We are developing & leveraging innovative modeling strategies to identify & translate targets into impact.

Phenotypic drug screening remains constrained by the vastness of chemical space and technical challenges scaling experimental workflows. To overcome these barriers, computational methods have been developed to prioritize compounds, but they rely on either single-task models lacking generalizability or heuristic-based genomic proxies that resist optimization. We designed an active deep-learning framework that leverages omics to enable scalable, optimizable identification of compounds that induce complex phenotypes. Our generalizable algorithm outperformed state-of-the-art models on classical recall, translating to a 13-17x increase in phenotypic hit-rate across two hematological discovery campaigns. Combining this algorithm […]

Inference of cell–cell communication from single-cell RNA sequencing data is a powerful technique to uncover intercellular communication pathways, yet existing methods perform this analysis at the level of the cell type or cluster, discarding single-cell-level information. Here we present Scriabin, a flexible and scalable framework for comparative analysis of cell–cell communication at single-cell resolution that is performed without cell aggregation or downsampling. We use multiple published atlas-scale datasets, genetic perturbation screens and direct experimental validation to show that Scriabin accurately recovers expected cell–cell communication edges and identifies communication networks that […]

The cellular composition of barrier epithelia is essential to organismal homoeostasis. In particular, within the small intestine, adult stem cells establish tissue cellularity, and may provide a means to control the abundance and quality of specialized epithelial cells. Yet, methods for the identification of biological targets regulating epithelial composition and function, and of small molecules modulating them, are lacking. Here we show that druggable biological targets and small-molecule regulators of intestinal stem cell differentiation can be identified via multiplexed phenotypic screening using thousands of miniaturized organoid models of intestinal stem […]

Prognostically relevant RNA expression states exist in pancreatic ductal adenocarcinoma (PDAC), but our understanding of their drivers, stability, and relationship to therapeutic response is limited. To examine these attributes systematically, we profiled metastatic biopsies and matched organoid models at single-cell resolution. In vivo, we identify a new intermediate PDAC transcriptional cell state and uncover distinct site- and state-specific tumor microenvironments (TMEs). Benchmarking models against this reference map, we reveal strong culture-specific biases in cancer cell transcriptional state representation driven by altered TME signals. We restore expression state heterogeneity by adding […]

There is pressing urgency to understand the pathogenesis of the severe acute respiratory syndrome coronavirus clade 2 (SARS-CoV-2) which causes the disease COVID-19. SARS-CoV- 2 spike (S)-protein binds ACE2, and in concert with host proteases, principally TMPRSS2, promotes cellular entry. The cell subsets targeted by SARS-CoV-2 in host tissues, and the factors that regulate ACE2 expression, remain unknown. Here, we leverage human, non-human primate, and mouse single-cell RNA-sequencing (scRNA-seq) datasets across health and disease to uncover putative targets of SARS-CoV-2 amongst tissue-resident cell subsets. We identify ACE2 and TMPRSS2 co-expressing […]

Background Human immunity relies on the coordinated responses of many cellular subsets and functional states. Inter-individual variations in cellular composition and communication could thus potentially alter host protection. Here, we explore this hypothesis by applying single-cell RNA-sequencing to examine viral responses among the dendritic cells (DCs) of three elite controllers (ECs) of HIV-1 infection. Results To overcome the potentially confounding effects of donor-to-donor variability, we present a generally applicable computational framework for identifying reproducible patterns in gene expression across donors who share a unifying classification. Applying it, we discover a […]