Jordan Rossen

This paper introduces CompBioBench, a benchmark for evaluating agentic systems in computational biology, showing strong performance across diverse tasks.

The paper shows that per-allele effect sizes of genetic variants in European populations are better predicted by African MAF than European MAF.

MultiSuSiE improves multi-ancestry fine-mapping in All of Us whole-genome sequencing data by allowing causal effect sizes to vary across ancestries, achieving higher power and lower computational costs.

This paper introduces TGFM, a method for fine-mapping causal tissues and genes at disease-associated loci using GWAS and eQTL data, identifying 147 causal elements per disease with 11% as gene-tissue pairs.