more of mean, standard deviation,
sample sizes, odds ratios, risk ratios, proportions of individuals
in the control and study groups with and without the condition
and/or any other suitable metrics. In a specific example,
determining one or more associations can be based on (e.g., using
the association, such as for downstream processing in determining
one or more significance index metrics, if the condition is met;
updating a microorganism-related database with the association;
etc.) the probability of regarding a false or spurious effect as
true is less than 5% (i.e. P-Value<0.05), such as indicating
there is a statistically significant association between the taxon
and a microorganism-related condition. In a specific example,
determining associations can be based on (e.g., using the
association, such as for downstream processing in determining one
or more significance index metrics, if the condition is met;
updating a microorganism-related database with the association;
etc.) the significance of the statistical comparison between the
study and the control groups, such as based on the P-value, where a
P-Value of less than 5% indicates the association is statistically
significant. However, conditions for determining associations can
be configured in any suitable manner.
[0048] In variations, determining one or more associations can
include determining one or more parameters describing the one or
more associations. Parameters describing the one or more
associations can include any one or more of: effect size metrics
(e.g., coefficients of correlation, such as between abundances of
one or more microorganism taxa and one or more
microorganism-related conditions; z-scores, etc.), data enabling
calculation of effect size metrics, mean, standard deviation,
sample sizes, odds ratios, risk ratios, proportions of individuals
in the control and study groups with and without the condition
and/or any other suitable metrics, experimental parameters,
confidence levels, sample characteristics, parameters associated
with types of conditions (e.g., subject parameters; sample
parameters; analytical technique parameters; metric parameters;
etc.), parameters provided by information sources, and/or any other
suitable types of parameters.
[0049] Determining a set of associations (and/or microbiome
features, reference features, user features, etc.) can be based on
microorganism datasets (e.g., microbiome features; associated with
one or more microorganism-related conditions; etc.), supplementary
datasets, and/or other suitable data, such as in a manner including
and/or analogous to that described in U.S. application Ser. No.
16/115,542 filed 28 Aug. 2018 and/or U.S. application Ser. No.
16/047,840 filed 27 Jul. 2018, which are herein incorporated in
their entireties by this reference.
[0050] Determining a set of associations can include generating
and/or updating (e.g., refining, adding to, deleting data, etc.)
one or more microorganism-related databases based on the determined
set of associations (e.g., adding additional associations to the
one or more microorganism-related databases), such as in response
to determining the set of associations, but generating and/or
updating microorganism-related databases can be performed in any
suitable manner at any suitable time and frequency.
[0051] However, determining a set of associations (e.g., between
one or more taxa and one or more microorganism-related conditions;
etc.) and/or any suitable parameters (e.g., effect size estimates,
other data, etc.) S110 can be performed in any suitable manner.
2.2 Determining a Set of Reference Features.
[0052] Embodiments of the method 100 can include determining a set
of reference features (e.g., reference abundance ranges, etc.) for
one or more microorganism taxa (e.g., microorganism taxa for which
associations were determined with one or more microorganism-related
conditions; etc.) S120, which can function to determine features
for use in determination of significance index metrics.
[0053] Reference features are preferably associated with (e.g.,
describe, correspond to, etc.) one or more microorganism taxa, such
as one or more microorganism taxa for which associations with one
or more microorganism-related conditions are determined (e.g., in
relation to S110). Additionally or alternatively, reference
features can be associated with any one or more of microbiome
composition (e.g., microbiome composition diversity, etc.),
microbiome function (e.g., microbiome functional diversity, etc.),
any suitable subjects and/or users (e.g., any suitable groups,
subgroups, and/or sets of subjects and/or users), and/or any other
suitable aspects.
[0054] Reference features preferably include reference abundance
ranges (e.g., reference relative abundance ranges) for one or more
taxa associated with one or more microorganism-related conditions.
In examples, reference abundance ranges (and/or user abundance
ranges corresponding to one or more users, and/or any suitable
abundance ranges; etc.) can include one or more healthy abundance
ranges (e.g., corresponding to a healthy range of abundance of a
taxon associated with a microorganism-related condition, such as a
health range derived based on subjects without the
microorganism-related condition; etc.), unhealthy abundance ranges
(e.g., corresponding to a unhealthy range of abundance of a taxon
associated with a microorganism-related condition, such as an
unhealthy range derived based on subjects with the
microorganism-related condition; etc.), low abundance ranges,
normal abundance ranges, high abundance ranges, absent abundance,
medium abundance ranges, percentiles for ranges (e.g., in relation
to any suitable group of subjects, samples, etc.), and/or any other
suitable types of abundance ranges. However, microorganism
abundance ranges can be configured in any suitable manner.
[0055] Reference features can be determined from the same or
different information sources used in variations of determining
associations between one or more taxa and one or more
microorganism-related conditions (e.g., a same information source
providing reference features such as reference abundance ranges for
associations between a set of taxa and a microorganism-related
condition.
[0056] Additionally or alternatively, reference features can be
determined in any suitable manner analogous to or different to
determining one or more associations between taxa and
microorganism-related conditions. In variations, determining
reference features can include determining reference features based
on sample processing and bioinformatics analysis for collected
samples from an aggregate population of subjects associated with
one or more microorganism-related conditions (e.g., including a
subgroup of subjects with the condition; a control subgroup of
subjects without the condition; etc.). In variations, determining
reference features can be based on one or more of information
sources, empirical analysis, sample processing, bioinformatics
analysis, and/or any other suitable processes.
[0057] Determining reference features, user features, and/or any
suitable portions of embodiments of the method 100 can include
applying pre-preprocessing (e.g., for data extracted from
information sources, for microorganism datasets, microbiome
features, and/or other suitable data for facilitation of downstream
processing such as determining significance index metrics, etc.).
In an example, performing a characterization process can include,
filtering a dataset (e.g., filtering a dataset extracted from an
information source, filtering a microorganism sequence dataset,
such as prior to applying a set of analytical techniques to
determine the microbiome features such as reference features,
etc.), by at least one of: removing first sample data corresponding
to first sample outliers of a set of biological samples (e.g.,
associated with one or more microorganism-related conditions,
etc.), such as where the first sample outliers are determined by at
least one of principal component analysis, a dimensionality
reduction technique, and a multivariate methodology; removing
second sample data corresponding to second sample outliers of the
set of biological samples, where the second sample outliers can
determined based on corresponding data quality for the set of
microbiome features (e.g., removing samples corresponding to a
number of microbiome features with high quality data below a
threshold condition, etc.); c) removing one or more microbiome
features from the set of microbiome features based on a sample
number for the microbiome feature failing to satisfy a threshold
sample number condition, where the sample number corresponds to a
number of samples associated with high quality data for the
microbiome feature; and/or any other suitable filtering techniques
for any suitable data described herein. However, pre-processing can
be performed with any suitable analytical techniques in any
suitable manner.
[0058] Determining reference features, user features, and/or other
suitable features (e.g., microbiome features, supplementary
features, etc.) can use computational methods (e.g., statistical
methods, machine learning methods, artificial intelligence methods,
bioinformatics methods, other approaches described herein, etc.) to
characterize a subject, sample, dataset, and/or other suitable
component as exhibiting and/or otherwise associated with one or
more features (e.g., where determining user microbiome features can
include determining feature values for microbiome features
identified as correlated with and/or otherwise associated with one
or more microorganism-related conditions, etc.), such as features
characteristic of a set of users with the one or more
microorganism-related conditions, etc.). However, any suitable
analytical techniques (e.g., described herein) can be used in
determining features and/or performing suitable portions of
embodiments of the method 100. In an example, determining reference
features and/or suitable features can include applying a set of
analytical techniques including at least one of a univariate
statistical test, a multivariate statistical test, a dimensionality
reduction technique, and an artificial intelligence approach, such
as where the features can improve computing system-related
functionality associated with the determining of significance index
metrics (e.g., in relation to accuracy, reducing error, processing
speed, scaling, etc.). In an example, determining microbiome
features (e.g., user microbiome features, etc.) can include
applying a set of analytical techniques to determine at least one
of presence of at least one of a microbiome composition diversity
feature and a microbiome functional diversity feature, absence of
the at least one of the microbiome composition diversity feature
and the microbiome functional diversity feature, a relative
abundance feature describing relative abundance of different
taxonomic groups associated with the first microorganism-related
condition, a ratio feature describing a ratio between at least two
microbiome features associated with the different taxonomic groups,
an interaction feature describing an interaction between the
different taxonomic groups, and a phylogenetic distance feature
describing phylogenetic distance between the different taxonomic
groups, such as in relation to (e.g., associated with) one or more
microorganism-related conditions, and such as where the set of
analytical techniques can include at least one of a univariate
statistical test, a multivariate statistical test, a dimensionality
reduction technique, and an artificial intelligence approach.
[0059] In variations, upon identification of represented groups of
microorganisms of the microbiome associated with one or more
samples (e.g., from subjects with or without one or more
microorganism-related conditions; etc.), features associated with
(e.g., derived from) compositional and/or functional aspects of the
microbiome can be determined. In a variation, generating features
can include generating features based upon multilocus sequence
typing (MSLT), in order to identify markers useful for significance
index metric determination and/or suitable portions of embodiments
of the method 100. Additionally or alternatively, determining
features can include determining features that describe the
presence or absence of certain taxonomic groups of microorganisms,
and/or ratios between exhibited taxonomic groups of microorganisms.
Additionally or alternatively, determining features can include
determining features describing one or more of: quantities of
represented taxonomic groups (e.g., taxa), networks of represented
taxonomic groups, correlations in representation of different
taxonomic groups, interactions between different taxonomic groups,
products produced by different taxonomic groups, interactions
between products produced by different taxonomic groups, ratios
between dead and alive microorganisms (e.g., for different
represented taxonomic groups, based upon analysis of RNAs),
phylogenetic distance (e.g., in terms of Kantorovich-Rubinstein
distances, Wasserstein distances etc.), any other suitable
taxonomic group-related feature(s), any other suitable genetic or
functional aspect(s).
