The characterization of cancer genomes has provided insight into somatically altered genes across tumors, transformed our understanding of cancer biology, and enabled tailoring of therapeutic strategies. However, the function of most cancer alleles remains mysterious, and many cancer features transcend their genomes. Consequently, tumor genomic characterization does not influence therapy for most patients. Approaches to understand the function and circuitry of cancer genes provide complementary approaches to elucidate both oncogene and non-oncogene dependencies. Emerging work indicates that the diversity of therapeutic targets engendered by non-oncogene dependencies is much larger than the list of recurrently mutated genes. Here we describe a framework for this expanded list of cancer targets, providing novel opportunities for clinical translation.

DNA-encoded library (DEL) screening and quantitative structure–activity relationship (QSAR) modeling are two techniques used in drug discovery to find novel small molecules that bind a protein target. Applying QSAR modeling to DEL selection data can facilitate the selection of compounds for off-DNA synthesis and evaluation. Such a combined approach has been done recently by training binary classifiers to learn DEL enrichments of aggregated “disynthons” in order to accommodate the sparse and noisy nature of DEL data. However, a binary classification model cannot distinguish between different levels of enrichment, and information is potentially lost during disynthon aggregation. Here, we demonstrate a regression approach to learning DEL enrichments of individual molecules, using a custom negative-log-likelihood loss function that effectively denoises DEL data and introduces opportunities for visualization of learned structure–activity relationships. Our approach explicitly models the Poisson statistics of the sequencing process used in the DEL experimental workflow under a frequentist view. We illustrate this approach on a DEL dataset of 108,528 compounds screened against carbonic anhydrase (CAIX), and a dataset of 5,655,000 compounds screened against soluble epoxide hydrolase (sEH) and SIRT2. Due to the treatment of uncertainty in the data through the negative-log-likelihood loss used during training, the models can ignore low-confidence outliers. While our approach does not demonstrate a benefit for extrapolation to novel structures, we expect our denoising and visualization pipeline to be useful in identifying structure–activity trends and highly enriched pharmacophores in DEL data. Further, this approach to uncertainty-aware regression modeling is applicable to other sparse or noisy datasets where the nature of stochasticity is known or can be modeled; in particular, the Poisson enrichment ratio metric we use can apply to other settings that compare sequencing count data between two experimental conditions.

Target- and phenotype-agnostic assessments of biological activity have emerged as viable strategies for prioritizing scaffolds, structural features, and synthetic pathways in screening sets, with the goal of increasing performance diversity. Here, we describe the synthesis of a small library of functionalized stereoisomeric azetidines and its biological annotation by "cell painting," a multiplexed, high-content imaging assay capable of measuring many hundreds of compound-induced changes in cell morphology in a quantitative and unbiased fashion. Using this approach, we systematically compare the degrees to which a core scaffold's biological activity, inferred from its effects on cell morphology, is affected by variations in stereochemistry and appendages. We show that stereoisomerism and appendage diversification can produce effects of similar magnitude, and that the concurrent use of these strategies results in a broader sampling of biological activity.

Organic chemists are able to synthesize molecules in greater number and chemical complexity than ever before. Yet, a majority of these compounds go untested in biological systems, and those that do are often tested long after the chemist can incorporate the results into synthetic planning. We propose the use of high-dimensional “multiplex” assays, which are capable of measuring thousands of cellular features in one experiment, to annotate rapidly and inexpensively the biological activities of newly synthesized compounds. This readily accessible and inexpensive “real-time” profiling method can be used in a prospective manner to facilitate, for example, the efficient construction of performance-diverse small-molecule libraries that are enriched in bioactives. Here, we demonstrate this concept by synthesizing ten triads of constitutionally isomeric compounds via complexity-generating photochemical and thermal rearrangements and measuring compound-induced changes in cellular morphology via an imaging-based “cell painting” assay. Our results indicate that real-time biological annotation can inform optimization efforts and library syntheses by illuminating trends relating to biological activity that would be difficult to predict if only chemical structure were considered. We anticipate that probe and drug discovery will benefit from the use of optimization efforts and libraries that implement this approach.