Explore the vast range of titles available.

Structure-based virtual ligand screening is emerging as a key paradigm for early drug discovery owing to the availability of high-resolution target structures 1 – 4 and ultra-large libraries of virtual compounds 5 , 6 . However, to keep pace with the rapid growth of virtual libraries, such as readily available for synthesis (REAL) combinatorial libraries 7 , new approaches to compound screening are needed 8 , 9 . Here we introduce a modular synthon-based approach—V-SYNTHES—to perform hierarchical structure-based screening of a REAL Space library of more than 11 billion compounds. V-SYNTHES first identifies the best scaffold–synthon combinations as seeds suitable for further growth, and then iteratively elaborates these seeds to select complete molecules with the best docking scores. This hierarchical combinatorial approach enables the rapid detection of the best-scoring compounds in the gigascale chemical space while performing docking of only a small fraction (<0.1%) of the library compounds. Chemical synthesis and experimental testing of novel cannabinoid antagonists predicted by V-SYNTHES demonstrated a 33% hit rate, including 14 submicromolar ligands, substantially improving over a standard virtual screening of the Enamine REAL diversity subset, which required approximately 100 times more computational resources. Synthesis of selected analogues of the best hits further improved potencies and affinities (best inhibitory constant ( K i ) = 0.9 nM) and CB 2 /CB 1 selectivity (50–200-fold). V-SYNTHES was also tested on a kinase target, ROCK1, further supporting its use for lead discovery. The approach is easily scalable for the rapid growth of combinatorial libraries and potentially adaptable to any docking algorithm. V-SYNTHES, a scalable and computationally cost-effective synthon-based approach to compound screening, identified compounds with a high affinity for CB2 and CB1 in a hierarchical structure-based screen of more than 11 billion compounds.

The σ 2 receptor has attracted intense interest in cancer imaging 1 , psychiatric disease 2 , neuropathic pain 3 – 5 and other areas of biology 6 , 7 . Here we determined the crystal structure of this receptor in complex with the clinical candidate roluperidone 2 and the tool compound PB28 8 . These structures templated a large-scale docking screen of 490 million virtual molecules, of which 484 compounds were synthesized and tested. We identified 127 new chemotypes with affinities superior to 1 μM, 31 of which had affinities superior to 50 nM. The hit rate fell smoothly and monotonically with docking score. We optimized three hits for potency and selectivity, and achieved affinities that ranged from 3 to 48 nM, with up to 250-fold selectivity versus the σ 1 receptor. Crystal structures of two ligands bound to the σ 2 receptor confirmed the docked poses. To investigate the contribution of the σ 2 receptor in pain, two potent σ 2 -selective ligands and one potent σ 1 /σ 2 non-selective ligand were tested for efficacy in a mouse model of neuropathic pain. All three ligands showed time-dependent decreases in mechanical hypersensitivity in the spared nerve injury model 9 , suggesting that the σ 2 receptor has a role in nociception. This study illustrates the opportunities for rapid discovery of in vivo probes through structure-based screens of ultra large libraries, enabling study of underexplored areas of biology. Crystal structures of the σ 2 receptor are determined and used to perform a docking screen of nearly 500 million molecules, identifying σ 2 -selective ligands and providing insight into the role of σ 2 in neuropathic pain.

Deep generative neural networks have been used increasingly in computational chemistry for de novo design of molecules with desired properties. Many deep learning approaches employ reinforcement learning for optimizing the target properties of the generated molecules. However, the success of this approach is often hampered by the problem of sparse rewards as the majority of the generated molecules are expectedly predicted as inactives. We propose several technical innovations to address this problem and improve the balance between exploration and exploitation modes in reinforcement learning. In a proof-of-concept study, we demonstrate the application of the deep generative recurrent neural network architecture enhanced by several proposed technical tricks to design inhibitors of the epidermal growth factor (EGFR) and further experimentally validate their potency. The proposed technical solutions are expected to substantially improve the success rate of finding novel bioactive compounds for specific biological targets using generative and reinforcement learning approaches. Deep generative neural networks are increasingly exploited for drug discovery, but often the majority of generated molecules are predicted to be inactive. Here, an optimized protocol for generative models with reinforcement learning is derived and applied to design potent epidermal growth factor inhibitors.