[0060] Additionally or alternatively, determining features can
include generating features describing relative abundance of
different microorganism groups, for instance, using a sparCC
approach, using Genome Relative Abundance and Average size (GAAS)
approach and/or using a Genome Relative Abundance using Mixture
Model theory (GRAMMy) approach that uses sequence-similarity data
to perform a maximum likelihood estimation of the relative
abundance of one or more groups of microorganisms. Additionally or
alternatively, determining features can include generating
statistical measures of taxonomic variation, as derived from
abundance metrics. Additionally or alternatively, determining
features can include determining features associated with (e.g.,
derived from) relative abundance factors (e.g., in relation to
changes in abundance of a taxon, which affects abundance of other
taxa). Additionally or alternatively, determining features can
include generation of qualitative features describing presence of
one or more taxonomic groups, in isolation and/or in combination.
Additionally or alternatively, determining features can include
generation of features related to genetic markers (e.g.,
representative 16S, 18S, and/or ITS sequences) characterizing
microorganisms of the microbiome associated with a biological
sample. Additionally or alternatively, determining features can
include generation of features related to functional associations
of specific genes and/or organisms having the specific genes.
Additionally or alternatively, determining features can include
generation of features related to pathogenicity of a taxon and/or
products attributed to a taxon. Additionally or alternatively,
determining features can include determination of any other
suitable feature(s), such as derived from information sources,
sequencing and mapping of nucleic acids of a biological sample,
and/or any suitable approaches. For instance, the feature(s) can be
combinatory (e.g. involving pairs, triplets), correlative (e.g.,
related to correlations between different features), and/or related
to changes in features (e.g., temporal changes, changes across
sample sites, etc., spatial changes, etc.).
[0061] In variations, determining features can include determining
one or more site-specific associated with one or more collection
sites (e.g., gut site, nose site, skin site, genital site, mouth
site, etc.). In an example, a set of site-specific features can
include a first subset of site-specific features associated with a
first body site, and a second subset of site-specific features
associated with a second body site. However, multi-site analyses
can be performed in any suitable manner.
[0062] In variations, determining features can include applying
computer-implemented rules (e.g., models, feature selection rules,
etc.) to process population-level data and/or other suitable data,
but can additionally or alternatively include applying
computer-implemented rules to process microbiome-related data on a
demographic characteristic-specific basis (e.g., subgroups sharing
one or more demographic characteristics such as therapy regimens,
dietary regimens, physical activity regimens, ethnicity, age,
gender, weight, behaviors, etc.), condition-specific basis (e.g.,
subgroups exhibiting a specific microorganism-related condition, a
combination of microorganism-related conditions, triggers for the
microorganism-related conditions, associated symptoms, etc.), a
sample type-specific basis (e.g., applying different
computer-implemented rules to process microbiome data derived from
different collection sites; etc.), a user basis (e.g., different
computer-implemented rules for different users; etc.) and/or any
other suitable basis. In examples, determining features can include
assigning users from the population of users to one or more
subgroups; and applying different computer-implemented rules for
determining features (e.g., the set of feature types used; the
types of characterization models generated from the features; etc.)
for the different subgroups. However, applying computer-implemented
rules can be performed in any suitable manner for any suitable
portions of embodiments of the method 100, such as for determining
significance index metrics S130.
[0063] Determining features can include process can include
determining one or more abundance ranges (e.g., reference
microbiome parameter ranges; a healthy reference relative abundance
range, where the range can be associated with a healthy microbiome
and/or the absence of one or more conditions; a risk reference
relative abundance range associated with the presence of and/or
risk of one or more conditions; microorganism composition range for
abundance of one or more taxa; phylogenetic diversity of the
microorganisms present in the sample; microorganism functional
diversity range for functional features associated with one or more
taxa; etc.), such as reference abundance range and/or user
abundance ranges, and/or any suitable abundance ranges, such as
where one or more significance index metrics can be based on a
comparison between the user microbiome parameter (e.g., user
abundance, etc.) and the reference microbiome parameter range
(e.g., characterizing a user as possessing an poor significance
index metric for microbiome composition in relation to bacterial
targets associated with microorganism-related conditions based on
the user microbiome parameter indicating an abundance outside of
the healthy reference ranges for the different bacterial targets;
etc.). Microbiome parameter ranges can have any suitable lower- and
upper-limits, in any suitable form (e.g., counts, etc.). Reference
microbiome parameter ranges can include ranges representing any
suitable confidence intervals (e.g., 99% confidence intervals
across a population of users). In an example, reference relative
abundance ranges can be calculated for any suitable taxa, such as
based on dividing the count of reads corresponding to that taxa by
the total number of reads (e.g., total number of clustered and
filtered reads); however, reference relative abundance ranges can
be calculated in any suitable manner.
[0064] In a variation, determining reference abundance ranges
and/or suitable features can be performed empirically. For example,
Block S130 can include collecting biological samples and
supplementary datasets from a population of users. The population
of users can include users associated with any suitable state of
microbiome composition, microbiome phylogenetic diversity,
microbiome functional diversity, conditions, and/or other suitable
characteristics, where the supplementary datasets (e.g., digitally
administered surveys at an application executing on mobile devices
associated with the users) can be informative of the
characteristics. In a specific example, the method 100 can include:
processing biological samples from a population of healthy users;
processing the biological sample to determine microorganism
sequences; determining relative abundance of each taxa (e.g., from
a set of taxa determined to be associated and/or potentially
associated with one or more microorganism-related conditions, etc.)
for each user; and generating healthy ranges (and/or unhealthy
ranges) for each of the taxa based on the relative abundances
across the population of healthy users. However, empirically
determining reference microbiome parameter ranges can be performed
in any suitable manner. In a specific example, the supplementary
data can indicate a lack of the at least one microorganism-related
condition for a subset of subjects from a set of subjects; where
determining the set of microbiome features can include determining
healthy reference microbiome parameters ranges associated with the
subset of subjects, based on the microorganism sequence dataset;
and where determining one or more significance index metrics can be
based on the on the supplementary data and/or the healthy reference
microbiome parameters ranges. In a variation, determining reference
microbiome parameter ranges can be performed non-empirically, such
as based on manually and/or automatically processing
condition-related information sources.
[0065] Comparing one or more reference features (e.g., abundance
ranges, etc.) to one or more user microbiome features (e.g.,
abundances, etc.) associated with one or more characteristics
(e.g., taxa, conditions, etc.) can be used in determining one or
more significance index metrics, such as including characterizing
the user as possessing the characteristic (e.g., a healthy
microbiome, etc.) or not possessing the characteristic based on
whether the user microbiome parameter values fall inside or outside
the reference microbiome parameter ranges.
[0066] Determining reference features can additionally or
alternatively include updating reference features (e.g., at one or
more microorganism-related databases; etc.), such as for improving
the set of reference features used in determining one or more
significance index metrics (e.g., for improving accuracy of
significance index metrics, such as in relation to characterizing
one or more associations between microorganism taxa and
microorganism-related conditions; etc.).
[0067] However, determining reference features S120 can be
performed in any suitable manner.
2.3 Determining a Significance Index Metric.
[0068] Embodiments of the method 100 can include determining one or
more significance index metrics (e.g., based on the set of
associations and the set of reference features, etc.) S130, which
can function to determine one or more metrics associated with
characterization of one or more associations between one or more
microorganism taxa and one or more microorganism-related
conditions.
[0069] Significance index metrics preferably describe a degree of
association between a set of taxa and one or more
microorganism-related conditions, but can additionally or
alternatively describe users, propensity for one or more
microorganism-related conditions, risk for one or more
microorganism-related conditions, characteristics useable for
determining one or more microorganism-related condition
characterizations (e.g., diagnoses, other suitable data for
facilitating diagnoses, etc.), characteristics usable for
determining one or more therapies (e.g., for facilitating
therapeutic intervention for one or more microorganism-related
conditions; etc.), and/or can be associated with any suitable
aspects.
[0070] Significance index metrics can include nay one or more of
scores (e.g., expressed as a range from 0 to 100 and/or any
suitable range), propensity scores for users (e.g., describing a
user propensity for one or more microorganism-related conditions
based on user microbiome features and/or other suitable user data;
etc.), classifications (e.g., by a machine learning model;
classifications of presence or absence of one or more
microorganism-related conditions; any suitable classifications
associated with one or more microorganism-related conditions, such
as in relation to condition severity; etc.), and/or can include any
suitable form of data described herein.
[0071] Significance index metrics can be for any number of
associations (e.g., between taxa and microorganism-related
conditions, etc.), users (e.g., propensity scores describing a
user's propensity for one or more microorganism-related conditions;
etc.), taxa, microorganism-related conditions, and/or any suitable
components.
[0072] Determining significance index metrics and/or any suitable
portions of embodiments of the method 100 and/or system 200 can
include employing one or more analytical techniques including any
one or more of: univariate statistical tests, multivariate
statistical tests, dimensionality reduction techniques, artificial
intelligence approaches (e.g., machine learning approaches, etc.),
performing pattern recognition on data (e.g., identifying
correlations between microorganism-related conditions and
microbiome features; etc.), fusing data from multiple sources
(e.g., generating characterization models based on microbiome data
and/or supplementary data from a plurality of users associated with
one or more microorganism-related conditions, such as based on
microbiome features extracted from the data; etc.), combination of
values (e.g., averaging values, etc.), compression, conversion
(e.g., digital-to-analog conversion, analog-to-digital conversion),
performing statistical estimation on data (e.g. ordinary least
squares regression, non-negative least squares regression,
principal components analysis, ridge regression, etc.), wave
modulation, normalization, updating (e.g., of characterization
models and/or therapy models based on processed biological samples
over time; etc.), ranking (e.g., microbiome features; therapies;
etc.), weighting (e.g., microbiome features; etc.), validating,
filtering (e.g., for baseline correction, data cropping, etc.),
noise reduction, smoothing, filling (e.g., gap filling), aligning,
model fitting, binning, windowing, clipping, transformations,
mathematical operations (e.g., derivatives, moving averages,
summing, subtracting, multiplying, dividing, etc.), data
association, multiplexing, demultiplexing, interpolating,
extrapolating, clustering, image processing techniques, other
signal processing operations, other image processing operations,
visualizing, and/or any other suitable processing operations.
