Jiuyao Lu

The paper introduces a non-parametric test for homogeneity of multivariate dispersions, showing it has higher power than existing methods in certain situations.

This paper introduces QuanT, a non-parametric method to identify unmeasured heterogeneity in microbiome data, improving downstream analysis accuracy.

The paper evaluates 47 highly variable gene selection methods for scRNA-seq across 19 datasets and finds hybrid methods outperform individual methods, proposing a new approach called mixHVG.

MOSAIC is a pipeline that compares microbiome integration strategies, finding that ComBat and ConQuR families perform best overall in balancing artifact removal and signal preservation.