Fragment-based screening can catalyze drug discovery by identifying novel scaffolds, but this approach is limited by the small chemical libraries studied by biophysical experiments and the challenging optimization process. To expand the explored chemical space, we employ structure-based docking to evaluate orders-of-magnitude larger libraries than those used in traditional fragment screening. We computationally dock a set of 14 million fragments to 8-oxoguanine DNA glycosylase (OGG1), a difficult drug target involved in cancer and inflammation, and evaluate 29 highly ranked compounds experimentally. Four of these bind to OGG1 and X-ray crystallography confirms the binding modes predicted by docking. Furthermore, we show how fragment elaboration using searches among billions of readily synthesizable compounds identifies submicromolar inhibitors with anti-inflammatory and anti-cancer effects in cells. Comparisons of virtual screening strategies to explore a chemical space of 10 22 compounds illustrate that fragment-based design enables enumeration of all molecules relevant for inhibitor discovery. Virtual fragment screening is hence a highly efficient strategy for navigating the rapidly growing combinatorial libraries and can serve as a powerful tool to accelerate drug discovery efforts for challenging therapeutic targets. Fragment-based drug design is an efficient yet challenging approach for developing therapeutics. Here, the authors employ structure-based docking screens of vast fragment libraries to identify inhibitors of 8-oxoguanine DNA glycosylase, a difficult drug target implicated in cancer and inflammation.

Virtual library docking can reveal unexpected chemotypes that complement the structures of biological targets. Seeking agonists for the cannabinoid-1 receptor (CB1R), we dock 74 million tangible molecules and prioritize 46 high ranking ones for de novo synthesis and testing. Nine are active by radioligand competition, a 20% hit-rate. Structure-based optimization of one of the most potent of these (K i = 0.7 µM) leads to ‘1350, a 0.95 nM ligand and a full CB1R agonist of G i/o signaling. A cryo-EM structure of ‘1350 in complex with CB1R-G i1 confirms its predicted docked pose. The lead agonist is strongly analgesic in male mice, with a 2-20-fold therapeutic window over hypolocomotion, sedation, and catalepsy and no observable conditioned place preference. These findings suggest that unique cannabinoid chemotypes may disentangle characteristic cannabinoid side-effects from analgesia, supporting the further development of cannabinoids as pain therapeutics. Tummino et al. dock 74 million molecules against the human cannabinoid-1 receptor to find uM ligands. Optimization led to a nM agonist conferring analgesia with reduced side effects in mice, highlighting its potential as a pain therapeutic and the promise of a structure-based approach.

Parallel Groebke–Blackburn–Bienaymé reaction was evaluated as a source of multimillion chemically accessible chemical space. Two most popular classical protocols involving the use of Sc(OTf) 3 and TsOH as the catalysts were tested on a broad substrate scope, and prevalence of the first method was clearly demonstrated. Furthermore, the scope and limitations of the procedure were established. A model 790-member library was obtained with 85% synthesis success rate. These results were used to generate a 271-Mln. readily accessible (REAL) heterocyclic chemical space mostly containing unique chemotypes, which was confirmed by comparative analysis with commercially available compound collections. Meanwhile, this chemical space contained 432 compounds that already showed biological activity according to the ChEMBL database.

On average, an approved drug currently costs US$2–3 billion and takes more than 10 years to develop 1 . In part, this is due to expensive and time-consuming wet-laboratory experiments, poor initial hit compounds and the high attrition rates in the (pre-)clinical phases. Structure-based virtual screening has the potential to mitigate these problems. With structure-based virtual screening, the quality of the hits improves with the number of compounds screened 2 . However, despite the fact that large databases of compounds exist, the ability to carry out large-scale structure-based virtual screening on computer clusters in an accessible, efficient and flexible manner has remained difficult. Here we describe VirtualFlow, a highly automated and versatile open-source platform with perfect scaling behaviour that is able to prepare and efficiently screen ultra-large libraries of compounds. VirtualFlow is able to use a variety of the most powerful docking programs. Using VirtualFlow, we prepared one of the largest and freely available ready-to-dock ligand libraries, with more than 1.4 billion commercially available molecules. To demonstrate the power of VirtualFlow, we screened more than 1 billion compounds and identified a set of structurally diverse molecules that bind to KEAP1 with submicromolar affinity. One of the lead inhibitors (iKeap1) engages KEAP1 with nanomolar affinity (dissociation constant ( K d ) = 114 nM) and disrupts the interaction between KEAP1 and the transcription factor NRF2. This illustrates the potential of VirtualFlow to access vast regions of the chemical space and identify molecules that bind with high affinity to target proteins. VirtualFlow, an open-source drug discovery platform, enables the efficient preparation and virtual screening of ultra-large ligand libraries to identify molecules that bind with high affinity to target proteins.

