Denoising DNA Encoded Library Screens with Sparse Learning

Compositional Deep Probabilistic Models of DNA-Encoded Libraries

DNA-encoded library (DEL) has proven to be a powerful tool that utilizes combinatorially constructed small molecules to facilitate highly efficient screening experiments. These selection experiments, involving multiple stages of washing, elution, and identification of potent binders via unique DNA barcodes, often generate complex data. This complexity can potentially mask the underlying signals, necessitating the application of computational tools, such as machine learning, to uncover valuable insights. We introduce a compositional deep probabilistic model of DEL data, DEL-Compose, which decomposes molecular representations into their monosynthon, disynthon, and trisynthon building blocks and capitalizes on the inherent hierarchical structure of these molecules by modeling latent reactions between embedded synthons. Additionally, we investigate methods to improve the observation models for DEL count data, such as integrating covariate factors to more effectively account for data noise. Across two popular public benchmark data sets (CA-IX and HRP), our model demonstrates strong performance compared to count baselines, enriches the correct pharmacophores, and offers valuable insights via its intrinsic interpretable structure, thereby providing a robust tool for the analysis of DEL data.

Evolution of chemistry and selection technology for DNA-encoded library

DNA-encoded chemical library (DEL) links the power of amplifiable genetics and the non-self-replicating chemical phenotypes, generating a diverse chemical world. In analogy with the biological world, the DEL world can evolve by using a chemical central dogma, wherein DNA replicates using the PCR reactions to amplify the genetic codes, DNA sequencing transcripts the genetic information, and DNA-compatible synthesis translates into chemical phenotypes. Importantly, DNA-compatible synthesis is the key to expanding the DEL chemical space. Besides, the evolution-driven selection system pushes the chemicals to evolve under the selective pressure, i.e., desired selection strategies. In this perspective, we summarized recent advances in expanding DEL synthetic toolbox and panning strategies, which will shed light on the drug discovery harnessing in vitro evolution of chemicals via DEL.

Building Block-Based Binding Predictions for DNA-Encoded Libraries

DNA-encoded libraries (DELs) provide the means to make and screen millions of diverse compounds against a target of interest in a single experiment. However, despite producing large volumes of binding data at a relatively low cost, the DEL selection process is susceptible to noise, necessitating computational follow-up to increase signal-to-noise ratios. In this work, we present a set of informatics tools to employ data from prior DEL screen(s) to gain information about which building blocks are most likely to be productive when designing new DELs for the same target. We demonstrate that similar building blocks have similar probabilities of forming compounds that bind. We then build a model from the inference that the combined behavior of individual building blocks is predictive of whether an overall compound binds. We illustrate our approach on a set of three-cycle OpenDEL libraries screened against soluble epoxide hydrolase (sEH) and report performance of more than an order of magnitude greater than random guessing on a holdout set, demonstrating that our model can serve as a baseline for comparison against other machine learning models on DEL data. Lastly, we provide a discussion on how we believe this informatics workflow could be applied to benefit researchers in their specific DEL campaigns.

Deep Learning Approach for the Discovery of Tumor-Targeting Small Organic Ligands from DNA-Encoded Chemical Libraries

DNA-Encoded Chemical Libraries (DELs) have emerged as efficient and cost-effective ligand discovery tools, which enable the generation of protein-ligand interaction data of unprecedented size. In this article, we present an approach that combines DEL screening and instance-level deep learning modeling to identify tumor-targeting ligands against carbonic anhydrase IX (CAIX), a clinically validated marker of hypoxia and clear cell renal cell carcinoma. We present a new ligand identification and hit-to-lead strategy driven by machine learning models trained on DELs, which expand the scope of DEL-derived chemical motifs. CAIX-screening datasets obtained from three different DELs were used to train machine learning models for generating novel hits, dissimilar to elements present in the original DELs. Out of the 152 novel potential hits that were identified with our approach and screened in an in vitro enzymatic inhibition assay, 70% displayed submicromolar activities (IC₅₀ < 1 μM). To generate lead compounds that are functionalized with anticancer payloads, analogues of top hits were prioritized for synthesis based on the predicted CAIX affinity and synthetic feasibility. Three lead candidates showed accumulation on the surface of CAIX-expressing tumor cells in cellular binding assays. The best compound displayed an in vitro K_D of 5.7 nM and selectively targeted tumors in mice bearing human renal cell carcinoma lesions. Our results demonstrate the synergy between DEL and machine learning for the identification of novel hits and for the successful translation of lead candidates for in vivo targeting applications.