[0073] Artificial intelligence approaches can include any one or
more of: supervised learning (e.g., using logistic regression,
using back propagation neural networks, using random forests,
decision trees, etc.), unsupervised learning (e.g., using an
Apriori algorithm, using K-means clustering), semi-supervised
learning, a deep learning algorithm (e.g., neural networks, a
restricted Boltzmann machine, a deep belief network method, a
convolutional neural network method, a recurrent neural network
method, stacked auto-encoder method, etc.) reinforcement learning
(e.g., using a Q-learning algorithm, using temporal difference
learning), a regression algorithm (e.g., ordinary least squares,
logistic regression, stepwise regression, multivariate adaptive
regression splines, locally estimated scatterplot smoothing, etc.),
an instance-based method (e.g., k-nearest neighbor, learning vector
quantization, self-organizing map, etc.), a regularization method
(e.g., ridge regression, least absolute shrinkage and selection
operator, elastic net, etc.), a decision tree learning method
(e.g., classification and regression tree, iterative dichotomiser
3, C4-5, chi-squared automatic interaction detection, decision
stump, random forest, multivariate adaptive regression splines,
gradient boosting machines, etc.), a Bayesian method (e.g., naive
Bayes, averaged one-dependence estimators, Bayesian belief network,
etc.), a kernel method (e.g., a support vector machine, a radial
basis function, a linear discriminate analysis, etc.), a clustering
method (e.g., k-means clustering, expectation maximization, etc.),
an associated rule learning algorithm (e.g., an Apriori algorithm,
an Eclat algorithm, etc.), an artificial neural network model
(e.g., a Perceptron method, a back-propagation method, a Hopfield
network method, a self-organizing map method, a learning vector
quantization method, etc.), an ensemble method (e.g., boosting,
boostrapped aggregation, AdaBoost, stacked generalization, gradient
boosting machine method, random forest method, etc.), and/or any
suitable artificial intelligence approach. However, data processing
can be employed in any suitable manner.
[0074] Significance index metrics can additionally or alternatively
include site-specific significance index metrics (e.g., specific to
one or more body sites, including any one or more of gut sites,
genital sites, nose sites, skin sites, mouth sites, and/or other
suitable sites; etc.), such as significance index metrics
characterizing associations between one or more
microorganism-related conditions and one or more taxa, where the
associations are specific to a specific body site. In an example,
for the same body site and microorganism-related conditions,
associations with taxa can differ (e.g., can be associations with
different taxa) based on one or more body sites involved (e.g.,
different associations for a gut site compared to a nose site;
etc.). In a specific example, significance index metrics can differ
(e.g., values of significance index metrics, types of significance
index metrics) based on the one or more body sites involved. In
examples, different site-specific significance index models can be
generated, applied, and/or otherwise processed. In specific
examples, different site-specific significance index models can be
generated, applied, and/or otherwise processed based on different
microbiome features, such as site-specific features associated with
the one or more body sites that the site-specific significance
index model is associated with (e.g., using gut site-specific
features derived from samples collected at gut collection sites of
subjects, and/or correlated with one or more microorganism-related
conditions, such as for determining gut site-specific features,
generating a gut site-specific significance index model that can be
applied for determining significance index metrics based on user
samples collected at user gut sites, and/or for any suitable
purpose; etc.). Site-specific models, site-specific features,
samples, site-specific therapies, and/or other suitable entities
(e.g., able to be associated with a body site, etc.) are preferably
associated with at least one body site (e.g., corresponding to a
sample collection site; etc.) including one or more of a nose site,
gut site (e.g., characterizable based on stool samples, etc.), skin
site, genital site (e.g., vaginal site, etc.), mouth site, and/or
any suitable body region. However, site-specific significance index
metrics can be configured in any manner and determined in any
suitable manner.
[0075] In a variation, determining one or more significance index
metrics can be based one or more effect size metrics associated
with (e.g., describing, characterizing, etc.) one or more
associations between one or more taxa and one or more
microorganism-related conditions. Effect size metrics preferably
include coefficients of correlation (e.g., between abundance for a
taxon and a microorganism-related condition) but can additionally
or alternatively include z-scores, and/or any suitable types of
metrics (e.g., described herein, etc.). In a specific example,
determining the significance index metric can include determining
effect size metrics for the set of associations between the set of
microorganism taxa and the at least one microorganism-related
condition, based on the set of reference abundance ranges (and/or
suitable reference features) for the set of microorganism taxa; and
determining the significance index metric based on the effect size
metrics.
[0076] In examples, data extracted from one or more information
sources (e.g., mean, standard deviation, sample sizes, proportions
of individuals in the control and study groups with and without the
condition, etc.) can be transformed into one or more types of
effect size metrics (e.g., coefficients of correlation, etc.). In
specific examples, transformations into types of effect size
metrics, and/or suitable portions of determining significance
metrics based on effect size metrics can be performed with one or
more computing systems (e.g., remote computing systems, such as
including microorganism-related databases, etc.), and/or through
any suitable components. In an example, a significance index can be
calculated based on an overall coefficient of correlation obtained
from combination of a plurality of individual effect size
metrics.
[0077] In specific examples, coefficients of correlation (e.g.,
obtained through transformations of data extracted from information
sources such as scientific peer-reviewed articles, etc.) can be
transformed to z-scores (e.g., using Fisher's transformation), such
as by using
z=0.5*[(ln(1+r))/(ln(1-r))]
where r corresponds to coefficient of correlation; where a
meta-analysis can be performed (e.g., where the z-score is regarded
as a dependent variable and taxa are included as independent
factors; where different taxa will have different associations with
(e.g., different effects on; etc.) the one or more
microorganism-related conditions; such as where, an assumption can
be established that the information sources used are a subset of
all information sources, and the true effect size is not supposed
to be the same in all cases, motivating a random effect model to be
fitted and where the information sources used are included as a
random effect; and where the output of the analysis can include the
predicted values of z-score for each taxon; and/or where the
z-scores can be transformed (e.g., transformed back) to
coefficients of correlation.
[0078] In an example (e.g., in a first example, as shown in Table
1, such as where Table 1 includes observed and theoretical maximum
and minimum values for each of the significance index metrics,
which are resealed between 0 and 100; in a first example, as shown
in FIG. 5, such as where FIG. 5 includes histograms of frequencies
of significance index metrics for processed samples processed, such
as using analytical techniques described herein, where X-axis
indicates significance index metrics scaled to the range 0-1 using
the maximum and minimum observed values from reference samples;
where results include a single major peak corresponding to samples
having 0 relative abundance on all relevant taxa; etc.),
determining a significance index can be based on
SI=1-(.PI.(1-r.sub.a)*.PI.(1-r.sub.ia))
where .PI.=product function (e.g., the first product function runs
over the directly associated taxa, while the second product
function runs over the inversely associated taxa),
r.sub.a=coefficient of correlation of the associated taxa (e.g.,
positive correlation; as shown in Table 2), r.sub.ia=coefficient of
correlation of the inversely associated organisms (e.g., negative
correlation; as shown in Table 2); such as where according to the
abundance and the direction of the association, the correlations
are classified as "protective" when associated (e.g., positively
associated) taxa are found in low or normal abundances, when they
are not found at all in the sample, or when inversely associated
(e.g., negatively associated) taxa are found in high abundance in
the sample; and/or where correlations are classified as "penalty"
when associated taxa are positive in the sample, when the taxa are
found in high abundance, or when inversely associated taxa are
found in low or normal abundance in the sample. In a specific
example, determining the effect size metrics can include
determining a set of coefficient of correlations for the set of
associations between the set of microorganism taxa and the at least
one microorganism-related condition, based on a meta-analysis; and
where determining the significance index metric based on the effect
size metrics can include determining the significance index metric
based on the set of coefficient of correlations. In a specific
example, determining the significance index metric for the user can
include determining the significance index metric based on the user
microbiome composition features and a set of coefficient of
correlations for the set of associations between the set of
microorganism taxa and the at least one microorganism-related
condition.
[0079] In an example (e.g., a second example, as shown in Table 1
and FIG. 5), each z-score (e.g., from taxa found an individual's
samples) can be multiplied by a factor dependent on the direction
of the association (e.g., positive or negative (inverse), between
the taxon and the microorganism-related condition, etc.) and on the
abundance of the organism (e.g., corresponding to the taxon, etc.)
in the sample (e.g., low, normal or high); where the total z-score
can be determined by multiplying the individual scores from each
taxon; and/or where the modified score is subtracted from 1 to get
the probability based on:
Probability=1-(z_score*abundance multiplier)
where the abundance multiplier is determined based on the abundance
of the taxon and the direction of the association, such as
according to:
Inversely associated/low abundance=0
Inversely associated/normal abundance=1
Inversely associated/high abundance=1
Associated/low abundance=0
Associated/normal abundance=0
Associated/high abundance=1
where additionally or alternative multipliers (e.g., weights) can
be added (e.g., if needed), such as before converting the z-scores
back to coefficients of correlation; and/or where the one or more
total z-scores can be transformed back to coefficients of
correlation such as through the inverse Fisher transformation:
r=[exp(2*z)-1]/[exp(2*z)+1]
where r corresponds to coefficient of correlation, and where a
score from -1 to 1 can be obtained, and where to obtain the
percentage of association, the score can be multiplied by 100. In a
specific example, determining the effect size metrics can include
determining a set of z-scores for the set of associations between
the set of microorganism taxa and the at least one
microorganism-related condition; and modifying the set of z-scores
based on the reference abundance ranges for the set of
microorganism taxa and at least one of the positive associations
and the negative associations between the set of microorganism taxa
and the at least one microorganism-related condition; and where
determining the significance index metric based on the effect size
metrics can include determining the significance index metric based
on the modified set of z-scores. In a specific example, determining
the significance index metric for the user can include determining
the significance index metric based on the user microbiome
composition features and a set of modified z-scores determined
based on the reference microbiome composition features and a set of
z-scores for the set of associations between the set of
microorganism taxa and the at least one microorganism-related
condition.
