Rahul Mazumder

This paper introduces Perturbscape, a framework that uses perturbation-induced gene programs to identify disease-relevant genes and pathways, showing higher heritability informativeness than steady-state approaches.

SPLENDID improves polygenic risk prediction across diverse populations by incorporating continuous genetic ancestry without requiring discrete ancestry labels.

This paper introduces graphREML, a likelihood-based method for heritability partitioning and enrichment analysis using GWAS summary statistics, showing greater power than S-LDSC.

STELLAR improves polygenic risk prediction by integrating rare and common variants using ensemble learning, enhancing accuracy and refining risk stratification.

Spacelink integrates spatial transcriptomics with genetics to identify spatially variable genes linked to disease, showing improved detection power and disease informativeness.

ALL-Sum improves polygenic risk prediction by combining efficient summary statistic-based regression with ensembling, outperforming existing methods in accuracy, speed, and memory usage.

The paper introduces HEELS, a new estimator for local heritability using GWAS summary statistics and LD matrix, achieving high statistical efficiency comparable to REML.

This paper introduces Robocov, a convex optimization framework for estimating gene-level tissue networks with missing data, showing it improves correlation and causal structure estimation and identifies disease-related genes.