Machine-Learning-Based Data Analysis Method for Cell-Based Selection of DNA-Encoded Libraries

DNA-encoded library (DEL) is a powerful ligand discovery technology that has been widely adopted in the pharmaceutical industry. DEL selections are typically performed with a purified protein target immobilized on a matrix or in solution phase. Recently, DELs have also been used to interrogate the targets in the complex biological environment, such as membrane proteins on live cells. However, due to the complex landscape of the cell surface, the selection inevitably involves significant nonspecific interactions, and the selection data are much noisier than the ones with purified proteins, making reliable hit identification highly challenging. Researchers have developed several approaches to denoise DEL datasets, but it remains unclear whether they are suitable for cell-based DEL selections. Here, we report the proof-of-principle of a new machine-learning (ML)-based approach to process cell-based DEL selection datasets by using a Maximum A Posteriori (MAP) estimation loss function, a probabilistic framework that can account for and quantify uncertainties of noisy data. We applied the approach to a DEL selection dataset, where a library of 7,721,415 compounds was selected against a purified carbonic anhydrase 2 (CA-2) and a cell line expressing the membrane protein carbonic anhydrase 12 (CA-12). The extended-connectivity fingerprint (ECFP)-based regression model using the MAP loss function was able to identify true binders and also reliable structure-activity relationship (SAR) from the noisy cell-based selection datasets. In addition, the regularized enrichment metric (known as MAP enrichment) could also be calculated directly without involving the specific machine-learning model, effectively suppressing low-confidence outliers and enhancing the signal-to-noise ratio. Future applications of this method will focus on de novo ligand discovery from cell-based DEL selections.

Combining DELs and machine learning for toxicology prediction

DNA-encoded libraries (DELs) allow starting chemical matter to be identified in drug discovery. The volume of experimental data generated also makes DELs an attractive resource for machine learning (ML). ML allows modeling complex relationships between compounds and numerical endpoints, such as the binding to a target measured by DELs. DELs could also empower other areas of drug discovery. Here, we propose that DELs and ML could be combined to model binding to off-targets, enabling better predictive toxicology. With enough data, ML models can make accurate predictions across a vast chemical space, and they can be reused and expanded across projects. Although there are limitations, more general toxicology models could be applied earlier during drug discovery, illuminating safety liabilities at a lower cost.

Recent advances in DNA-encoded dynamic libraries

The DNA-encoded chemical library (DEL) has emerged as a powerful technology platform in drug discovery and is also gaining momentum in academic research. The rapid development of DNA-/DEL-compatible chemistries has greatly expanded the chemical space accessible to DELs. DEL technology has been widely adopted in the pharmaceutical industry and a number of clinical drug candidates have been identified from DEL selections. Recent innovations have combined DELs with other legacy and emerging techniques. Among them, the DNA-encoded dynamic library (DEDL) introduces DNA encoding into the classic dynamic combinatorial libraries (DCLs) and also integrates the principle of fragment-based drug discovery (FBDD), making DEDL a novel approach with distinct features from static DELs. In this Review, we provide a summary of the recently developed DEDL methods and their applications. Future developments in DEDLs are expected to extend the application scope of DELs to complex biological systems with unique ligand-discovery capabilities.

Strategies for developing DNA-encoded libraries beyond binding assays

DNA-encoded chemical libraries (DELs) have emerged as a powerful technology in drug discovery. The wide adoption of DELs in the pharmaceutical industry and the rapid advancements of DEL-compatible chemistry have further fuelled its development and applications. In general, a DEL has been considered as a massive binding assay to identify physical binders for individual protein targets. However, recent innovations demonstrate the capability of DELs to operate in the complex milieu of biological systems. In this Perspective, we discuss the recent progress in using DNA-encoded chemical libraries to interrogate complex biological targets and their potential to identify structures that elicit function or possess other useful properties. Future breakthroughs in these aspects are expected to catapult DEL to become a momentous technology platform not only for drug discovery but also to explore fundamental biology.

Sequence Fusion Algorithm of Tumor Gene Sequencing and Alignment Based on Machine Learning

With the rapid development of DNA high-throughput testing technology, there is a high correlation between DNA sequence variation and human diseases, and detecting whether there is variation in DNA sequence has become a hot research topic at present. DNA sequence variation is relatively rare, and the establishment of DNA sequence sparse matrix, which can quickly detect and reason fusion variation point, has become an important work of tumor gene testing. Because there are differences between the current comparison software and mutation detection software in detecting the same sample, there are errors between the results of derivative sequence comparison and the detection of mutation. In this paper, SNP and InDel detection methods based on machine learning and sparse matrix detection are proposed, and VarScan 2, Genome Analysis Toolkit (GATK), BCFtools, and FreeBayes are compared. In the research of SNP and InDel detection with intelligent reasoning, the experimental results show that the detection accuracy and recall rate are better when the depth is increasing. The reasoning fusion method proposed in this paper has certain advantages in comparison effect and discovery in SNP and InDel and has good effect on swelling and pain gene detection.

Analysis of DNA-Encoded Library Screening Data: Selection of Molecules for Synthesis

DNA-encoded library (DEL) screens are used to discover novel chemical matter capable of modulating the activity of pharmaceutically interesting protein targets. DEL selections are accomplished by immobilizing a target protein on a resin and capturing library molecules that bind to the target. The barcodes of the captured library molecules are then amplified and sequenced. This chapter outlines simple methods for visualizing the resulting screening data (using free open-source software), such that enriched molecules can be selected for synthesis and follow-up activity confirmation. Measures of enrichment and the concept of sub-libraries are also illustrated.