[0080] In an example, determining a significance index metric can
be based on a Lowry method (e.g., for determining concentration of
proteins in a sample, etc.), such as including performing a
meta-analysis and weighing the outputs with one or more abundance
multipliers (e.g., described above, etc.); generating mock samples
and random combinations of abundances of the set of taxa (e.g.,
corresponding to associations with microorganism-related
conditions; etc.); determining a calibration curve based on the
abundances; and using real samples (e.g., from subjects, users,
etc.) in an interpolation process with the calibration curve to
determine significance index. In a specific example, determining
the effect size metrics can include performing an interpolation
process based on the reference abundance ranges and a calibration
curve derived from a random set of abundances for the set of
microorganism taxa; and where determining the significance index
metric based on the effect size metrics can include determining the
significance index metric based on the interpolation process. In a
specific example, determining the significance index metric for the
user can include determining the significance index metric based on
user microbiome composition features (and/or suitable user
microbiome features) and an interpolation process with the
reference microbiome composition features (and/or suitable
reference features) and a calibration curve derived from a random
set of abundances for the set of microorganism taxa. However
determining one or more significance index metrics based on effect
size metrics can be performed in any suitable manner.
[0081] In a variation, determining one or more significance index
metrics can include determining one or more propensity scores, such
as based on effect size metrics and user features (e.g., user
abundances, user microbiome features such as user microbiome
composition features, etc.). In a specific example, determining the
significance index metric can include determining a propensity
score for a user describing an association between a user
microbiome and the at least one microorganism-related condition,
based on the effect size metrics and user abundances for the set of
microorganism taxa. In a specific example, user microbiome
composition features include user abundances for the set of
microorganism taxa, where the reference microbiome composition
features include reference abundance ranges for the set of
microorganism taxa, and where determining the significance index
metric includes determining a propensity score for the user
characterizing the association between the user microbiome and the
at least one microorganism-related condition, based on the user
abundances and effect size metrics determined based on the
reference abundance ranges and the set of associations between the
set of microorganism taxa and the at least one
microorganism-related condition.
[0082] In an example (e.g., third example, as shown in Table 1 and
FIG. 5), determining a propensity score can be based on:
.SIGMA..sub.i=1.sup.Tf.sub.thc.sub.t
where f.sub.th corresponds to the relative abundance of the t-th
taxa on the h-th user, and where c.sub.t corresponds to the effect
size. In an example (e.g., a fourth example, as shown in Table 1
and FIG. 5), determining a propensity score can be based on
significance of the effect size, where determining the propensity
score can be based on:
i = 1 T f th p t c t with p t = BF 10 1 + BF 10 ##EQU00001##
where the BF parameter corresponds to the Bayes Factor (e.g., as
shown in Table 2), which transforms the coefficients of correlation
into probabilities (e.g., as shown in Table 2). A propensity score
associated with (e.g., calculated using the equation in, etc.) the
third example and a propensity score associated with (e.g.,
calculated using the equation in, etc.) the fourth example can be
formulated, respectively in a fifth and a sixth example (e.g., as
shown in Table 1 and FIG. 6), by transforming the taxa frequency
into a continuous range spanning from minus to positive infinity
using the quantiles of a standard normal distribution (e.g., to
improve apparent discontinuity on the score produced by taxa having
0 relative abundance, etc.); such as where a pseudocount of 1/10000
can be applied to the abundances for the two scores. In a specific
example, determining the propensity score includes determining the
propensity score based on the user abundances, the effect size
metrics, and significance metrics for the effect sizes.
[0083] In examples (e.g., where variation of significance index can
be affected by variation on the taxa abundance among users and
magnitude of effect size metrics; etc.), propensity scores can be
normalized, such as based on bounding with a predictable minimum
and maximum value. In a specific example, a normalized propensity
score (and/or other suitable normalized significance index metric)
can be determined based on:
normalised score = score h - score minimum score maximum - score
minimum ##EQU00002##
such as where the minimum and maximum values can be determined
empirically based on processed samples from users (e.g., where
scores for the samples can cover the observed range of abundance of
each taxon, etc.), determined based on information sources, and/or
otherwise determined; and where the minimum and maximum values can
be updatable (e.g., based on newly processed samples, etc.). In a
specific example, determining the significance index metric
includes normalizing the propensity score based on a set of
empirical abundance ranges for the set of microorganism taxa. In a
specific example, a set of vaginal samples (and/or other suitable
samples) can be processed to determine empirical abundances for a
set of taxa associated with human papilloma virus (HPV) (and/or
other suitable microorganism-related conditions; etc.), where the
minimum and maximum abundances and/or abundances corresponding to
minimum and maximum significance index metric scores can be used in
determining upper and lower boundaries of significance index
metrics, that can be translated respectively into a score of 100
and a score of 0. In examples normalization of propensity score
and/or any other suitable metrics can include one or more of:
scaling the determined metrics based on observed minimum and/or
maximum values of the metric type and/or parameters used in
determining the metric (e.g., based on processing samples; updating
based on processing new samples; etc.); filtering samples used in
determining minimum and/or maximum values of the metric type, such
as based on manual and/or automatic analysis (e.g., in relation to
appropriateness for use, such as based on correlation with
microorganism-related conditions related to the target
microorganism-related condition; etc.); choosing maximum and/or
minimum values as an empirical percentile of distribution of
observed values from a set of samples (e.g., lower or higher
99.sup.th percentile; etc.), such as for lowering the effect of
extreme values and allowing estimation of new maximum and/or
minimum value as additional data (e.g., from additional processed
samples) become available; down-weighting the effect of extreme
outliers without having to choose a particular percentile, such as
for enabling use of all data and weighing with higher importance to
a larger portion of the set of data, for reducing the effect of
extreme values and enabling estimation of new maximum and minimum
values automatically as more data becomes available; and/or any
other suitable processes. However, determining propensity scores
can be performed in any suitable manner.
[0084] In a variation, determining one or more significance index
metrics can be based on one or more labels determined for (e.g.,
identified for, assigned to, etc.) one or more samples. In an
example, labels can be assigned based on abundance of taxa, in the
sample, corresponding to the set of taxa associated with the at
least one microorganism-related condition. In a specific example,
the method 100 can include: determining abundances for a set of
taxa for a given sample; comparing the abundances to healthy
abundance ranges for the each taxon associated with the one or more
microorganism-related conditions; and assigning a label (e.g., a
"flag") for each taxon-condition association based on
[0085] Low-normal-high range of abundance: the flag applies when
the taxon is directly correlated with the health condition, and
abundance is `high`;
[0086] Low-normal-high range of abundance: the flag applies when
the taxon is inversely correlated with the health condition, and
abundance is `low`;
[0087] Absent-medium-high range of abundance: the flag applies when
the taxon abundance is `high`;
[0088] Low-normal range of abundance: the flag applies when the
taxon is inversely correlated with the health condition, and
abundance is `low`;
[0089] Normal-high range of abundance: the flag applies when the
taxon is directly correlated with the health condition, and
abundance is `high`;
[0090] Negative-positive: the flag applies when the taxon is
`positive` (e.g., it has a non-zero abundance in the sample);
[0091] Where, for each microorganism-related condition, the number
of assigned labels (e.g., "flags") can be counted, and the
significance can be calculated based on:
(A/B).times.100
where A=Number of taxa with a label for a condition; and B=Total
taxa associated with that condition. In a specific example,
determining the significance index metric includes determining a
set of labels for a user sample, where determining the set of
labels includes determining a label of the set of labels for a
taxon of the set of microorganism taxa based on satisfaction of an
abundance condition by a user abundance for the taxon in relation
to a reference abundance range for the taxon, and satisfaction of
an association type condition by an association, of the set of
associations, between the taxon and the at least one
microorganism-related condition, and determining the significance
index metric for a user associated with the user sample, based on
the set of labels. However, determining significance index metrics
based on one or more labels can be determined in any suitable
manner.
[0092] In a variation, determining one or more significance index
metrics can be based on one or more artificial intelligence
approaches (e.g., machine learning models; such as significance
index models applying artificial intelligence approaches; etc.),
such as to calculate a sample's probability of coming from a user
with a certain microorganism-related condition of interest or being
healthy. Determining one or more significance index metrics based
on one or more artificial intelligence approaches can include any
one or more of: transforming (e.g., centered log ratio
transformation, isometric log ratio transformation; filtering such
as in relation to feature selection for selecting features with
greatest contribution to classification and/or improved output
accuracy; applying any suitable machine learning algorithms (e.g.,
for training and/or processing types of machine learning models
described herein; etc.); model selection (e.g., for selecting
between different types of machine learning models, such as based
on accuracy comparisons; etc.); applying machine learning models,
such as to classify one or more components (e.g., samples), such as
for classification of a sample (and/or user, etc.) as healthy or
presenting one or more microorganism-related conditions; and/or
other suitable processes. In a specific example, the method 100 can
include performing one or more centered log ratio and/or isometric
log ratio transformation of the abundance of microorganisms present
in one or more samples; filtering datasets, such as to select only
the features corresponding to the taxa with greatest contribution
to classification of samples as either healthy or presenting a
microorganism-related condition; performing a set of different
artificial intelligence approaches (e.g., random forest
classifiers, support vector machines, logistic regression, K-means,
closest neighbors, etc.) and/or other suitable analytical
techniques for training a machine learning model for classifying
samples (e.g., novel samples) from users as either healthy or
presenting a microorganism-related condition; and applying the one
or more selected machine learning models, such as using user
features (e.g., derived a microorganism sequence dataset generated
based on a user sample; etc.) as inputs, for classifying one or
more samples as either healthy or presenting a
microorganism-related condition, based on the probability output by
the machine learning model. In a specific example, the machine
learning model can output a probability of belonging to either
group (e.g., healthy or microorganism-related condition), where the
significance index metric can be determined based on:
Significance Index=(probability of being in the
"microorganism-related condition" group).times.100
where the determined significance index metric can be used not just
as a score of closeness to one or another group (e.g., healthy
group or microorganism-related condition group, etc.), but
additionally or alternatively as the probability of belonging to
one category or another; and where the significance index metrics
can be used to classify microbial samples as coming from healthy
users or users with one or more microbial-related condition (e.g.,
performing or facilitating one or more diagnoses; etc.).
