ChemMORT: an automatic ADMET optimization platform using deep learning and multi-objective particle swarm optimization.

Drug discovery and development constitute a laborious and costly undertaking. The success of a drug hinges not only good efficacy but also acceptable absorption, distribution, metabolism, elimination, and toxicity (ADMET) properties. Overall, up to 50% of drug development failures have been contributed from undesirable ADMET profiles. As a multiple parameter objective, the optimization of the ADMET properties is extremely challenging owing to the vast chemical space and limited human expert knowledge. In this study, a freely available platform called Chemical Molecular Optimization, Representation and Translation (ChemMORT) is developed for the optimization of multiple ADMET endpoints without the loss of potency (https://cadd.nscc-tj.cn/deploy/chemmort/). ChemMORT contains three modules: Simplified Molecular Input Line Entry System (SMILES) Encoder, Descriptor Decoder and Molecular Optimizer. The SMILES Encoder can generate the molecular representation with a 512-dimensional vector, and the Descriptor Decoder is able to translate the above representation to the corresponding molecular structure with high accuracy. Based on reversible molecular representation and particle swarm optimization strategy, the Molecular Optimizer can be used to effectively optimize undesirable ADMET properties without the loss of bioactivity, which essentially accomplishes the design of inverse QSAR. The constrained multi-objective optimization of the poly (ADP-ribose) polymerase-1 inhibitor is provided as the case to explore the utility of ChemMORT.

3D-SMGE: a pipeline for scaffold-based molecular generation and evaluation.

In the process of drug discovery, one of the key problems is how to improve the biological activity and ADMET properties starting from a specific structure, which is also called structural optimization. Based on a starting scaffold, the use of deep generative model to generate molecules with desired drug-like properties will provide a powerful tool to accelerate the structural optimization process. However, the existing generative models remain challenging in extracting molecular features efficiently in 3D space to generate drug-like 3D molecules. Moreover, most of the existing ADMET prediction models made predictions of different properties through a single model, which can result in reduced prediction accuracy on some datasets. To effectively generate molecules from a specific scaffold and provide basis for the structural optimization, the 3D-SMGE (3-Dimensional Scaffold-based Molecular Generation and Evaluation) work consisting of molecular generation and prediction of ADMET properties is presented. For the molecular generation, we proposed 3D-SMG, a novel deep generative model for the end-to-end design of 3D molecules. In the 3D-SMG model, we designed the cross-aggregated continuous-filter convolution (ca-cfconv), which is used to achieve efficient and low-cost 3D spatial feature extraction while ensuring the invariance of atomic space rotation. 3D-SMG was proved to generate valid, unique and novel molecules with high drug-likeness. Besides, the proposed data-adaptive multi-model ADMET prediction method outperformed or maintained the best evaluation metrics on 24 out of 27 ADMET benchmark datasets. 3D-SMGE is anticipated to emerge as a powerful tool for hit-to-lead structural optimizations and accelerate the drug discovery process.

Integrating transformers and many-objective optimization for drug design.

BACKGROUND: Drug design is a challenging and important task that requires the generation of novel and effective molecules that can bind to specific protein targets. Artificial intelligence algorithms have recently showed promising potential to expedite the drug design process. However, existing methods adopt multi-objective approaches which limits the number of objectives. RESULTS: In this paper, we expand this thread of research from the many-objective perspective, by proposing a novel framework that integrates a latent Transformer-based model for molecular generation, with a drug design system that incorporates absorption, distribution, metabolism, excretion, and toxicity prediction, molecular docking, and many-objective metaheuristics. We compared the performance of two latent Transformer models (ReLSO and FragNet) on a molecular generation task and show that ReLSO outperforms FragNet in terms of reconstruction and latent space organization. We then explored six different many-objective metaheuristics based on evolutionary algorithms and particle swarm optimization on a drug design task involving potential drug candidates to human lysophosphatidic acid receptor 1, a cancer-related protein target. CONCLUSION: We show that multi-objective evolutionary algorithm based on dominance and decomposition performs the best in terms of finding molecules that satisfy many objectives, such as high binding affinity and low toxicity, and high drug-likeness. Our framework demonstrates the potential of combining Transformers and many-objective computational intelligence for drug design.

A robust computational quest: Discovering potential hits to improve the treatment of pyrazinamide-resistant Mycobacterium tuberculosis.