DNA-Encoded Chemical Libraries: A Comprehensive Review with Succesful Stories and Future Challenges

DNA-encoded chemical libraries (DELs) represent a versatile and powerful technology platform for the discovery of small-molecule ligands to protein targets of biological and pharmaceutical interest. DELs are collections of molecules, individually coupled to distinctive DNA tags serving as amplifiable identification barcodes. Thanks to advances in DNA-compatible reactions, selection methodologies, next-generation sequencing, and data analysis, DEL technology allows the construction and screening of libraries of unprecedented size, which has led to the discovery of highly potent ligands, some of which have progressed to clinical trials. In this Review, we present an overview of diverse approaches for the generation and screening of DEL molecular repertoires. Recent success stories are described, detailing how novel ligands were isolated from DEL screening campaigns and were further optimized by medicinal chemistry. The goal of the Review is to capture some of the most recent developments in the field, while also elaborating on future challenges to further improve DEL technology as a therapeutic discovery platform.

Recent Advances on the Selection Methods of DNA-Encoded Libraries

DNA-encoded libraries (DEL) have come of age and become a major technology platform for ligand discovery in both academia and the pharmaceutical industry. Technological maturation in the past two decades and the recent explosive developments of DEL-compatible chemistries have greatly improved the chemical diversity of DELs and fueled its applications in drug discovery. A relatively less-covered aspect of DELs is the selection method. Typically, DEL selection is considered as a binding assay and the selection is conducted with purified protein targets immobilized on a matrix, and the binders are separated from the non-binding background via physical washes. However, the recent innovations in DEL selection methods have not only expanded the target scope of DELs, but also revealed the potential of the DEL technology as a powerful tool in exploring fundamental biology. In this Review, we first cover the "classic" DEL selection methods with purified proteins on solid phase, and then we discuss the strategies to realize DEL selections in solution phase. Finally, we focus on the emerging approaches for DELs to interrogate complex biological targets.

Triaging of DNA-Encoded Library Selection Results by High-Throughput Resynthesis of DNA-Conjugate and Affinity Selection Mass Spectrometry

DNA encoded library (DEL) technology allows for rapid identification of novel small-molecule ligands and thus enables early-stage drug discovery. DEL technology is well-established, numerous cases of discovered hit molecules have been published, and the technology is widely employed throughout the pharmaceutical industry. Nonetheless, DEL selection results can be difficult to interpret, as library member enrichment may derive from not only desired products, but also DNA-conjugated byproducts and starting materials. Note that DELs are generally produced using split-and-pool combinatorial chemistry, and DNA-conjugated byproducts and starting materials cannot be removed from the library mixture. Herein, we describe a method for high-throughput parallel resynthesis of DNA-conjugated molecules such that byproducts, starting materials, and desired products are produced in a single pot, using the same chemical reactions and reagents as during library production. The low-complexity mixtures of DNA-conjugate are then assessed for protein binding by affinity selection mass spectrometry and the molecular weights of the binding ligands ascertained. This workflow is demonstrated to be a practical tool to triage and validate potential hits from DEL selection data.

Trends in Hit-to-Lead Optimization Following DNA-Encoded Library Screens

DNA-encoded library (DEL) screens have emerged as a powerful hit-finding tool for a number of biological targets. In this Innovations article, we review published hit-to-lead optimization studies following DEL screens. Trends in molecular property changes from hit to lead are identified, and specific optimization tactics are exemplified in case studies. Across the studies, physicochemical property and structural changes post-DEL screening are similar to those which occur during hit-to-lead optimization following high throughputscreens (HTS). However, unique aspects of DEL-the combinatorial synthetic methods which enable DEL synthesis and the linker effects at the DNA attachment point-impact the strategies and outcomes of hit-to-lead optimizations.

Synthetic Studies toward DNA-Encoded Heterocycles Based on the On-DNA Formation of α,β-Unsaturated Ketones

Taking advantage of the diversity-oriented synthesis strategy with α,β-unsaturated carbonyl compounds, we have successfully established the DNA-compatible transformations for various heterocyclic scaffolds. The ring-closure reactions for pyrrole, pyrrolidine, pyrazole, pyrazoline, isoxazoline, pyridine, piperidine, cyclohexenone, and 5,8-dihydroimidazo[1,2-a]pyrimidine were elegantly demonstrated in a DNA-compatible format. These efforts paved the way for preparing DNA-encoded libraries with more extensive chemical space.

The maturation of DNA encoded libraries: opportunities for new users

DNA-encoded combinatorial libraries (DECLs) represent an exciting new technology for high-throughput screening, significantly increasing its capacity and cost-effectiveness. Historically, DECLs have been the domain of specialized academic groups and industry; however, there has recently been a shift toward more drug discovery academic centers and institutes adopting this technology. Key to this development has been the simplification, characterization and standardization of various DECL subprotocols, such as library design, affinity screening and data analysis of hits. This review examines the feasibility of implementing DECL screening technology as a first-time user, particularly in academia, exploring the some important considerations for this, and outlines some applications of the technology that academia could contribute to the field.