[0093] In a specific example, determining the significance index
metric includes determining a microorganism-related condition
classification associated with a health state of a user for the at
least one microorganism-related condition, based on user microbiome
composition features (e.g., user abundances for a set of taxa
associated with one or more microorganism-related conditions; etc.)
and a machine learning model derived from the set of associations
and the set of reference abundance ranges (e.g., for the set of
taxa associated with one or more microorganism-related conditions;
etc.). In a specific example, determining the significance index
metric includes determining a microorganism-related condition
classification associated with a health state of the user for the
at least one microorganism-related condition, based on the user
microbiome composition features (e.g., user abundances for a set of
taxa associated with one or more microorganism-related conditions;
etc.) and a machine learning model derived from the set of
associations and the set of reference microbiome composition
features (e.g., reference abundance ranges for the set of taxa
associated with one or more microorganism-related conditions;
etc.).
[0094] Artificial intelligence approaches can be used to determine
any suitable number and type of classifications, probabilities,
and/or other suitable outputs, such as for determining any suitable
types of significance index metrics (e.g., described herein, etc.).
For example, machine learning models can be trained with any
suitable number of labels for one or more microorganism-related
conditions (e.g., labels of healthy and different severities of a
condition; different labels for different conditions, where outputs
of the model can include probabilities of presenting each of the
different microorganism-related conditions; etc.).
[0095] Different significance index models (e.g., different
combinations of significance index models; different models
applying different analytical techniques; different inputs and/or
output types; applied in different manners such as in relation to
time and/or frequency; different significance index approaches,
such as in relation to variants described herein, etc.) can be
applied (e.g., executed, selected, retrieved, stored, trained,
generated; as shown in FIGS. 14-15, etc.) based on one or more of:
microorganism-related conditions (e.g., using different
significance index models depending on the microorganism-related
condition or conditions being characterized, such as where
different significance index models possess differing levels of
suitability for processing data in relation to different
microorganism-related conditions and/or combinations of conditions,
etc.), taxa (e.g., using different significance index models
depending on the types of taxa involved with determining
significance index metrics, such as the types of taxa with
associations to the relevant one or more microorganism-related
conditions; etc.), users (e.g., different significance index models
based on different user data and/or characteristics, demographic
characteristics, genetics, environmental factors, etc.),
microorganism-related characterizations (e.g., different
significance index models for different types of characterizations,
such as a therapy-related characterization versus a
diagnosis-related characterization, such as for identifying a
classification versus determining a propensity score for a
microorganism-related condition; etc.), therapies (e.g., different
significance index models for determining and/or monitoring
efficacy of different therapies, etc.), body sites (e.g., different
significance index models for processing microorganism datasets
corresponding to biological samples from different sample
collection sites; etc.), supplementary data (e.g., different models
for predicting different types of user data, etc.), and/or any
other suitable components. However, significance index models
(e.g., as shown in FIG. 10) can be tailored and/or used in any
suitable manner for facilitating significance index metric
determination.
[0096] Additionally or alternatively, determining one or more
significance index metrics S130 and/or any suitable portions of
embodiments of the method 100 can employ any suitable combination
of analytical techniques (e.g., in any suitable manner) described
in U.S. application Ser. No. 16/047,840 filed 27 Jul. 2018, which
is herein incorporated in its entirety by this reference. However,
determining significance index metrics S130 can be performed in any
suitable manner.
2.4 Facilitating Diagnosis.
[0097] Embodiments of the method 100 can additionally or
alternatively include facilitating diagnosis of one or more
microorganism-related conditions based on one or more significance
index metrics (and/or associated data) S140, which can function to
use significant index metrics and/or associated data to diagnose
and/or aid in diagnosis of one or more microorganism-related
conditions. In examples, calculated propensity scores,
classifications (e.g., using machine learning models), sets of
assigned labels, and/or other suitable significance index metrics
and/or associated data can be used in diagnosis. In an example, the
method 100 can include facilitating diagnosis of the user for the
at least one microorganism-related condition based on the
significance index metric.
[0098] Facilitating diagnosis can include using significance index
metrics in an additional or alternative manner to using other
suitable diagnostic data (e.g., supplementary data provided by a
user; etc.) and/or diagnostic procedures (e.g., computed tomography
(CT scan), ultrasound, biopsy, blood test, cancer screening exams,
urine test diagnostic imaging, other suitable diagnostic procedures
associated with microorganism-related conditions, survey-related
information, and/or any other suitable test, etc.). In a specific
example, facilitating diagnosis can include recommending a user to
undergo one or more diagnostic procedures and/or to request
additional diagnostic-related data, such as based on one or more
conditions (e.g., conditions indicating a likelihood of presenting
one or more microorganism-related conditions, etc.). In specific
examples, diagnosis, recommending additional diagnostic procedures,
and/or requesting additional diagnostic-related data can be in
response to a calculated propensity score satisfying a threshold
condition; a number of assigned labels (e.g., "flags") to taxa
satisfying a threshold condition; or a probability output (e.g.,
corresponding to a classification) satisfying a threshold
condition. Additionally or alternatively, threshold conditions
and/or other suitable types of conditions can be used in any
suitable manner for any suitable portions of embodiments of the
method 100. However, using significance index metrics with one or
more other processes can be performed in any suitable manner.
[0099] Facilitating diagnosis S140, facilitating therapeutic
intervention S150, and/or other suitable portions of embodiments of
the method 100 can be based on (e.g., can use as inputs into a
model, can user as inputs into calculations; etc.) one or more
features (e.g., described herein; determined in relation to S120;
user features; reference features; supplementary features; etc.),
and/or any other suitable data.
[0100] Facilitating diagnosis can include any one or more of:
providing a diagnosis; providing a diagnostic recommendation (e.g.,
to seek a care provider to perform a diagnostic procedure; etc.);
transmitting significance index metrics and/or other associated
data to one or more entities (e.g., care providers, for use by care
providers in performing a diagnosis; etc.); providing reports to
users (e.g., at user devices; etc.); and/or any other suitable
diagnostic-related processes.
[0101] Facilitating diagnosis can include facilitating detection of
microorganism-related conditions for a user, which can motivate
subsequent promotion of therapies, such as for modulation of a user
microbiome for improving a user health state associated with one or
more microorganism-related conditions (e.g., modulation of a user
microbiome towards healthy abundance ranges for taxa associated
with the one or more microorganism-related conditions; etc.).
Additionally or alternatively, diagnostic procedures can include
any one or more of: medical history analyses, imaging examinations,
cell culture tests, antibody tests, skin prick testing, patch
testing, blood testing, challenge testing, performing portions of
embodiments of the method 100, and/or any other suitable procedures
for facilitating the detecting (e.g., observing, predicting, etc.)
of microorganism-related conditions. Additionally or alternatively,
diagnostic device-related information and/or other suitable
diagnostic information can be processed in relation to facilitating
diagnosis and/or therapeutic intervention, and/or collected, used,
and/or otherwise processed in relation to any suitable portions of
embodiments of the method 100.
2.5 Facilitating Therapeutic Intervention.
[0102] Embodiments of the method 100 can additionally or
alternatively include facilitating therapeutic intervention for the
one or more microorganism-related conditions (e.g., based on the
one or more significance index metrics and/or associated data,
etc.) S150, which can function to use significant index metrics
and/or associated data to facilitate therapeutic intervention
(e.g., promote therapies, provide therapies, etc.), such as for
improving a health state of a user in relation to one or more
microorganism-related conditions. In an example, the method 100 can
include facilitating therapeutic intervention for the user for the
at least one microorganism-related condition based on the
significance index metric.
[0103] Facilitating therapeutic intervention can include
identifying, selecting, ranking, prioritizing, predicting,
discouraging, and/or otherwise facilitating therapeutic
intervention. For example, facilitating therapeutic intervention
can include determining one or more of probiotic-based therapies,
bacteriophage-based therapies, small molecule-based therapies,
and/or other suitable therapies, such as therapies that can shift a
subject's microbiome composition, function, diversity, and/or other
characteristics (e.g., microbiomes at any suitable sites, etc.)
toward a desired state (e.g., equilibrium state, etc.), such as a
towards a healthy microbiome composition (e.g., abundances in
healthy reference abundance ranges; etc.), such as in promotion of
a user's health, such as for modifying a state of one or more
microorganism-related conditions, and/or for other suitable
purposes.
[0104] Therapies (e.g., microorganism-related therapies, etc.) can
include any one or more of: consumables (e.g., probiotic therapies,
prebiotic therapies, medication such as antibiotics, allergy or
cold medication, bacteriophage-based therapies, consumables for
underlying conditions, small molecule therapies, etc.);
device-related therapies (e.g., monitoring devices; sensor-based
devices; medical devices; implantable medical devices; etc.);
surgical operations; psychological-associated therapies (e.g.,
cognitive behavioral therapy, anxiety therapy, talking therapy,
psychodynamic therapy, action-oriented therapy, rational emotive
behavior therapy, interpersonal psychotherapy, relaxation training,
deep breathing techniques, progressive muscle relaxation,
meditation, etc.); behavior modification therapies (e.g., physical
activity recommendations such as increased exercise; dietary
recommendations such as reducing sugar intake, increased vegetable
intake, increased fish intake, decreased caffeine consumption,
decreased alcohol consumption, decreased carbohydrate intake;
smoking recommendations such as decreasing tobacco intake;
weight-related recommendations; sleep habit recommendations etc.);
topical administration therapies (e.g., topical probiotic,
prebiotic, and/or antibiotics; bacteriophage-based therapies);
environmental factor modification therapies; modification of any
other suitable aspects associated with one or more
microorganism-related conditions; and/or any other suitable
therapies (e.g., for improving a health state associated with one
or more microorganism-related conditions, such as therapies for
improving one or more microorganism-related conditions, therapies
for reducing the risk of one or more microorganism-related
conditions, etc.). In examples, types of therapies can include any
one or more of: probiotic therapies, bacteriophage-based therapies,
small molecule-based therapies, cognitive/behavioral therapies,
physical rehabilitation therapies, clinical therapies,
medication-based therapies, diet-related therapies, and/or any
other suitable therapy designed to operate in any other suitable
manner in promoting a user's health.