The rise of pyrazinamide (PZA)-resistant strains of Mycobacterium tuberculosis (MTB) poses a major challenge to conventional tuberculosis (TB) treatments. PZA, a cornerstone of TB therapy, must be activated by the mycobacterial enzyme pyrazinamidase (PZase) to convert its active form, pyrazinoic acid, which targets the ribosomal protein S1. Resistance, often associated with mutations in the RpsA protein, complicates treatment and highlights a critical gap in the understanding of structural dynamics and mechanisms of resistance, particularly in the context of the G97D mutation. This study utilizes a novel integration of computational techniques, including multiscale biomolecular and molecular dynamics simulations, physicochemical and medicinal chemistry predictions, quantum computations and virtual screening from the ZINC and Chembridge databases, to elucidate the resistance mechanism and identify lead compounds that have the potential to improve treatment outcomes for PZA-resistant MTB, namely ZINC15913786, ZINC20735155, Chem10269711, Chem10279789 and Chem10295790. These computational methods offer a cost-effective, rapid alternative to traditional drug trials by bypassing the need for organic subjects while providing highly accurate insight into the binding sites and efficacy of new drug candidates. The need for rapid and appropriate drug development emphasizes the need for robust computational analysis to justify further validation through in vitro and in vivo experiments.

Identification of Helicobacter pylori-carcinogenic TNF-alpha-inducing protein inhibitors via daidzein derivatives through computational approaches.

Gastric cancer is considered a class 1 carcinogen that is closely linked to infection with Helicobacter pylori (H. pylori), which affects over 1 million people each year. However, the major challenge to fight against H. pylori and its associated gastric cancer due to drug resistance. This research gap had led our research team to investigate a potential drug candidate targeting the Helicobacter pylori-carcinogenic TNF-alpha-inducing protein. In this study, a total of 45 daidzein derivatives were investigated and the best 10 molecules were comprehensively investigated using in silico approaches for drug development, namely pass prediction, quantum calculations, molecular docking, molecular dynamics simulations, Lipinski rule evaluation, and prediction of pharmacokinetics. The molecular docking study was performed to evaluate the binding affinity between the target protein and the ligands. In addition, the stability of ligand-protein complexes was investigated by molecular dynamics simulations. Various parameters were analysed, including root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), radius of gyration (Rg), hydrogen bond analysis, principal component analysis (PCA) and dynamic cross-correlation matrix (DCCM). The results has confirmed that the ligand-protein complex CID: 129661094 (07) and 129664277 (08) formed stable interactions with the target protein. It was also found that CID: 129661094 (07) has greater hydrogen bond occupancy and stability, while the ligand-protein complex CID 129664277 (08) has greater conformational flexibility. Principal component analysis revealed that the ligand-protein complex CID: 129661094 (07) is more compact and stable. Hydrogen bond analysis revealed favourable interactions with the reported amino acid residues. Overall, this study suggests that daidzein derivatives in particular show promise as potential inhibitors of H. pylori.

In Silico Prediction of Potential Inhibitors for Targeting RNA CAG Repeats via Molecular Docking and Dynamics Simulation: A Drug Discovery Approach.

Spinocerebellar ataxia (SCA) is a rare neurological illness inherited dominantly that causes severe impairment and premature mortality. While each rare disease may affect individuals infrequently, collectively they pose a significant healthcare challenge. It is mainly carried out due to the expansion of RNA triplet (CAG) repeats, although missense or point mutations can also be induced. Unfortunately, there is no cure; only symptomatic treatments are available. To date, SCA has about 48 subtypes, the most common of these being SCA 1, 2, 3, 6, 7, 12, and 17 having CAG repeats. Using molecular docking and molecular dynamics (MD) simulation, this study seeks to investigate effective natural herbal neuroprotective compounds against CAG repeats, which are therapeutically significant in treating SCA. Initially, virtual screening followed by molecular docking was used to estimate the binding affinity of neuroprotective natural compounds toward CAG repeats. The compound with the highest binding affinity, somniferine, was then chosen for MD simulation. The structural stability, interaction mechanism, and conformational dynamics of CAG repeats and somniferine were investigated via MD simulation. The MD study revealed that during the simulation period, the interaction between CAG repeats and somniferine stabilizes and results in fewer conformational variations. This in silico study suggests that Somniferine can be used as a therapeutic medication against RNA CAG repeats in SCA.

Identification of coumarin - benzimidazole hybrids as potential antibacterial agents: Synthesis, in vitro and in vivo biological assessment, and ADMET prediction.

