Justin Silverman
Pennsylvania State University, PA·Mathematics & Statistics
DNA sequencing is now the default way to ask which microbes live in a sample or which genes a cell is expressing, but the numbers it returns are slipperier than they look. A sequencing run reports counts of reads, and the total number of reads has nothing to do with the total number of bacteria in a gut or molecules in a cell.
- Fields
Mathematics & Statistics
Computer Science
Biology
- Topics
sequence count data
compositional data analysis
measurement bias
- Grants
- 1 known grant, last ended Aug 2026
What they work on
- Build a theoretical framework for objectively evaluating existing analysis methods for sequence count data in light of scale limitations and measurement bias, the two problems that can dominate study conclusions.
- Develop Simulated Inference, a computational framework that lets researchers keep their preferred statistical models and software while explicitly propagating uncertainty from unmeasured scale and biased measurement.
- Validate the proposed tools on three case studies of real sequence count data, such as 16S rRNA microbiome surveys and single-cell RNA sequencing, and test whether they reduce false positive and false negative rates.
More about the research
Separately, the measurement process itself favors some sequences over others, so the observed mix of counts is systematically skewed. These two problems have been shown to influence, and sometimes dominate, the conclusions researchers draw, yet there are few tools for carrying out analyses that account for them or for judging which existing methods hold up.
Why it matters
Sequencing-based measurements of microbial communities and gene expression underpin a large share of biomedical research, and conclusions drawn from them can hinge on unstated assumptions about scale and measurement bias. Methods that make those assumptions explicit and carry their uncertainty through the analysis could make published findings about human health and disease more reproducible and reduce both missed effects and spurious ones.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Amy Willis
University of Washington, WA·Mathematics & Statistics
Develop robust statistical methods for differential abundance, which test whether microbes are more or less common between groups while accounting for bias and variability introduced by sequencing.
David Chan
University of California-Berkeley, CA·Social Science
Validate data-driven measures of pneumonia diagnoses and outcomes among 5.5 million chest X-ray visits in Veterans Health Administration emergency departments. Use quasi-random assignment of patients to radiologists to infer differences in false-positive and false-negative rates.
Hok Chio Lai
University of Southern California, CA·Psychology·Early-career
Develop a robust optimization algorithm for data harmonization that puts different instruments measuring the same concept on a common scale while also correcting for measurement bias across groups.
The school
Pennsylvania State University: 60 labs on PI Pages, $1.3B a year on research, 743 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.