prs-net-deep-learning-agent

Geometric deep learning-based polygenic risk score prediction using PRS-Net for modeling gene interactions, enhanced disease prediction, and cross-ancestry portability.

Jump to install

Repository: mdbabumiamssm/LLMs-Universal-Life-Science-and-Clinical-Skills-
Last source activity: February 7, 2026 at 16:51
Detected SKILL.md language: English
Stars: 31
Forks: 8

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

Back to search Back to installation

name	prs-net-deep-learning-agent
description	Geometric deep learning-based polygenic risk score prediction using PRS-Net for modeling gene interactions, enhanced disease prediction, and cross-ancestry portability.

PRS-Net Deep Learning Agent

The PRS-Net Deep Learning Agent implements interpretable geometric deep learning for polygenic risk score prediction. PRS-Net models non-linear gene-gene interactions and biological network relationships to enhance disease prediction accuracy and improve cross-ancestry portability compared to traditional linear PRS methods.

When to Use This Skill

When linear PRS methods show limited performance.
For modeling complex gene-gene interactions.
To improve PRS portability across ancestries.
When biological interpretability of PRS is needed.
For integrating pathway and network information.

Core Capabilities

Non-Linear PRS: Capture gene-gene interactions via deep learning.
Network Integration: Incorporate protein-protein interaction networks.
Interpretability: Identify important pathways and gene modules.
Cross-Ancestry Transfer: Improved portability via learned biology.
Multi-Task Learning: Joint modeling of related traits.
Uncertainty Quantification: Provide prediction confidence.

PRS-Net Architecture

Component	Function	Innovation
Input Layer	Gene-level summaries	Aggregated variant effects
Network Encoder

Aspect	Linear PRS	PRS-Net
Gene Interactions	Not modeled	GNN captures
Network Biology	Ignored	Integrated
Interpretability	Limited (SNP weights)	Pathway-level
Cross-Ancestry	Often poor	Improved
Computational Cost	Low	Moderate
Training Data Needed	Low	Moderate

python3 Skills/Precision_Medicine/PRS_Net_Deep_Learning_Agent/prs_net_predict.py \
    --genotypes cohort_genotypes.vcf.gz \
    --ppi_network string_ppi.graphml \
    --trait type2_diabetes \
    --model_weights prs_net_t2d_v1.pt \
    --interpret_pathways true \
    --ancestry_calibration multi \
    --output prs_net_results/

Component	Function	Innovation
Input Layer	Gene-level summaries	Aggregated variant effects
Network Encoder

Interpretation Level	Output	Clinical Use
Gene	Top contributing genes	Target identification
Pathway	Enriched pathways	Mechanism understanding
Module	Network subgraphs	Biological insight
Hub Genes	Central genes	Druggable targets

Application	PRS-Net Advantage	Benefit
Risk Stratification	Higher accuracy	Better prediction
Biological Insight	Pathway interpretation	Mechanism
Drug Targets	Hub gene identification	Therapeutic targets
Ancestry Equity	Better portability	Fairer prediction

Limitation	Impact	Future Direction
Training Data	EUR-dominated	Diverse cohorts
Network Completeness	Missing edges	Multi-network integration
Rare Variants	Not well captured	WGS + rare variant methods
Clinical Validation	Limited trials	Prospective studies