Analysis of Molecular and Cellular Interactions by Combining Network Algorithms and Human Insight
To develop a “human-in-the-loop” system for analysis of Human Cell Atlas data, combining statistically rigorous network algorithms with an interactive web platform for visualization, exploration, and annotation of molecular data.
Results & Resources
The Raphael Lab published several resources to advance the “human-in-the-loop” system by combining human intelligence with machine learning to efficiently analyze single-cell sequencing datasets with missing data and low coverage sequencing reads. They developed netNMF, a cell clustering algorithm, and CHISEL, an algorithm developed to improve data coverage in single-cell DNA sequencing. They also developed SCARLET to construct phylogenies from single-nucleotide variants simultaneously and copy number aberrations (CNAs).