Michelle H. Badri

This paper introduces FactorIV, a causal inference method for high-dimensional multi-omics data, demonstrating its ability to recover causal effects in biological systems.

This study shows shrinkage estimation improves microbial association estimates across different normalization methods using the American Gut Project data.

The paper compares correlation/proportionality methods with MMvec for identifying microbe-metabolite interactions and finds MMvec performs better on real data due to its handling of sparsity.