In recent years, the debate in the field of applications of Deep Learning to Virtual Screening has focused on the use of neural embeddings with respect to classical descriptors in order to encode both structural and physical properties of ligands and/or targets. The attention on embeddings with the increasing use of Graph Neural Networks aimed at overcoming molecular fingerprints that are short range embeddings for atomic neighborhoods. Here, we present EMBER, a novel molecular embedding made by seven molecular fingerprints arranged as different “spectra” to describe the same molecule, and we prove its effectiveness by using deep convolutional architecture that assesses ligands’ bioactivity on a data set containing twenty protein kinases with similar binding sites to CDK1. The data set itself is presented, and the architecture is explained in detail along with its training procedure. We report experimental results and an explainability analysis to assess the contribution of each fingerprint to different targets.

NLRP3 (NOD-like receptor family, pyrin domain-containing protein 3) activation has been linked to several chronic pathologies, including atherosclerosis, type-II diabetes, fibrosis, rheumatoid arthritis, and Alzheimer’s disease. Therefore, NLRP3 represents an appealing target for the development of innovative therapeutic approaches. A few companies are currently working on the discovery of selective modulators of NLRP3 inflammasome. Unfortunately, limited structural data are available for this target. To date, MCC950 represents one of the most promising noncovalent NLRP3 inhibitors. Recently, a possible region for the binding of MCC950 to the NLRP3 protein was described but no details were disclosed regarding the key interactions. In this communication, we present an in silico multiple approach as an insight useful for the design of novel NLRP3 inhibitors. In detail, combining different computational techniques, we propose consensus-retrieved protein residues that seem to be essential for the binding process and for the stabilization of the protein–ligand complex.

The family of protein kinases comprises more than 500 genes involved in numerous functions. Hence, their physiological dysfunction has paved the way toward drug discovery for cancer, cardiovascular, and inflammatory diseases. As a matter of fact, Kinase binding sites high similarity has a double role. On the one hand it is a critical issue for selectivity, on the other hand, according to poly-pharmacology, a synergistic controlled effect on more than one target could be of great pharmacological interest. Another important aspect of binding similarity is the possibility of exploit it for repositioning of drugs on targets of the same family. In this study, we propose our approach called Kinase drUgs mAchine Learning frAmework (KUALA) to automatically identify kinase active ligands by using specific sets of molecular descriptors and provide a multi-target priority score and a repurposing threshold to suggest the best repurposable and non-repurposable molecules. The comprehensive list of all kinase-ligand pairs and their scores can be found at https://github.com/molinfrimed/multi-kinases.

In the last decades, HOX proteins have been extensively studied due to their pivotal role in transcriptional events. HOX proteins execute their activity by exploiting a cooperative binding to PBX proteins and DNA. Therefore, an increase or decrease in HOX activity has been associated with both solid and haematological cancer diseases. Thus, inhibiting HOX-PBX interaction represents a potential strategy to prevent these malignancies, as demonstrated by the patented peptide HTL001 that is being studied in clinical trials. In this work, a computational study is described to identify novel potential peptides designed by employing a database of non-natural amino acids. For this purpose, residue scanning of the HOX minimal active sequence was performed to select the mutations to be further processed. According to these results, the peptides were point-mutated and used for Molecular Dynamics (MD) simulations in complex with PBX1 protein and DNA to evaluate complex binding stability. MM-GBSA calculations of the resulting MD trajectories were exploited to guide the selection of the most promising mutations that were exploited to generate twelve combinatorial peptides. Finally, the latter peptides in complex with PBX1 protein and DNA were exploited to run MD simulations and the ΔG_binding average values of the complexes were calculated. Thus, the analysis of the results highlighted eleven combinatorial peptides that will be considered for further assays.

The COVID-19 pandemic continues to pose a substantial threat to human lives and is likely to do so for years to come. Despite the availability of vaccines, searching for efficient small-molecule drugs that are widely available, including in low- and middle-income countries, is an ongoing challenge. In this work, we report the results of a community effort, the “Billion molecules against Covid-19 challenge”, to identify small-molecule inhibitors against SARS-CoV-2 or relevant human receptors. Participating teams used a wide variety of computational methods to screen a minimum of 1 billion virtual molecules against 6 protein targets. Overall, 31 teams participated, and they suggested a total of 639,024 potentially active molecules, which were subsequently ranked to find ‘consensus compounds’. The organizing team coordinated with various contract research organizations (CROs) and collaborating institutions to synthesize and test 878 compounds for activity against proteases (Nsp5, Nsp3, TMPRSS2), nucleocapsid N, RdRP (Nsp12 domain), and (alpha) spike protein S. Overall, 27 potential inhibitors were experimentally confirmed by binding-, cleavage-, and/or viral suppression assays and are presented here. All results are freely available and can be taken further downstream without IP restrictions. Overall, we show the effectiveness of computational techniques, community efforts, and communication across research fields (i.e., protein expression and crystallography, in silico modeling, synthesis and biological assays) to accelerate the early phases of drug discovery.

AGAPE (computational G-quadruplex Affinity Prediction) is a novel machine learning (ML)-based tool designed to predict the binding and stabilizing potential of small molecules targeting G-quadruplexes (G4s). G4s, prevalent in telomeres and oncogene promoters, are promising therapeutic targets, but designing selective binders remains challenging. Building upon a curated dataset of 1217 compounds labelled through Förster Resonance Energy Transfer (FRET) melting assays, AGAPE integrates 5666 molecular descriptors, both classical and quantum chemical. It captures features relevant to G4 recognition, driving researcher to predict the potential G4 stabilization of small molecules, including both organic ligands and metal complexes. Among the trained ML models, XGBoost achieved the best performance with an accuracy of nearly 91%, using 489 selected features. SHAP analysis highlighted descriptors related to molecular topology, polarizability, and electrostatic potential as key contributors to the classification. AGAPE is deployed through a user-friendly web interface supporting batch prediction and secure data handling and provides a robust and interpretable tool to accelerate the discovery of G4-stabilizing compounds, integrating quantum chemical information within an ML-driven cheminformatics framework.