[0105] In variations, therapies can include site-specific therapies
associated with one or more body sites, such as for facilitating
modification of microbiome composition and/or function at one or
more different body sites of a user (e.g., one or more different
collection sites, etc.), such as targeting and/or transforming
microorganisms associated with one or more of a nose site, gut
site, skin site, mouth site, and/or genital site; such as by
facilitating therapeutic intervention in relation to one or more
therapies configured to specifically target one or more user body
sites, such as microbiome at one or more of the user body sites;
such as for facilitating improvement of one or more
microorganism-related conditions (e.g., by modifying user
microbiome composition and/or function at a particular user body
site towards a target microbiome composition and/or function, such
as microbiome composition and/or function at a particular body site
and associated with a healthy microbiome status and/or lack of the
one or more microorganism-related condition; etc.). Facilitating
therapeutic intervention in relation to site-specific therapies can
be based on site-specific significance index metrics (e.g.,
facilitating a site-specific therapy to modulate a microbiome
composition at a specific site towards a healthy microbiome
composition including taxa used in determining a corresponding
site-specific significance index metric, such as taxa associated
with the relevant microorganism-related condition at the site;
etc.). Site-specific therapies can include any one or more of
consumables (e.g., targeting a body site microbiome and/or
microbiomes associated with any suitable body sites; etc.); topical
therapies (e.g., for modifying a skin microbiome, a nose
microbiome, a mouth microbiome, a genitals microbiome, etc.);
and/or any other suitable types of therapies. In an example, the
method 100 can include collecting a sample associated with a first
body site (e.g., including at least one of a nose site, gut site, a
skin site, a genital site, a mouth site, etc.) from a user;
determining site-specific composition features associated with the
first body site; determining a significance index metric for the
user for the microorganism-related condition based on the
site-specific composition features (e.g., and site-specific
reference features; etc.); and facilitating therapeutic
intervention in relation to a first site-specific therapy for the
user (e.g., providing the first site-specific therapy to the user;
etc.) for facilitating improvement of the microorganism-related
condition, based on the significance index metric, where the first
site-specific therapy is associated with the first body site. In an
example, the method 100 can include collecting a post-therapy
sample from the user after the facilitation of the therapeutic
intervention in relation to the first site-specific therapy (e.g.,
after the providing of the first site-specific therapy; etc.),
where the post-therapy sample is associated with a second body site
(e.g., including at least one of the nose site, gut site, the skin
site, the genital site, the mouth site; etc.); determining a
post-therapy significance index metric for the user for the
microorganism-related condition based on site-specific features
associated with the second body site; and facilitating therapeutic
intervention in relation to a second site-specific therapy for the
user (e.g., providing a second site-specific therapy to the user;
etc.) for facilitating improvement of the microorganism-related
condition, based on the post-therapy significance index metric,
where the second site-specific therapy is associated with the
second body site. However, significance index metrics (e.g.,
site-specific or site-independent) can be determined and/or used at
any suitable time and frequency (e.g., pre-therapy, post-therapy,
at any suitable stage of a user's microbiome; at any suitable
temporal indicator; etc.).
[0106] In a variation, therapies can include one or more
bacteriophage-based therapies (e.g., in the form of a consumable,
in the form of a topical administration therapy, etc.), where one
or more populations (e.g., in terms of colony forming units) of
bacteriophages specific to a certain bacteria (or other
microorganism) represented in the subject can be used to
down-regulate or otherwise eliminate populations of the certain
bacteria. As such, bacteriophage-based therapies can be used to
reduce the size(s) of the undesired population(s) of bacteria
represented in the subject. Additionally or alternatively,
bacteriophage-based therapies can be used to increase the relative
abundances of bacterial populations not targeted by the
bacteriophage(s) used. However, bacteriophage-based therapies can
be used to modulate characteristics of microbiomes (e.g.,
microbiome composition, microbiome function, etc.) in any suitable
manner, and/or can be used for any suitable purpose.
[0107] In variations, therapies can include one or more probiotic
therapies and/or prebiotic therapies associated with any
combination of at least one or more of (e.g., including any
combination of one or more of, at any suitable amounts and/or
concentrations, such as any suitable relative amounts and/or
concentrations; etc.) any suitable taxa described herein (e.g., in
relation to one or more microbiome composition features associated
with one or more microorganism-related conditions, etc.), and/or
any other suitable microorganisms associated with any suitable
taxonomic groups (e.g., microorganisms from taxa described herein,
such as in relation to microbiome features; taxa associated with
functional features described herein, etc.). For one or more
probiotic therapies and/or other suitable therapies, microorganisms
associated with a given taxonomic group, and/or any suitable
combination of microorganisms can be provided at dosages of 0.1
million to 10 billion CFU, and/or at any suitable amount (e.g., as
determined from a therapy model that predicts positive adjustment
of a patient's microbiome in response to the therapy; different
amounts for different taxa; same or similar amounts for different
taxa; etc.). In an example, a subject can be instructed to ingest
capsules including the probiotic formulation according to a regimen
tailored to one or more of his/her: physiology (e.g., body mass
index, weight, height), demographic characteristics (e.g., gender,
age), severity of dysbiosis, sensitivity to medications, and any
other suitable factor. In examples, probiotic therapies and/or
prebiotic therapies can be used to modulate a user microbiome
(e.g., in relation to composition, function, etc.) for facilitating
improvement of one or more microorganism-related conditions. In
examples, facilitating therapeutic intervention can include
promoting (e.g., recommending, informing a user regarding,
providing, administering, facilitating obtainment of, etc.) one or
more probiotic therapies and/or prebiotic therapies to a user, such
as for facilitating improvement of one or more
microorganism-related conditions.
[0108] In a specific example of probiotic therapies, as shown in
FIG. 11, candidate therapies can perform one or more of: blocking
pathogen entry into an epithelial cell by providing a physical
barrier (e.g., by way of colonization resistance), inducing
formation of a mucous barrier by stimulation of goblet cells,
enhance integrity of apical tight junctions between epithelial
cells of a subject (e.g., by stimulating up regulation of
zona-occludens 1, by preventing tight junction protein
redistribution), producing antimicrobial factors, stimulating
production of anti-inflammatory cytokines (e.g., by signaling of
dendritic cells and induction of regulatory T-cells), triggering an
immune response, and performing any other suitable function that
adjusts a subject's microbiome away from a state of dysbiosis.
However, probiotic therapies and/or prebiotic therapies can be
configured in any suitable manner.
[0109] In another specific example, therapies can include
medical-device based therapies (e.g., associated with human
behavior modification, associated with treatment of disease-related
conditions, etc.).
[0110] Additionally or alternatively, facilitating therapeutic
intervention can be based on identification of a "normal" or
baseline microbiome composition and/or functional features, as
assessed from subjects of a population of subjects who are
identified to be in good health. Upon identification of a subset of
subjects of the population of subjects who are characterized to be
in good health (e.g., using features of the characterization
process), therapies that modulate microbiome compositions and/or
functional features toward those of subjects in good health (e.g.,
using significance index metrics; etc.) can be determined and/or
promoted. Therapies can additionally or alternatively include
therapies that can shift microbiomes of subjects who are in a state
of dysbiosis toward one of the identified baseline microbiome
compositions and/or functional features.
[0111] Microorganism compositions associated with probiotic
therapies and/or prebiotic therapies (e.g., associated with
probiotic therapies, such as determined based on one or more
significance index metrics, etc.) can include microorganisms that
are culturable (e.g., able to be expanded to provide a scalable
therapy) and/or non-lethal (e.g., non-lethal in their desired
therapeutic dosages). Furthermore, microorganism compositions can
include a single type of microorganism that has an acute or
moderated effect upon a subject's microbiome. Additionally or
alternatively, microorganism compositions can include balanced
combinations of multiple types of microorganisms that are
configured to cooperate with each other in driving a subject's
microbiome toward a desired state. For instance, a combination of
multiple types of bacteria in a probiotic therapy can include a
first bacteria type that generates products that are used by a
second bacteria type that has a strong effect in positively
affecting a subject's microbiome. Additionally or alternatively, a
combination of multiple types of bacteria in a probiotic therapy
can include several bacteria types that produce proteins with the
same functions that positively affect a subject's microbiome.
[0112] Probiotic and/or prebiotic compositions can be naturally or
synthetically derived. For instance, in one application, a
probiotic composition can be naturally derived from fecal matter or
other biological matter (e.g., of one or more subjects having a
baseline microbiome composition and/or functional features, as
identified using one or more processes of embodiments of the method
100; etc.). Additionally or alternatively, probiotic compositions
can be synthetically derived (e.g., derived using a benchtop
method) based upon a baseline microbiome composition and/or
functional features. In variations, microorganism agents that can
be used in probiotic therapies can include one or more of: yeast
(e.g., Saccharomyces boulardii), gram-negative bacteria (e.g., E.
coli Nissle), gram-positive bacteria (e.g., Bifidobacteria bifidum,
Bifidobacteria infantis, Lactobacillus rhamnosus, Lactococcus
lactis, Lactobacillus plantarum, Lactobacillus acidophilus,
Lactobacillus casei, Bacillus polyfermenticus, etc.), and any other
suitable type of microorganism agent. However, probiotic therapies,
prebiotic therapies and/or other suitable therapies can include any
suitable combination of microorganisms associated with any suitable
taxa described herein, and/or therapies can be configured in any
suitable manner.
