This study introduces a new de novo design algorithm called GENERA that combines the capabilities of a deep-learning algorithm for automated drug-like analogue design, called DeLA-Drug, with a genetic algorithm for generating molecules with desired target-oriented properties. Specifically, GENERA was applied to the angiotensin-converting enzyme 2 (ACE2) target, which is implicated in many pathological conditions, including COVID-19. The ability of GENERA to de novo design promising candidates for a specific target was assessed using two docking programs, PLANTS and GLIDE. A fitness function based on the Pareto dominance resulting from computed PLANTS and GLIDE scores was applied to demonstrate the algorithm's ability to perform multiobjective optimizations effectively. GENERA can quickly generate focused libraries that produce better scores compared to a starting set of known ACE-2 binders. This study is the first to utilize a DL-based algorithm designed for analogue generation as a mutational operator within a GA framework, representing an innovative approach to target-oriented de novo design.

In this paper, we introduce DeLA-DrugSelf, an upgraded version of DeLA-Drug [J. Chem. Inf. Model. 62 (2022) 1411-1424], which incorporates essential advancements for automated multi-objective de novo design. Unlike its predecessor, which relies on SMILES notation for molecular representation, DeLA-DrugSelf employs a novel and robust molecular representation string named SELFIES (SELF-referencing Embedded String). The generation process in DeLA-DrugSelf not only involves substitutions to the initial string representing the starting query molecule but also incorporates insertions and deletions. This enhancement makes DeLA-DrugSelf significantly more adept at executing data-driven scaffold decoration and lead optimization strategies. Remarkably, DeLA-DrugSelf explicitly addresses the SELFIES-related collapse issue, considering only collapse-free compounds during generation. These compounds undergo a rigorous quality metrics evaluation, highlighting substantial advancements in terms of drug-likeness, uniqueness, and novelty compared to the molecules generated by the previous version of the algorithm. To evaluate the potential of DeLA-DrugSelf as a mutational operator within a genetic algorithm framework for multi-objective optimization, we employed a fitness function based on Pareto dominance. Our objectives focused on target-oriented properties aimed at optimizing known cannabinoid receptor 2 (CB2R) ligands. The results obtained indicate that DeLA-DrugSelf, available as a user-friendly web platform (https://www.ba.ic.cnr.it/softwareic/delaself/), can effectively contribute to the data-driven optimization of starting bioactive molecules based on user-defined parameters.

Although the approved vaccines are proving to be of utmost importance in containing the Coronavirus disease 2019 (COVID-19) threat, they will hardly be resolutive as new severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2, a single-stranded RNA virus) variants might be insensitive to the immune response they induce. In this scenario, developing an effective therapy is still a dire need. Different targets for therapeutic antibodies and diagnostics have been identified, among which the SARS-CoV-2 spike (S) glycoprotein, particularly its receptor-binding domain, has been defined as crucial. In this context, we aim to focus attention also on the role played by the S N-terminal domain (S1-NTD) in the virus attachment, already recognized as a valuable target for neutralizing antibodies, in particular, building on a cavity mapping indicating the presence of two druggable pockets and on the recent literature hypothesizing the presence of a ganglioside-binding domain. In this perspective, we aim at proposing S1-NTD as a putative target for designing small molecules hopefully able to hamper the SARS-CoV-2 attachment to host cells.

Targeting pathogenic mechanisms, rather than essential processes, represents a very attractive approach for the development of new antimycobacterial drugs. In this context, iron acquisition routes have recently emerged as potentially druggable pathways. However, the importance of siderophore biosynthesis in the virulence and pathogenicity of M. abscessus (Mab) is still poorly understood. In this study, we investigated the Salicylate Synthase (SaS) of Mab as an innovative molecular target for the development of inhibitors of siderophore production. Notably, Mab-SaS does not have any counterpart in human cells, making it an interesting candidate for drug discovery. Starting from the analysis of the binding of a series of furan-based derivatives, previously identified by our group as inhibitors of MbtI from M. tuberculosis (Mtb), we successfully selected the lead compound 1, exhibiting a strong activity against Mab-SaS (IC₅₀ ≈ 5 µM). Computational studies characterized the key interactions between 1 and the enzyme, highlighting the important roles of Y387, G421, and K207, the latter being one of the residues involved in the first step of the catalytic reaction. These results support the hypothesis that 5-phenylfuran-2-carboxylic acids are also a promising class of Mab-SaS inhibitors, paving the way for the optimization and rational design of more potent derivatives.

Peroxisome proliferator-activated receptor gamma (PPARγ) is a nuclear receptor whose ligand-induced conformational changes, primarily driven by helix 12 (H12) repositioning, regulate transcriptional activity. However, the precise mechanism remains elusive. In this study, we performed classical molecular dynamics (cMD) simulations of the PPARγ ligand binding domain (LBD) in complex with two agonists (BRL, 3EA), a partial agonist (GW0072), and an antagonist (EKP), generating 3 μs trajectories for each system. To gain deeper insights, we integrated machine learning-assisted clustering with MD simulations, revealing a favorable trend in binding free energy (ΔG _b), suggesting enhanced complex stability. A case study on EKP demonstrated that, despite fitting within the binding site, it failed to induce rapid LBD or H12 rearrangements in the apo agonist-induced conformation. Additionally, we investigated the apo-state conformations of PPARγ-LBD influenced by agonist and antagonist ligands, utilizing cMD and Gaussian accelerated molecular dynamics (GaMD) over a cumulative 6 μs (3 μs cMD + 3 μs GaMD). Key residues known to modulate PPARγ function upon mutation were analyzed, and simulations confirmed the high stability of both apo and ligand-bound conformations. Notably, in the apo state, specific H12 residues interacted with other PPARγ-LBD regions, preventing disorder and abrupt transitions. These findings guided the selection of collective variables (CVs) for well-tempered metadynamics (WT-MetaD) simulations, which-in the apo-agonist state-captured the H12 shift from agonist- to antagonist-like conformations, consistent with resolved X-ray structures. Overall, this computational framework provides novel insights into PPARγ-LBD conformational dynamics and establishes a valuable approach for rationally assessing the effects of modulators on PPARγ activity.

The development of small molecules that selectively target the cannabinoid receptor subtype 2 (CB2R) is emerging as an intriguing therapeutic strategy to treat neurodegeneration, as well as to contrast the onset and progression of cancer. In this context, in-silico tools able to predict CB2R affinity and selectivity with respect to the subtype 1 (CB1R), whose modulation is responsible for undesired psychotropic effects, are highly desirable. In this work, we developed a series of machine learning classifiers trained on high-quality bioactivity data of small molecules acting on CB2R and/or CB1R extracted from ChEMBL v30. Our classifiers showed strong predictive power in accurately determining CB2R affinity, CB1R affinity, and CB2R/CB1R selectivity. Among the built models, those obtained using random forest as algorithm proved to be the top-performing ones (AUC in validation ≥0.96) and were made freely accessible through a user-friendly web platform developed ad hoc and called ALPACA (https://www.ba.ic.cnr.it/softwareic/alpaca/). Due to its user-friendly interface and robust predictive power, ALPACA can be a valuable tool in saving both time and resources involved in the design of selective CB2R modulators.