DeepEXPOKE: A Deep Learning Framework with Polygenic Risk Scores as Knockoffs for Deconvoluting Genetic and Non-Genetic Exposure Risks in Sepsis and Coronary Heart Disease
medRxiv (Cold Spring Harbor Laboratory),
Год журнала:
2024,
Номер
unknown
Опубликована: Окт. 16, 2024
ABSTRACT
The
exposome
refers
to
the
totality
of
environmental,
behavioral,
and
lifestyle
exposures
an
individual
experiences
throughout
one’s
lifetime.
Due
modifiability
exposures,
identifying
risk
on
a
disease
is
crucial
for
effective
intervention
prevention
disease.
However,
traditional
analytical
methods
struggle
capture
complexities
data:
nonlinear
effects,
correlated
potential
interplay
with
genetic
effects.
To
address
these
challenges
accurately
estimate
exposure
effects
complex
diseases,
we
developed
DeepEXPOKE,
deep
learning
framework
integrating
two
types
knockoff
features:
statistical
knockoffs
(statKO)
polygenic
score
as
(PRSKO).
DeepEXPOKE-statKO
controls
correlation
DeepEXPOKE-PRSKO
isolates
while
both
can
We
applied
DeepEXPOKE
predict
outcomes
significant
diseases
distinct
etiology
clinical
presentation:
sepsis
coronary
heart
(CHD),
demonstrating
its
performance
in
comparison
existing
machine
methods.
Furthermore,
identified
metabolites
such
glucose
triglycerides
factors
suggested
that
their
are
primarily
at
non-genetic
level,
consistent
role
responding
environmental
factors.
Additionally,
uniquely
asthma
factor
effect
partially
offering
insights
into
conflicting
associations
observed
between
genome
data
studies
patient
analysis
regarding
risk.
Overall,
offers
novel
DNN
approach
interpreting
factors,
advancing
our
understanding
diseases.
Язык: Английский
THE EFFECTIVENESS OF EXERCISE APPLICATION IN GENETICALLY PREDISPOSED OBESITY AND DIABETES
Bulletin of Problems Biology and Medicine,
Год журнала:
2024,
Номер
1(3), С. 36 - 36
Опубликована: Янв. 1, 2024
Язык: Английский
Effects of Gene–Lifestyle Interaction on Obesity Among Students
Genes,
Год журнала:
2024,
Номер
15(12), С. 1506 - 1506
Опубликована: Ноя. 24, 2024
Obesity
is
a
global
health
issue
influenced
primarily
by
genetic
variants
and
environmental
factors.
This
study
aimed
to
examine
the
relationship
between
lifestyle
factors
their
interaction
with
obesity
risk
among
university
students.
A
total
of
658
students
from
same
participated
in
this
study,
including
531
females
(mean
age
(SD):
21.6
(3.9)
years)
127
males
(21.9
(4.6)
years).
Among
them,
550
were
classified
as
normal
weight
or
underweight
(456
94
males),
while
108
identified
overweight
obese
(75
33
males).
All
participants
underwent
anthropometric
screening
completed
sleep
quality
questionnaires.
The
polygenic
score,
based
on
seven
(
revealed
that
individuals
higher
predisposition,
defined
polymorphic
loci,
are
more
susceptible
becoming
under
certain
conditions.
Язык: Английский