[0113] In variations, facilitating therapeutic intervention (e.g.,
providing therapies, etc.) and/or other suitable portions of
embodiments of the method 100 can include provision of
notifications (e.g., as shown in FIGS. 8A-8D, 12, and 16) to a user
regarding one or more recommended therapies, other forms of
therapy, significance index metrics, diagnoses, diagnostic
recommendations, and/or any other suitable data (e.g., described
herein; etc.). In a specific example, facilitating therapeutic
intervention (e.g., providing a therapy; etc.) can include
providing therapy recommendations (e.g., substantially
concurrently, such as in the same report, as providing significance
index metrics and/or data derived from significance index metrics;
etc.) and/or other suitable therapy-related information (e.g.,
therapy efficacy; comparisons to other individual users, subgroups
of users, and/or populations of users; therapy comparisons;
historic therapies and/or associated therapy-related information;
psychological therapy guides such as for cognitive behavioral
therapy; etc.). Notifications can be presented a web interface
(e.g., through a user account associated with and identifying a
user; etc.). Notifications can be provided to a user by way of an
electronic device (e.g., personal computer, mobile device, tablet,
wearable, head-mounted wearable computing device, wrist-mounted
wearable computing device, etc.), such as a device that executes an
application, web interface, and/or messaging client configured for
notification provision. In one example, a web interface of a
personal computer or laptop associated with a user can provide
access, by the user, to a user account of the user, where the user
account can be associated with information regarding the user's
microbiome (e.g., detailed characterization of aspects of the
user's microbiome in relation to correlations with
microorganism-related conditions; etc.), and/or notifications
regarding suggested therapeutic measures (e.g., generated in Blocks
S140 and/or S170, etc.). In another example, an application
executing at a personal electronic device (e.g., smart phone, smart
watch, head-mounted smart device) can be configured to provide
notifications (e.g., at a display, haptically, in an auditory
manner, etc.), such as regarding therapy suggestions based on
significance index metrics. Notifications and/or probiotic
therapies can additionally or alternatively be provided directly
through an entity associated with a user (e.g., a caretaker, a
spouse, a significant other, a healthcare professional, etc.). In
variations, notifications can additionally or alternatively be
provided to an entity (e.g., healthcare professional) associated
with a user, such as where the entity is able to facilitate
provision of the therapy (e.g., by way of prescription, by way of
conducting a therapeutic session, through a digital telemedicine
session using optical and/or audio sensors of a computing device,
etc.). Providing notifications and/or otherwise facilitating
therapeutic intervention, however, be performed in any suitable
manner.
[0114] However, facilitating therapeutic intervention S150 can be
performed in any suitable manner.
3. EXAMPLES
[0115] In specific examples, one or more diet-related condition
models (e.g., for determining diet-relation condition
classifications such as whether a user consumes a certain food or
drink, follows a certain diet, etc.), such as a caffeine
consumption model (e.g., and/or other suitable significance index
models and/or related models; and/or other suitable models for any
suitable diet-related conditions, such as described herein, etc.)
can be processed (e.g., generated, trained, applied, etc.), such as
for predicting, based on set of microbiome composition features
(e.g., user microbiome composition features such as user abundances
for taxa associated with caffeine consumption, etc.) and/or other
suitable features (e.g., supplementary features, microbiome
functional features, etc.) a diet-related condition metric, such as
a caffeine consumption metric (e.g., a classification as to whether
the user is a caffeine consumer or not, etc.). In specific
examples, taxa (and/or associated abundances) can be associated
with one or more diet-related conditions (e.g., dietary profiles;
types of consumed foods; etc.), such as where the taxa can
additionally or alternatively be associated with one or more other
microorganism-related conditions (e.g., disease; etc.), such as
where determination of a diet's effect on microbiome can enable
inferences and/or other suitable insights into a microbiome, a
status of a microorganism-related condition, and/or other suitable
microbiome and/or microorganism-related condition insights, such as
in relation to diagnostics and/or therapeutics. In a specific
example, a caffeine consumer model can be trained upon and/or
otherwise processed based on features describing taxa abundances
for caffeine consumers and non-caffeine consumers. In a specific
example, different artificial intelligence approaches can be
applied for training different machine learning models, such as
including support vector machine (SVM) models (e.g., as shown in
FIGS. 7B-7E) and/or random forest classifier models (e.g., as shown
in FIG. 7A). In specific examples, microbiome composition features
(and/or other suitable microbiome features) for caffeine consumer
prediction (e.g., for processing caffeine consumer models; etc.)
and/or other suitable diet-related condition predictions (e.g.,
classifications, etc.) can be associated with one or more taxa
including: Alistipes; Anaerotruncus; Bacteroides; Bifidobacterium;
Bilophila; Blautia; Butyricimonas; Clostridium; Collinsella;
Erysipelatoclostridium; Faecalibacterium; Flavobacterium;
Flavonifractor; Granulicatella; Hespellia; Intestinimonas;
Kluyvera; Lachnospira; Marvinbryantia; Odoribacter; Oscillibacter;
Parabacteroides; Phascolarctobacterium; Pseudobutyrivibrio;
Roseburia; Streptococcus; Subdoligranulum; Sutterella; and/or
Terrisporobacter. In a specific example, taxa for microbiome
composition features can be selected through any suitable means,
such as based on "importance score" in random forest approaches. In
a specific example, determining the microorganism-related condition
classification includes determining a caffeine consumption
classification for the user based on the machine learning model and
the user microbiome composition features associated with the set of
microorganism taxa, where the set of microorganism taxa includes at
least one of: Alistipes; Anaerotruncus; Bacteroides;
Bifidobacterium; Bilophila; Blautia; Butyricimonas; Clostridium;
Collinsella; Erysipelatoclostridium; Faecalibacterium;
Flavobacterium; Flavonifractor; Granulicatella; Hespellia;
Intestinimonas; Kluyvera; Lachnospira; Marvinbryantia; Odoribacter;
Oscillibacter; Parabacteroides; Phascolarctobacterium;
Pseudobutyrivibrio; Roseburia; Streptococcus; Subdoligranulum;
Sutterella; and Terrisporobacter. In a specific example,
determining the microorganism-related condition classification
includes determining a diet-related condition classification,
associated with a diet-related condition, for the user based on the
machine learning model and the user microbiome composition features
associated with the set of microorganism taxa. In a specific
example, the diet-related condition can include at least one of
caffeine consumption, alcohol consumption, artificial sweetener
consumption, and sugar consumption; where determining the
microorganism-related condition classification comprises
determining at least one of a caffeine consumption classification,
an alcohol consumption classification, an artificial sweetener
consumption classification, and a sugar consumption
classification.
[0116] In specific examples, one or more significance index metrics
can be determined for one or more HPV conditions and/or any
suitable women's health-related conditions (e.g., described herein;
as shown in FIGS. 8A-8D), such as for characterizing how a user's
vaginal microbiome composition is similar to or departs from that
of healthy individuals without one or more of the women's
health-related conditions.
[0117] Microbiome analysis can enable accurate and/or efficient
microorganism-related characterization (e.g., of a user microbiome,
of a user sample, of a user, etc.) and/or therapy provision (e.g.,
according to portions of embodiments of the method 100, etc.) for
microorganism-related conditions caused by, correlated with, and/or
otherwise associated with microorganisms, such as through
determination and/or use of significance index metrics. Specific
examples of the technology can overcome several challenges faced by
conventional approaches. First, conventional approaches can require
patients to visit one or more care providers to receive a
characterization and/or a therapy recommendation, such as for a
microorganism-related condition, which can amount to inefficiencies
and/or health-risks associated with the amount of time elapsed
before diagnosis and/or treatment, with inconsistency in healthcare
quality, and/or with other aspects of care provider visitation.
Second, conventional genetic sequencing and analysis technologies
for human genome sequencing can be incompatible and/or inefficient
when applied to the microbiome (e.g., where the human microbiome
can include over 10 times more microbial cells than human cells;
where viable analytical techniques and the means of leveraging the
analytical techniques can differ; where optimal sample processing
techniques can differ, such as for reducing amplification bias;
where different approaches to microorganism-related
characterizations can be employed; where the types of conditions
and correlations can differ; where causes of the associated
conditions and/or viable therapies for the associated conditions
can differ; where sequence reference databases can differ; where
the microbiome can vary across different body regions of the user
such as at different collection sites; etc.). Third, the onset of
sequencing technologies (e.g., next-generation sequencing,
associated technologies, etc.) has given rise to technological
issues (e.g., data processing and analysis issues for the plethora
of generated sequence data; issues with processing a plurality of
biological samples in a multiplex manner; information display
issues; therapy prediction issues; therapy provision issues, etc.)
that would not exist but for the unprecedented advances in speed
and data generation associated with sequencing genetic material.
Specific examples of the method 100 and/or system 200 can confer
technologically-rooted solutions to at least the challenges
described above.
[0118] First, specific examples of the technology can transform
entities (e.g., users, biological samples, therapy facilitation
systems including medical devices, etc.) into different states or
things. For example, the technology can transform a biological
sample into components able to be sequenced and analyzed to
generate significance index metrics, such as usable for
characterizing users in relation to one or more
microorganism-related conditions (e.g., such as through use of
next-generation sequencing systems, multiplex amplification
operations; etc.). In another example, the technology can identify,
discourage and/or promote (e.g., present, recommend, provide,
administer, etc.), therapies (e.g., personalized therapies based on
a microorganism-related characterization; etc.) and/or otherwise
facilitate therapeutic intervention (e.g., facilitating
modification of a user's microbiome composition, microbiome
functionality, etc.), such as based on significance index metrics,
which can prevent and/or ameliorate one or more
microorganism-related conditions, such as thereby transforming the
microbiome and/or health of the patient (e.g., improving a health
state associated with a microorganism-related condition; etc.),
such as based on applying one or more microbiome features (e.g.,
applying correlations, relationships, and/or other suitable
associations between microbiome features and one or more
microorganism-related conditions; etc.). In another example, the
technology can transform microbiome composition and/or function at
one or more different body sites of a user (e.g., one or more
different collection sites; etc.), such as targeting and/or
transforming microorganisms associated with the nose, gut, skin,
mouth, genitals, and/or other sites associated with a microbiome
(e.g., by facilitating therapeutic intervention in relation to one
or more site-specific therapies; etc.). In another example, the
technology can control therapy facilitation systems (e.g., dietary
systems; automated medication dispensers; behavior modification
systems; diagnostic systems; disease therapy facilitation systems;
etc.) to promote therapies (e.g., by generating control
instructions for the therapy facilitation system to execute; etc.),
thereby transforming the therapy facilitation system.