Drug discovery starts with the identification of a “hit” compound that, following a long and expensive optimization process, evolves into a drug candidate. Bigger screening collections increase the odds of finding more and better hits. For this reason, large pharmaceutical companies have invested heavily in high-throughput screening (HTS) collections that can contain several million compounds. However, this figure pales in comparison with the emergent on-demand chemical collections, which have recently reached the trillion scale. These chemical collections are potentially transformative for drug discovery, as they could deliver many diverse and high-quality hits, even reaching lead-like starting points. But first, it will be necessary to develop computational tools capable of efficiently navigating such massive virtual collections. To address this challenge, we have conceived an innovative strategy that explores the chemical universe from the bottom up, performing a systematic search on the fragment space (exploration phase), to then mine the most promising areas of on-demand collections (exploitation phase). Using a hierarchy of increasingly sophisticated computational methods to remove false positives, we maximize the success probability and minimize the overall computational cost. A basic implementation of the concept has enabled us to validate the strategy prospectively, allowing the identification of new BRD4 (BD1) binders with potencies comparable to stablished drug candidates. The vast scale of emerging on-demand chemical collections presents a challenge for efficiently identifying promising drug candidates. Here, the authors develop a bottom-up computational strategy to first explore fragment space and then exploit the most promising scaffolds, successfully identifying diverse and potent BRD4 binders.

Three promising antibacterial peptides were studied with regard to their ability to inhibit the growth and kill the cells of clinical strains of Staphylococcus aureus , Enterococcus faecalis and Enterococcus faecium . The multifunctional gramicidin S (GS) was the most potent, compared to the membranotropic temporin L (TL), being more effective than the innate-defence regulator IDR-1018 (IDR). These activities, compared across 16 strains as minimal bactericidal and minimal inhibitory concentrations (MIC), are independent of bacterial resistance pattern, phenotype variations and/or biofilm-forming potency. For S. aureus strains, complete killing is accomplished by all peptides at 5 × MIC. For E. faecalis strains, only GS exhibits a rapid bactericidal effect at 5 × MIC, while TL and IDR require higher concentrations. The biofilm-preventing activities of all peptides against the six strains with the largest biofilm biomass were compared. GS demonstrates the lowest minimal biofilm inhibiting concentrations, whereas TL and IDR are consistently less effective. In mature biofilms, only GS completely kills the cells of all studied strains. We compare the physicochemical properties, membranolytic activities, model pharmacokinetics and eukaryotic toxicities of the peptides and explain the bactericidal, antipersister and antibiofilm activities of GS by its elevated stability, pronounced cell-penetration ability and effective utilization of multiple modes of antibacterial action.

An implementation of the three-component one-pot approach to unsymmetrical 1,3,5-trisubstituted-1,2,4-triazoles into combinatorial chemistry is described. The procedure is based on the coupling of amidines with carboxylic acids and subsequent cyclization with hydrazines. After the preliminary assessment of the reagent scope, the method had 81% success rate in parallel synthesis. It was shown that over a billion-sized chemical space of readily accessible (“REAL”) compounds may be generated based on the proposed methodology. Analysis of physicochemical parameters shows that the library contains significant fractions of both drug-like and “beyond-rule-of-five” members. More than 10 million of accessible compounds meet the strictest lead-likeness criteria. Additionally, 195 Mln of sp3-enriched compounds can be produced. This makes the proposed approach a valuable tool in medicinal chemistry.Graphic abstract