Single Neuron Analyzer for Multi-modal, Cross-dataset (Epi)genomic Cell Type Datasets
Identifying and characterizing brain cell types is a prerequisite to understanding and controlling neural circuits and is a key aim of the U.S. BRAIN Initiative. The Brain Initiative Cell Census Network (BICCN) leverages recent advances in single cell high-throughput molecular profiling to build a comprehensive atlas of mouse brain cell types. BICCN laboratories are using multiple complementary methods to generate high-resolution profiles of brain cell transcriptomes (RNA-Seq) and epigenomes, including DNA methylation (mC-Seq) and chromatin accessibility (ATAC-Seq). These data will constitute a definitive reference atlas for brain cell types, akin to the human genome reference sequence and annotation, which will have a transformative impact on neuroscience research. Whereas each molecular analysis modality has advanced rapidly and in parallel in terms of resolution (number of genes and/or genomic regions assayed) and throughput (number of cells profiled), computational resources and software for validating and integrating these complementary datasets are currently lacking. Combining data from single cell transcriptome and epigenome profiles, collected in different laboratories, goes beyond the standard tasks of normalization and batch correction. Integrated analysis necessitates a sophisticated framework for merging intimately related but distinct cell type signatures. Moreover, a valid cell type atlas must meet strict standards of statistical and biological reproducibility, both within and across datasets and modalities. In this project, we are developing performant, horizontally scalable, cloud-based solutions for statistical validation of single cell datasets, and for integrating transcriptomic and epigenomic information to identify and characterize neuronal cell types. Our system will enable neuroscientists to quickly and easily compare their datasets and analyses with other public and private data.
Project Leadership
Eran A. Mukamel, Ph.D. (Principal Investigator)
Associate Professor of Cognitive Science
University of California, San Diego
https://brainome.ucsd.edu
Jesse Gillis, Ph.D. (Multiple Principal Investigator)
Associate Professor
Stanley Institute for Cognitive Genomics, Cold Spring Harbor Laboratory
https://www.cshl.edu/research/faculty-staff/jesse-gillis/
Tim Tickle, Ph.D. (Multiple Principal Investigator)
Head of Scientific Partnerships
Data Sciences Platform
The Broad Institute of MIT and Harvard