[0119] Second, specific examples of the technology can confer
improvements in computer-related technology (e.g., improving
computational efficiency in storing, retrieving, and/or processing
microorganism-related data for microorganism-related conditions
such as associations and/or features; computational processing
associated with biological sample processing, etc.) such as by
facilitating computer performance of functions not previously
performable. For example, the technology can apply a set of
analytical techniques in a non-generic manner to non-generic
microorganism datasets and/or microbiome features (e.g., that are
recently able to be generated and/or are viable due to advances in
sample processing techniques and/or sequencing technology, etc.)
for determining significance index metrics, such as for improving
microorganism-related characterizations (e.g., diagnoses, etc.)
and/or facilitating therapeutic intervention for
microorganism-related conditions.
[0120] Third, specific examples of the technology can confer
improvements in processing speed, microorganism-related
characterization, accuracy, microbiome-related therapy
determination and promotion, and/or other suitable aspects, such as
in relation to microorganism-related conditions. For example, the
technology can leverage non-generic microorganism datasets to
determine, select, and/or otherwise process microbiome features of
particular relevance to one or more types of microorganism-related
conditions (e.g., processed microbiome features relevant to a
microorganism-related condition; cross-condition microbiome
features with relevance to a plurality of microorganism-related
conditions, etc.), which can facilitate improvements in accuracy
(e.g., by using the most relevant microbiome features; by
leveraging tailored analytical techniques; etc.), processing speed
(e.g., by selecting a subset of relevant microbiome features; by
performing dimensionality reduction techniques; by leveraging
tailored analytical techniques; etc.), and/or other computational
improvements (e.g., in relation to phenotypic prediction, such as
indications of the microorganism-related conditions, etc.), other
suitable characterizations, therapeutic intervention facilitation,
and/or other suitable purposes. In a specific example, the
technology can apply feature-selection rules (e.g., microbiome
feature-selection rules for composition, function; for supplemental
features extracted from supplementary datasets; etc.) to select an
optimized subset of features (e.g., microbiome functional features
relevant to one or more microorganism-related conditions;
microbiome composition diversity features such as reference
relative abundance features indicative of healthy, presence,
absence, and/or other suitable ranges of taxonomic groups
associated with microorganism-related conditions; user relative
abundance features that can be compared to reference relative
abundance features correlated with microorganism-related conditions
and/or therapy responses; etc.) out of a vast potential pool of
features (e.g., extractable from the plethora of microbiome data
such as sequence data; identifiable by univariate statistical
tests; etc.) for generating, applying, and/or otherwise
facilitating characterization and/or therapies (e.g., through
models, etc.). The potential size of microbiomes (e.g., human
microbiomes, animal microbiomes, etc.) can translate into a
plethora of data, giving rise to questions of how to process and
analyze the vast array of data to generate actionable microbiome
insights in relation to microorganism-related characterizations.
However, the feature-selection rules and/or other suitable
computer-implementable rules can enable one or more of: shorter
generation and execution times (e.g., for generating and/or
applying models; for determining microorganism-related
characterizations and/or associated therapies; etc.); optimized
sample processing techniques (e.g., improving transformation of
microorganism nucleic acids from biological samples through using
primer types, other biomolecules, and/or other sample processing
components identified through computational analysis of taxonomic
groups, sequences, and/or other suitable data associated with
microorganism-related conditions, such as while optimizing for
improving specificity, reducing amplification bias, and/or other
suitable parameters; etc.); model simplification facilitating
efficient interpretation of results; reduction in overfitting;
network effects associated with generating, storing, and applying
microorganism-related characterizations for a plurality of users
over time in relation to microorganism-related conditions (e.g.,
through collecting and processing an increasing amount of
microbiome-related data associated with an increasing number of
users to improve predictive power of the microorganism-related
characterizations and/or therapy determinations; etc.);
improvements in data storage and retrieval (e.g., storing and/or
retrieving significance index models; storing specific models such
as in association with different users and/or sets of users, with
different microorganism-related conditions; storing microorganism
datasets in association with user accounts; storing therapy
monitoring data in association with one or more therapies and/or
users receiving the therapies, such as in relation to significance
index metrics; storing features, microorganism-related
characterizations, and/or other suitable data in association with a
user, set of users, and/or other entities to improve delivery of
personalized characterizations and/or treatments for the
microorganism-related conditions, etc.), and/or other suitable
improvements to technological areas.
[0121] Fourth, specific examples of the technology can amount to an
inventive distribution of functionality across components including
a sample handling system, a microorganism-related characterization
system, and a plurality of users, where the sample handling system
can handle substantially concurrent processing of biological
samples (e.g., in a multiplex manner) from the plurality of users,
which can be leveraged by the microorganism-related
characterization system in generating personalized
characterizations, and/or therapies (e.g., customized to the user's
microbiome such as in relation to the user's dietary behavior,
probiotics-associated behavior, medical history, demographic
characteristics, other behaviors, preferences, etc.) for
microorganism-related conditions.
[0122] Fifth, specific examples of the technology can improve the
technical fields of at least genomics, microbiology,
microbiome-related computation, diagnostics, therapeutics,
microbiome-related digital medicine, digital medicine generally,
modeling, and/or other relevant fields. In an example, the
technology can model and/or characterize different
microorganism-related conditions, such as through computational
identification of relevant microorganism features (e.g., which can
act as biomarkers to be used in diagnoses, facilitating therapeutic
intervention, etc.) for microorganism-related conditions. In
another example, the technology can perform cross-condition
analysis to identify and evaluate cross-condition microbiome
features associated with (e.g., shared across, correlated across,
etc.) a plurality of a microorganism-related conditions (e.g.,
diseases, phenotypes, etc.). Such identification and
characterization of microbiome features can facilitate improved
health care practices (e.g., at the population and individual
level, such as by facilitating diagnosis and therapeutic
intervention, etc.), by reducing risk and prevalence of comorbid
and/or multi-morbid microorganism-related conditions (e.g., which
can be associated with environmental factors, and thereby
associated with the microbiome, etc.). In specific examples, the
technology can apply unconventional processes (e.g., sample
processing processes; computational analysis processes; etc.), such
as to confer improvements in technical fields.
[0123] Sixth, the technology can leverage specialized computing
devices (e.g., devices associated with the sample handling system,
such as next-generation sequencing systems; microorganism-related
characterization systems; therapy facilitation systems; etc.) in
performing suitable portions associated with embodiments of the
method 100 and/or system 200.
[0124] Specific examples of the technology can, however, provide
any suitable improvements in the context of using non-generalized
components and/or suitable components of embodiments of the system
200 for microorganism-related characterization, microbiome
modulation, and/or for performing suitable portions of embodiments
of the method 100.
4. OTHER
[0125] Embodiments of the method 100 can, however, include any
other suitable blocks or steps configured to facilitate reception
of biological samples from subjects, processing of biological
samples from subjects, analyzing data derived from biological
samples, and generating models that can be used to provide
customized diagnostics and/or probiotic-based therapeutics
according to specific microbiome compositions and/or functional
features of subjects.
[0126] Embodiments of the method 100 and/or system 200 can include
every combination and permutation of the various system components
and the various method processes, including any variants (e.g.,
embodiments, variations, examples, specific examples, figures,
etc.), where portions of embodiments of the method 100 and/or
processes described herein can be performed asynchronously (e.g.,
sequentially), concurrently (e.g., in parallel), or in any other
suitable order by and/or using one or more instances, elements,
components of, and/or other aspects of the system 200 and/or other
entities described herein.
[0127] Any of the variants described herein (e.g., embodiments,
variations, examples, specific examples, figures, etc.) and/or any
portion of the variants described herein can be additionally or
alternatively combined, aggregated, excluded, used, performed
serially, performed in parallel, and/or otherwise applied.
[0128] Portions of embodiments of the method 100 and/or system 200
can be embodied and/or implemented at least in part as a machine
configured to receive a computer-readable medium storing
computer-readable instructions. The instructions can be executed by
computer-executable components that can be integrated with the
system. The computer-readable medium can be stored on any suitable
computer-readable media such as RAMs, ROMs, flash memory, EEPROMs,
optical devices (CD or DVD), hard drives, floppy drives, or any
suitable device. The computer-executable component can be a general
or application specific processor, but any suitable dedicated
hardware or hardware/firmware combination device can alternatively
or additionally execute the instructions.
[0129] As a person skilled in the art will recognize from the
previous detailed description and from the figures and claims,
modifications and changes can be made to embodiments of the method
100, system 200, and/or variants without departing from the scope
defined in the claims.
TABLE-US-00001 TABLE 1 Significance Index Samples Theoretical
Metric/Score Minimum Maximum Minimum Maximum First example 1.4
87.38 0 88.43 Second example 0 99.99 0 99.99 Third example -40.71
45.45 -41.0 64.0 Fourth example -17.34 45.45 -17.46 63.9 Fifth
example -4.59 1.42 -5.10 2.38 Sixth example -4.20 0.34 -5.92
2.38
TABLE-US-00002 TABLE 2 Probability of association; Calculated with
Bayes factor (BF) as probability of Correlation association = Taxa
coefficient Bayes factor BF/(1 + BF) Fusobacterium -0.1 0.14 0.12
nucleatum Gardnerella 0.083 0.42 0.29 Gardnerella 0.34 5.81 0.85
vaginalis Lactobacillus iners -0.41 0.74 0.42 Sneathia 0.64
35453.65 0.99
* * * * *