3/3 With mcRigor, downstream analyses become more reliable:
✅ gene co-expression modules
✅ enhancer–gene associations
✅ temporal expression patterns
Benchmarking shows MetaCell & SEACells are better-performing methods.
I am deeply honored by this recognition from the ISCB, which would not have been possible without the immense trust and hard work of my trainees. I am also thankful for the strong support of my colleagues and administrative team.
I’m deeply thankful for the incredible support and opportunities at UCLA over the past 12 years—ever since finishing my PhD. Grateful for the journey so far and looking forward to this new chapter at the intersection of statistics and biomedical science.
I’m honored to join Fred Hutch as Professor and Program Head of Biostatistics, and as the Donald and Janet K. Guthrie Endowed Chair in Statistics. Excited to be part of a deeply collaborative and scientifically vibrant community with a rich legacy of impact. @fredhutchbiostat.bsky.social
I’ll speak on July 16 (2:30pm ET) at NSF@75: Advancing Statistical Science for a Data‑Driven World Conference, by ASA & Instats
My talk is about an info-theoretic criterion (ITCA) for combining ambiguous class labels: jmlr.org/papers/v23/2...
Free registration: instats.org/seminar/nsfa...
1/3 Metacells boost power in single-cell RNA-seq & multiome analysis. But without checking homogeneity, they risk forming dubious metacells that bias discoveries.
We introduce mcRigor: a statistical safeguard for rigorous metacell analysis.
👉 www.nature.com/articles/s41...
2/3 At the core of mcRigor is a feature-correlation statistic with a null derived from a double permutation scheme — a brilliant idea by my postdoc Pan Liu jsb-lab.org/people/pan-l...
mcRigor detects dubious metacells and optimizes method & hyperparameter for a specific dataset.
I hope that Nullstrap-DE can be a useful add-on to calibrate the FDR control of popular parametric DE methods, including DESeq2 and edgeR, while maintaining the strong power. Looking forward to feedback and comments from the community.
We will post the scDesignPop preprint soon.
Excited to share our method ClipperQTL published in Genome Biology.
Built on our p-value-free FDR control framework Clipper, ClipperQTL performs on par with FastQTL and runs up to 500× faster.
Big thanks to my former PhD student Heather Zhou!
genomebiology.biomedcentral.com/articles/10....
#eQTL
@guggfellows.bsky.social I’m deeply honored to be named a 2025 Guggenheim Fellow—especially as part of the Foundation’s historic 100th class. Grateful to be in the company of so many brilliant artists, scholars, and scientists. #guggfellows2025
