Michigan Center for Single-Cell Genomic Data Analytics
Published:
Jun Z. Li (Departments of Human Genetics and Computational Medicine and Bioinformatics) and I are the co-directors of the Michigan Center for Single-Cell Genomic Data Analytics. Our research team connects mathematicians and data scientists with biological researchers to develop, evaluate, and implement a variety of cutting-edge methodologies in sparse data analysis.
I have focused on genetic marker selection for single-cell RNA seq data. PicturedRocks is a tool for the analysis of single cell RNA-seq data. Currently, we implement two marker selection approaches: a 1-bit compressed sensing based sparse SVM algorithm and a mutual information-based Max Relevance min Redundancy algorithm. Marker selection is defined simply as feature selection (in machine learning).
We emphasize that this software and its attendant rigorous analysis is very much a work in progress. We will be updating it as we determine better its empirical performance. Please contact us with your feedback and performance results!
- Software: Anna C. Gilbert, Alexander Vargo, Umang Varma, Pictured Rocks: Single Cell RNA-seq analysis tool, 2017.