Journal of Scientific Agriculture,
Journal Year:
2023,
Volume and Issue:
unknown, P. 63 - 76
Published: Sept. 11, 2023
The
goal
of
this
research
was
to
create
a
more
accurate
and
efficient
method
for
selecting
plants
with
disease
resistance
using
combination
genetic
markers
advanced
machine
learning
algorithms.
A
multi-disciplinary
approach
incorporating
genomic
data,
algorithms
high-performance
computing
employed.
First,
highly
associated
were
identified
next-generation
sequencing
data
statistical
analysis.
Then,
an
adiabatic
quantum
algorithm
developed
integrate
these
into
single
predictor
susceptibility.
results
demonstrate
that
the
integrative
use
significantly
improved
accuracy
efficiency
resistance-based
marker-assisted
plant
selection.
By
leveraging
power
markers,
effective
strategies
selection
can
be
developed.
PLoS ONE,
Journal Year:
2024,
Volume and Issue:
19(7), P. e0305654 - e0305654
Published: July 18, 2024
Even
though
deep
learning
shows
impressive
results
in
several
applications,
its
use
on
problems
with
High
Dimensions
and
Low
Sample
Size,
such
as
diagnosing
rare
diseases,
leads
to
overfitting.
One
solution
often
proposed
is
feature
selection.
In
learning,
along
selection,
network
sparsification
also
used
improve
the
when
dealing
high
dimensions
low
sample
size
data.
However,
most
of
time,
they
are
tackled
separate
problems.
This
paper
proposes
a
new
approach
that
integrates
based
sparsification,
into
training
process
neural
network.
uses
constrained
biobjective
gradient
descent
method.
It
provides
set
Pareto
optimal
networks
make
trade-off
between
sparsity
model
accuracy.
Results
both
artificial
real
datasets
show
using
increases
without
degrading
classification
performances.
With
approach,
an
dataset,
selection
score
reached
0.97
0.92
accuracy
0.9.
For
same
accuracy,
none
other
methods
above
0.20
0.35.
Finally,
statistical
tests
validate
obtained
all
datasets.
Journal of Scientific Agriculture,
Journal Year:
2023,
Volume and Issue:
unknown, P. 63 - 76
Published: Sept. 11, 2023
The
goal
of
this
research
was
to
create
a
more
accurate
and
efficient
method
for
selecting
plants
with
disease
resistance
using
combination
genetic
markers
advanced
machine
learning
algorithms.
A
multi-disciplinary
approach
incorporating
genomic
data,
algorithms
high-performance
computing
employed.
First,
highly
associated
were
identified
next-generation
sequencing
data
statistical
analysis.
Then,
an
adiabatic
quantum
algorithm
developed
integrate
these
into
single
predictor
susceptibility.
results
demonstrate
that
the
integrative
use
significantly
improved
accuracy
efficiency
resistance-based
marker-assisted
plant
selection.
By
leveraging
power
markers,
effective
strategies
selection
can
be
developed.