The direct-linked coumarin-benzimidazole hybrids, featuring aryl and n-butyl substituents at the N1-position of benzimidazole were synthesized through a Knoevenagel condensation reaction. This reaction involved the condensation of 1,2-diaminobenzene derivatives with coumarin-3-carboxylic acids in the presence of polyphosphoric acid (PPA) at 154 °C. The in vitro antibacterial potency of the hybrid molecules against different gram-positive and gram-negative bacterial strains led to the identification of the hybrids 6m and 6p with a MIC value of 6.25 µg/mL against a gram-negative bacterium, Klebsiella pneumonia ATCC 27736. Cell viability studies on THP-1 cells demonstrated that the compounds 6m and 6p were non-toxic at a concentration of 50 µM. Furthermore, in vivo efficacy studies using a murine neutropenic thigh infection model revealed that both compounds significantly reduced bacterial (Klebsiella pneumonia ATCC 27736) counts (more than 2 log) compared to the control group. Additionally, both compounds exhibited favorable physicochemical properties and drug-likeness characteristics. Consequently, these compounds hold promise as lead candidates for further development of effective antibacterial drugs.

Synthesis and cytotoxicity of novel 6,8,9-trisubstituted purine analogs against liver cancer cells.

A series of novel 6-(substituted phenyl piperazine)-8-(4-substituted phenyl)-9-cyclopentyl purines, 10-51, were synthesized by a four-step synthesis, achieving an overall yield of about 43 %. The reaction conditions were effectively optimized, and the final products were obtained with high purity and yield in all synthesis steps. The synthesized nucleobases were evaluated for their in vitro cytotoxic activities on selected human cancer cell lines (HUH7 (liver), HCT116 (colon), and MCF7 (breast)) using the Sulforhodamine B (SRB) assay. Among these analogs, compounds bearing 4-trifluoromethyl phenyl (19, 20 and 21), 4-methoxy phenyl (27) and 4-fluoro phenyl (34) substitutions at C-8 of purine were the most potent, and they were also analyzed in drug-resistance and drug-sensitive hepatocellular cancer cell (HCC) panels. Compound 19 displayed remarkable anticancer activities (IC50 = 2.9-9.3 µM) against Huh7, FOCUS, SNU475, SNU182, HepG2, and Hep3B cells compared to the positive control, Fludarabine. Additionally, the pharmacological properties and toxicity profiles of the molecules were investigated computationally by the Swiss-ADME and Pro-Tox II online tools, respectively. Results showed that our compounds have favorable physicochemical characteristics for oral bioavailability and do not reveal any toxicity endpoints such as carcinogenicity, immunotoxicity, mutagenicity, or cytotoxicity.

Chemoinformatics Study of Benzodiazepine-1, 2, 3-triazole Derivatives Targeting Butyrylcholinesterase.

This study aims to assess the potential bioactivity of newly designed benzodiazepine-1,2,3-triazole derivatives using in-silico methodologies, with a primary focus on elucidating their inhibitory interactions with the butyrylcholinesterase (BuChE) enzyme, which is implicated in Alzheimer's disease. We employed multiple linear regression (MLR) methods to conduct a quantitative structure-activity relationship (QSAR) analysis on a collection of 31 benzodiazepine-1,2,3-triazole derivatives, with the goal of investigating, assessing, and predicting their activities, as well as designing novel compounds. This approach yielded highly accurate results, with coefficients of determination (R²) of 0.77 and 0.81 for the training and test datasets, respectively. Additionally, the optimized compounds were subjected to an Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) analysis, demonstrating their potential as non-hepatotoxic agents with enhanced absorption and blood-brain barrier permeability. To further validate these findings, the most favorable docking conformations were analyzed using molecular dynamics (MD) simulations with GROMACS software, predicting the stability of the formed complexes. These simulations underscored the critical role of hydrogen bonds in stabilizing the compounds at the BuChE receptor binding site. The results hold great promise for the development of innovative benzodiazepine-1,2,3-triazole derivatives as effective BuChE inhibitors, potentially leading to therapeutic interventions for Alzheimer's disease.

Synthesis and Characterization of Zn(II) Complex of 4-chloro-2-(((2-phenoxyphenyl)imino)methyl)phenol and its Biological Efficacies: DNA Interaction, ADMET, DFT and Molecular Docking Study.

