Up to 97% of drug-indication pairs that are tested in clinical trials in oncology never advance to receive FDA approval. Mischaracterization of the mechanisms-of-action of these therapies likely contributes to this extremely high failure rate. We have developed a set of genomic techniques to aid in the identification of the true target(s) of small-molecule cancer drugs. By applying these approaches, we can shed light on the likely reasons why certain drugs failed during clinical testing and we can identify suitable patient populations for future drug re-purposing studies.

The systematic profiling of small molecules across information-rich cellular assays has revealed unexpected drug activities. My talk will describe how perturbational gene expression and cell viability profiling can be coupled with genome-scale CRISPR/Cas9 genetic modifier screens to discover novel anti-cancer drug mechanisms. I will also provide an overview of the Drug Repurposing Hub compound library and information resource.

The identification of drug-resistant mutants in an enzyme of interest is a key step to understanding the functional relationship between the enzyme and an inhibitor. Such mutants can inform the design of next-generation inhibitors, or can be used to confirm that a drug's phenotype is on-target. However, discovering such mutants is often challenging, particularly when no crystal structure of the protein bound to the molecule exists. In this talk, we will explore the use of CRISPR base editor technology as a novel and unbiased tool to functionally identify drug-resistant enzyme mutants.

Target identification and validation is a key aspect of the drug discovery paradigm. With the advent of various genome editing and artificial intelligence tools, we live in a world where the discovery of novel validated targets for treatment of human disease can be greatly augmented. In my talk, I will review the key challenges and the unprecedented opportunities we face in target discovery. I will exemplify this with a few real-life project examples.

Functionally characterizing the genetic alterations that drive pancreatic cancer progression is a prerequisite for Precision Medicine. Here, we developed a somatic CRISPR/Cas9 mutagenesis screen to assess the transforming potential of 125 recurrently mutated ‘long-tail’ pancreatic cancer genes, which revealed USP15 and SCAF1 as novel and potent PDAC tumor suppressors. Interestingly, we found that USP15 functioning in a haploinsufficient manner and that the SCAF1-USP15 axis regulates sensitivity to PARPi and gemcitabine.

While CRISPR screens are helping uncover genes regulating many cell-intrinsic processes, existing approaches are suboptimal for identifying gene functions operating extracellularly or within a tissue context. To address this, we recently developed an approach for spatial functional genomics called Perturb-map, which enables resolution of CRISPR screens by multiplex tissue imaging and spatial transcriptomics. We are now applying Perturb-map to identify genetic determinants of ovarian tumor growth, metastasis, and immune composition.

Pseudouridine (Ψ) is an abundant mRNA modification in mammalian transcriptome, but its functions have remained elusive due to the difficulty of transcriptome-wide mapping. We developed a nanopore native RNA sequencing method for quantitative Ψ prediction (NanoPsu). Biologically we find interferon inducible Ψ modifications in interferon stimulated gene transcripts which is consistent with a role of Ψ in enabling efficacy of mRNA vaccines.

There are over 7,000 known rare diseases and over 80% have a simple Mendelian pattern of inheritance. Despite our knowledge of the genetic mutations that cause rare diseases, little is known of the cellular consequences of these genetic mutations. To bridge the gap between genetic and functional information, I systematically phenotyped a collection of human rare pathogenic genetic variants for protein localization, protein interactions, protein degradation in the endoplasmic reticulum, and upregulation of the unfolded protein response (UPR). The data collected from this study, combined with previously generated data on protein-protein interactions, protein-DNA interactions and protein stability, will generate a unique, publicly available database for molecular phenotypes in Mendelian disease. Importantly, this resource will provide a platform for future efforts in drug discovery.

The oncogenic transformation of normal tissue and its ultimate metastatic dissemination are believed to be driven by clonal competition resulting in the success of some lineages and loss of others. Understanding the molecular mechanisms governing clonal dominance is critical for the identification of points for therapeutic intervention. We discuss the development and application of novel genome editing tools to map and interrogate lineage in models of colon and pancreatic adenocarcinomas.

In recent years, we have seen significant advancements in the field of precision medicine, primarily in oncology settings where druggable driver mutations are present. New genomic technologies hold great promise for the identification of actionable drug targets and associated biomarkers for several complex diseases, such as autoimmunity. However, the discovery of novel targets in these settings is often complicated by multigenic effects and the involvement of multiple cell types in disease progression. Our approach uses single-cell RNAseq and machine learning to elucidate the precise cell types involved in the progression of complex diseases and to identify novel therapeutic targets.

In this talk, I will describe our continued efforts to use genomics and AI to identify the targets of compounds that may not have entirely known mechanisms of action. I will describe how these approaches can be used to screen libraries of compounds in silico to uncover repositioning opportunities. I will then describe the successful application to an anticancer compound, followed by precise clinical positioning in pediatric brain cancers.

Recursion is a clinical-stage pharmatech company, mapping human biology at scale to bring better medicines to patients. Enabled by 14 petabytes of imaging and other omics data, we use deep learning to build biological representations across multiple cell types, a whole-genome CRISPR library, and nearly 1 million compounds. Here, we demonstrate the capability of this platform to model complex disease and identify and optimize compounds as potential cancer therapies. Integrating recent insights on genetic correlates of drug response, we demonstrate the application of inferential search to uncover novel mechanisms and small molecules that enhance response in immunotherapy and HRD- cancers.