P. Mila Jhamai

The paper develops interpretable neural networks for multi-omics data to predict smoking status, age, and LDL levels, showing improved performance and generalizability across cohorts.

The paper identifies 818 genes influencing DNA methylation in blood using genetic anchors and causal inference, revealing both transcription factors and non-DNA-binding proteins with regulatory roles.

This paper introduces TWMR, a Mendelian Randomization method that identifies gene-trait associations by integrating GWAS and eQTL data, revealing 3,913 novel associations across 43 human phenotypes.

The paper identifies trans-meQTLs linking disease variants to methylation changes at transcription factor binding sites, revealing eight-fold more downstream effects than trans-eQTLs.

The paper identifies context-dependent eQTLs in peripheral blood RNA-seq data from Dutch individuals using a hypothesis-free approach, revealing gene expression influenced by cell type and signaling pathways.