Condensing 2-phenoxyaniline with 5-chlorosalicyldehyde under reflux conditions, a 4-chloro-2-(((2-phenoxyphenyl)imino)methyl)phenol Schiff base has been Synthesized. A zinc complex was synthesized by combining the ligand in a 1:1 molar ratio with zinc sulphateheptahydrate. Mass spectroscopy, NMR, infrared, and elemental analysis were used to characterize the ligand and zinc complex. By measuring the molar conductance, the non-electrolytic character of the complex was confirmed. The zinc ion is coordinated in a pentadentate manner, according to an IR and NMR investigation. Viscosity measurements, absorption and fluorescence spectroscopy were utilized to examine the complex's interaction with CT (calf thymus) DNA. Furthermore, the ligand and complex's ADMET characteristics were ascertained through the use of ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) study. Calculation of the different electronic parameters of the optimized structure through Density Functional Theory (DFT) indicated the stability of the Zn(II) complex. Molecular docking study reflected the future opportunity for the consideration of Zn(II) complex to fight against Alzheimer and Glaucoma diseases.

Synthesis of Novel [{(2-Amino-5-Nitro-N-[(E)-Thiophen-2-yl-Methylidene]Aniline-κ3N1:N4:S)(Sulphato-κ2O1:O3)}Zinc(II)] Complex with Physico-Chemical and Biological Perspective Exploration: A Combined Experimental and Computational Studies.

A novel metal complex was synthesized using freshly prepared 2-Amino-5-nitro-N-[(E)-thiophen-2-yl-methylidene]aniline ligand with Zn (II) sulphate heptahydrate in a 1:1 molar ratio. The ligand and the complex were characterized using different spectroscopic techniques, and the complex was assigned a distorted square pyramidal geometry. Additionally, DNA binding assays and antibacterial activity were used to assess the biological perspectives for the synthesized complex, including the ligand and complex which was further confirmed by molecular docking. Fluorescence Spectroscopy, viscosity measurement, and adsorption measurement were used to investigate the interaction of the Zn (II) complex with CT-DNA. A comparative in vitro antibacterial activity study against Escherichia coli, Klebsiella pneumoniae, Bacillus subtilis, and Staphylococcus aureus strains were studied with free ligand and Zn (II) metal complex. The stable geometry of the complex was additionally established through computational simulation utilizing density functional theory, which was followed by the calculation of several electronic properties. The ADMET characteristics of the complex and ligand were also assessed using ADMET analysis. The in-silico ADMET properties pointed to a significant drug-likeness feature in the synthesized compounds, based on the Lipinski criteria.

Iodoquinazoline-derived VEGFR-2 and EGFRT790M dual inhibitors: Design, synthesis, molecular docking and anticancer evaluations.

Herein, we report the synthesis of a series of new fourteen iodoquinazoline derivatives 7a-c to 13a-e and their evaluation as potential anticancer agents via dual targeting of EGFRT790M and VEGFR-2. The new derivatives were designed according to the target receptors structural requirements. The compounds were evaluated for their cytotoxicity against HepG2, MCF-7, HCT116 and A549 cancer cell lines using MTT assay. Compound 13e showed the highest anticancer activities with IC50 = 5.70, 7.15, 5.76 and 6.50 µM against HepG2, MCF-7, HCT116 and A549 cell lines correspondingly. Compounds 7c, 9b and 13a-d exhibited very good anticancer effects against the tested cancer cell lines. The highly effective six derivatives 7c, 10, 13b, 13c, 13d and 13e were examined against VERO normal cell lines to estimate their cytotoxic capabilities. Our conclusion revealed that compounds 7c, 10, 13b, 13c, 13d and 13e possessed low toxicity against VERO normal cells with IC50 prolonging from 41.66 to 53.99 µM. Also compounds 7a-c to 13a-e were further evaluated for their inhibitory activity against EGFRT790M and VEGFR-2. Also, their ability to bind with both EGFR and VEGFR-2 receptors was examined by molecular modeling. Compounds 13e, 13d, 7c and 13c excellently inhibited VEGFR-2 activity with IC50 = 0.90, 1.00, 1.25 and 1.50 µM respectively. Moreover, Compounds 13e, 7c, 10 and 13d excellently inhibited EGFRT790M activity with IC50 = 0.30, 0.35, 0.45 and 0.47 µM respectively. Finally, our derivatives 7b, 13d and 13e showed good in silico calculated ADMET profile.

