Machine learning-guided design of potent darunavir analogs targeting HIV-1 proteases: A computational approach for antiretroviral drug discovery

In the pursuit of novel antiretroviral therapies for human immunodeficiency virus type-1 (HIV-1) proteases (PRs), recent improvements in drug discovery have embraced machine learning (ML) techniques to guide the design process. This study employs ensemble learning models to identify crucial substructures as significant features for drug development. Using molecular docking techniques, a collection of 160 darunavir (DRV) analogs was designed based on these key substructures and subsequently screened using molecular docking techniques. Chemical structures with high fitness scores were selected, combined, and one-dimensional (1D) screening based on beyond Lipinski's rule of five (bRo5) and ADME (absorption, distribution, metabolism, and excretion) prediction implemented in the Combined Analog generator Tool (CAT) program. A total of 473 screened analogs were subjected to docking analysis through convolutional neural networks scoring function against both the wild-type (WT) and 12 major mutated PRs. DRV analogs with negative changes in binding free energy (ΔΔ G bind ) compared to DRV could be categorized into four attractive groups based on their interactions with the majority of vital PRs. The analysis of interaction profiles revealed that potent designed analogs, targeting both WT and mutant PRs, exhibited interactions with common key amino acid residues. This observation further confirms that the ML model-guided approach effectively identified the substructures that play a crucial role in potent analogs. It is expected to function as a powerful computational tool, offering valuable guidance in the identification of chemical substructures for synthesis and subsequent experimental testing.

Profiling the interaction of 1-phenylbenzimidazoles to cyclooxygenases

In the design of 1-phenylbenzimidazoles as model cyclooxygenase (COX) inhibitors, docking to a series of crystallographic COX structures was performed to evaluate their potential for high-affinity binding and to reproduce the interaction profile of well-known COX inhibitors. The effect of ligand-specific induced fit on the calculations was also studied. To quantitatively compare the pattern of interactions of model compounds to the profile of several cocrystallized COX inhibitors, a geometric parameter, denominated ligand-receptor contact distance (LRCD), was developed. The interaction profile of several model complexes showed similarity to the profile of COX complexes with inhibitors such as iodosuprofen, iodoindomethacin, diclofenac, and flurbiprofen. Shaping of high-affinity binding sites upon ligand-specific induced fit mostly determined both the affinity and the binding mode of the ligands in the docking calculations. The results suggest potential of 1-phenylbenzimidazole derivatives as COX inhibitors on the basis of their predicted affinity and interaction profile to COX enzymes. The analyses also provided insights into the role of induced fit in COX enzymes. While inhibitors produce different local structural changes at the COX ligand binding site, induced fit allows inhibitors in diverse chemical classes to share characteristic interaction patterns that ensure key contacts to be achieved. Different interaction patterns may also be associated with different inhibitory mechanisms.

In silico identification of potential drug compound against Peroxisome proliferator-activated receptor-gamma by virtual screening and toxicity studies for the treatment of Diabetic Nephropathy

Diabetic Nephropathy is a serious complication of diabetes mellitus. Current therapeutic strategies of Diabetic Nephropathy are based on control of modifiable risks like hypertension, glucose levels, and dyslipidemia. Peroxisome proliferator activated receptor-gamma (PPAR-γ) is implicated in several metabolic syndromes including Diabetic Nephropathy, besides obesity, insulin insensitivity, dislipidemia, inflammation, and hypertension. In the present study, virtual screening of 617 compounds from two different public databases was done against PPAR-γ with an objective to find a possible lead compound. Two softwares, PyRx and iGEMDOCK, were used to achieve the docking accuracy in order to avoid loss of candidate compounds. Rosiglitazone (used to treat Diabetic Nephropathy) was taken as the standard compound. A total of 30 compounds with good binding affinity with PPAR-γ were selected for further filtering, on the basis of absorption, distribution, metabolism, excretion, and toxicity (ADMET). The interaction profiling of these 30 compounds, showed a minimum of one and maximum of three interactions with reference to rosiglitazone (SER-289, HIS-449, HIS-323, TYR-473). The fulfilling of ADMET analysis criteria of 30 compounds led to the selection of four compounds (ZINC ID 00181552, 00276456, 00298314, 00448009). Molecular dynamics simulation of these lead compounds in complex with PPAR-γ revealed that three of the four compounds formed a stable complex in the ligand-binding pocket of PPAR-γ during 20-ns simulation. Hence, these three (ZINC ID 00181552, 00276456, 00298314) of the four compounds are potential candidates for experimental validation of biological activity against PPAR-γ in future drug discovery studies.

Specificity of broad protein interaction surfaces for proteins with multiple binding partners

Analysis of protein-protein interaction networks has revealed the presence of proteins with multiple interaction ligand proteins, such as hub proteins. For such proteins, multiple ligands would be predicted as interacting partners when predicting all-to-all protein-protein interactions (PPIs). In this work, to obtain a better understanding of PPI mechanisms, we focused on protein interaction surfaces, which differ between protein pairs. We then performed rigid-body docking to obtain information of interfaces of a set of decoy structures, which include many possible interaction surfaces between a certain protein pair. Then, we investigated the specificity of sets of decoy interactions between true binding partners in each case of alpha-chymotrypsin, actin, and cyclin-dependent kinase 2 as test proteins having multiple true binding partners. To observe differences in interaction surfaces of docking decoys, we introduced broad interaction profiles (BIPs), generated by assembling interaction profiles of decoys for each protein pair. After cluster analysis, the specificity of BIPs of true binding partners was observed for each receptor. We used two types of BIPs: those involved in amino acid sequences (BIP-seqs) and those involved in the compositions of interacting amino acid residue pairs (BIP-AAs). The specificity of a BIP was defined as the number of group members including all true binding partners. We found that BIP-AA cases were more specific than BIP-seq cases. These results indicated that the composition of interacting amino acid residue pairs was sufficient for determining the properties of protein interaction surfaces.