Project Summary Linking genotype to phenotype by predicting the functional effects of genomic variation is crucial to realizing the potential of genomic medicine. Over the past two decades, consortium efforts have characterized common and rare population-scale genetic variation and functional gene regulatory elements across cell types. More recently, single-cell technologies have enabled organism-scale surveys of molecular cell states. The availability of these three data types means that the goal of general models to predict the effects of variants is finally within reach. Currently, integrating these diverse biological data sets to build predictive models is difficult. While resources such as RegulomeDB help researchers annotate variants with putative regulatory function, they often lack cell type specificity and predict variant function in a general sense. Similarly, GTEx effectively links specific variants to changes in gene expression, but these variants are primarily SNPs, and the predicted effects are mostly pairwise interactions. Furthermore, previous efforts rely primarily on bulk measurements, with limited exploration of the impact of genomic variation at the single-cell level. We propose quantitative shifts in cellular state as a new paradigm for defining and predicting variant function. Single-cell transcriptomic and epigenomic data from healthy individuals provide a reference atlas of cell states. By comparing cell state distributions against this reference, we can identify quantitative shifts resulting from genetic variation and explore these deviations as potential disease states. We will then build models to predict shifts in cell state by combining single-cell data with background germline genetic variation, chromatin structure, and supporting functional data.

Project Narrative The goal of this project is to develop novel computational methods to predict the impact of multiple changes in DNA sequence on overall cellular state and help IGVF consortium data collection and analysis strategies. Ultimately, we will generate an atlas of not only individual genetic variants to phenotype associations but, more importantly, entire genotype to phenotype predictions. These efforts represent crucial steps toward enabling genomic medicine by comprehensively linking genotype with cellular phenotype.