Novel multi-target angiogenesis inhibitors as potential anticancer agents: Design, synthesis and preliminary activity evaluation.

Based on the crucial role of histone deacetylase (HDAC) and receptor tyrosine kinase in angiogenesis, in situ assembly, skeletal transition, molecular hybridization, and pharmacophore fusion were employed to yield seventy-six multi-target angiogenesis inhibitors. Biological evaluation indicated that most of the compounds exhibited potent proliferation inhibitory activity on MCF-7 cells, with the TH series having the highest inhibitory activity on MCF-7 cells. In addition, the IC50 values of TA11 and TH3 against HT-29 cellswere 0.078 µmol/L and 0.068 µmol/L, respectively. The cytotoxicity evaluation indicated that TC9, TA11, TM4, and TH3 displayed good safety against HEK293T cells. TH2 and TH3 could induce apoptosis of MCF-7 cells. Molecular modeling and ADMET prediction results indicated that most of target compounds showed promising medicinal properties, which was consistent with the experimental results. Our findings provided new lead compounds for the structural optimization of multi-target angiogenesis inhibitors.

Synthesis, molecular docking and biological evaluation of 1,2,4-oxadiazole based novel non-steroidal derivatives against prostate cancer.

Prostate cancer is one of the most prevalent cancers in men leading to second most death causing cancer in men. Despite the availability of multiple treatment still the prevalence is high for prostate cancer. Steroidal antagonists associated with poor bioavailability, side effects while non-steroidal antagonists show serious side effects like gynecomastia. Therefore, there is a need of potential candidate for the treatment of prostate cancer with better bioavailability, good therapeutic effect and minimal side effects. In the same context, we have designed the series, SP1-SP25 based 3-phenyl-5-styryl-1,2,4-oxadiazole as the core structure. We successfully synthesized all 25 molecules in this series and characterized them using 1H, 13C NMR, and mass spectroscopy. Subsequently, we conducted MTT assays using PC-3 cells and observed that all the compounds exhibited a dose-dependent decrease in cell viability. Notably, compounds SP04, SP16, and SP19 demonstrated a significant decrease in cell viability and exhibited potent activity compared to the other synthesized molecules and standard drug bicalutamide. Among them, SP04 emerged as the one of the most potent compounds with an IC50 value of 238.13 nM and an 89.99 % inhibition of PC-3 cells, compared to synthesized molecules and standard drug bicalutamide. Furthermore, we conducted ROS assays and androgen receptor inhibition assays using the potent compound SP04 and bicalutamide. The results indicated that SP04 increased ROS production and decreased androgen receptor expression dose-dependent manner. Additionally, we conducted a docking study to analyse the interaction patterns within the active site of the androgen receptor. ADMET analysis revealed that all the compounds exhibited favorable physicochemical properties and manageable toxicity profiles.

Pyrano[2,3-b]chromone derivatives as novel dual inhibitors of α-glucosidase and α-amylase: Design, synthesis, biological evaluation, and in silico studies.

Inhibition of α-glucosidase and α-amylase is an important target for treatment of type 2 diabetes. In this work, a novel series of pyrano[2,3-b]chromene derivatives 5a-m was designed based on potent α-glucosidase and α-amylase inhibitors and synthesized by simple chemical reactions. These compounds were evaluated against the latter enzymes. Most of the title compounds exhibited high inhibitory activity against α-glucosidase and α-amylase in comparison to standard inhibitor (acarbose). Representatively, the most potent compound, 4-methoxy derivative 5d, was 30.4 fold more potent than acarbose against α-glucosidase and 6.1 fold more potent than this drug against α-amylase. In silico molecular modeling demonstrated that compound 5d attached to the active sites of α-glucosidase and α-amylase with a favorable binding energies and established interactions with important amino acids. Dynamics of compound 5d also showed that this compound formed a stable complex with the α-glucosidase active site. In silicodrug-likeness as well as ADMET prediction of this compound was also performed and satisfactory results were obtained.

Design, synthesis, and biological evaluation of novel 2,3-Di-O-Aryl/Alkyl sulfonate derivatives of l-ascorbic acid: Efficient access to novel anticancer agents via in vitro screening, tubulin polymerization inhibition, molecular docking study and ADME predictions.

