Aritra Bose

The paper predicts end-stage renal disease from asymptomatic individuals using multimodal data and identifies novel genomic biomarkers associated with kidney function and disease risk.

COMICAL discovers genetic associations with brain imaging phenotypes using contrastive learning and multi-omics data from the UK Biobank.

MaSk-LMM uses matrix sketching to efficiently estimate LMM parameters in GWAS, providing fast and accurate results compared to existing methods.

This paper introduces FairPRS, a method to create polygenic risk scores that are fair across ethnic groups by using invariant risk minimization, showing improved phenotype prediction and reduced racial bias.

ThreSPCA provides a fast and accurate algorithm for sparse PCA, revealing genetic diversity across the world using genotype data.

CuNA identifies genes associated with Parkinson's Disease symptoms by analyzing higher-order relationships between eQTLs and clinical phenotypes.

The paper introduces EpiSimRA, a coalescent model with multi-way epistasis, validated by a forward Wright-Fisher model, showing superior performance in simulating evolutionary scenarios with selection and recombination.

The paper models genetic diversity in India by integrating geography, language, and social structure, finding language and social factors most influential in shaping genetic patterns.

CluStrat improves population stratification correction in GWAS by using clustering and randomized regression, identifying more true causal variants than PCA-based methods.