A series of novel l-ascorbic acid derivatives bearing aryl and alkyl sulfonate substituents were synthesized and characterized. In vitro anticancer evaluation against MCF-7 (breast) and A-549 (lung) cancer cell lines revealed potent activity for most of the compounds, with 2b being equipotent to the standard drug colchicine against MCF-7 (IC50 = 0.04 µM). Notably, compound 2b displayed 89-fold selectivity for MCF-7 breast cancer over MCF-10A normal breast cells. Derivatives with two sulfonate groups (2a-g, 3a-g) exhibited superior potency over those with one sulfonate (4a-c,5g, 6b). Compounds 2b and 2c potently inhibited tubulin polymerization in A-549 cancer cells (73.12 % and 62.09 % inhibition, respectively), substantiating their anticancer potential through microtubule disruption. Molecular docking studies showed higher binding scores and affinities for these compounds at the colchicine-binding site of α, ß-tubulin compared to colchicine itself. In-silico ADMET predictions indicated favourable drug-like properties, with 2b exhibiting the highest binding affinity. These sulfonate derivatives of l-ascorbic acid represents promising lead scaffolds for anticancer drug development.

Organohalogen chalcones: design, synthesis, ADMET prediction, molecular dynamics study and inhibition effect on acetylcholinesterase and carbonic anhydrase.

In an effort to discover potential acetylcholinesterase (AChE) and carbonic anhydrase (CA) inhibitors, a novel series of organohalogen chalcone derivatives (12-20, 23-30) was synthesized, and their chemical structures were characterized by spectral analysis. They showed a highly potent inhibition effect on AChE and hCAs (Ki values range from 5.07 ± 0.062 to 65.53 ± 4.36 nM for AChE, 13.54 ± 2.55 to 94.11 ± 10.39 nM for hCA I, and 5.21 ± 0.54 to 57.44 ± 3.12 nM for hCA II). In addition, the chalcone derivatives with the highest inhibitor score docked into the active site of the indicated metabolic enzyme receptors, and their absorption, metabolism, and toxic properties were evaluated according to ADMET's estimation.Compounds 16 and 19 exhibited the highest inhibition score, emerged as lead compounds, and inspired the development of more potent compounds.

A green one-pot synthetic protocol of hexahydropyrimido[4,5-d]pyrimidin-4(1H)-one derivatives: molecular docking, ADMET, anticancer and antimicrobial studies.

Ten hexahydropyrimido[4,5-d]pyrimidine derivatives have been synthesized by using a green and time-efficient microwave method. The synthesized motifs were evaluated for their anticancer activity, antimicrobial activity, molecular docking, drug likeliness and ADMET studies. Comparatively, the hetero-aromatic pyrazole substituted compound 4a exhibited the highest anticancer activity [Mean growth percent: 35.57], while EDG [-N(CH3)2] substituted compound 4i indicated very good activity [Mean growth percent: 60.92] against various cell lines. From the computational studies, Compound 4a passed the drug-likeness and ADME properties, fewer toxic properties, and potent inhibitory potential against the RIPK2 with significant binding affinity. In-silico molecular docking revealed that the compound 4a has significant binding energy (- 9.8 kcal/mol) and dissociation constant (0.54 µM) properties. Additionally, synthesized motifs were evaluated for antimicrobial activity by MIC referencing the standards. According to the SAR evaluations, the compounds 4f (4-NO2), 4g (3-NO2), and 4h (2-Cl) that include EWGs substituted aldehydes performed well as antimicrobials against selected bacterial and fungal strains. Thus, the synthesized pyrimido[4,5-d]pyrimidine with the heterocyclic and EWGs substituents could act as a potential candidate after further structural optimization for anticancer and antimicrobial drug discovery, respectively.

Construction of IRAK4 inhibitor activity prediction model based on machine learning.

Interleukin-1 receptor-associated kinase 4 (IRAK4) is a crucial serine/threonine protein kinase that belongs to the IRAK family and plays a pivotal role in Toll-like receptor (TLR) and Interleukin-1 receptor (IL-1R) signaling pathways. Due to IRAK4's significant role in immunity, inflammation, and malignancies, it has become an intriguing target for discovering and developing potent small-molecule inhibitors. Consequently, there is a pressing need for rapid and accurate prediction of IRAK4 inhibitor activity. Leveraging a comprehensive dataset encompassing activity data for 1628 IRAK4 inhibitors, we constructed a prediction model using the LightGBM algorithm and molecular fingerprints. This model achieved an R2 of 0.829, an MAE of 0.317, and an RMSE of 0.460 in independent testing. To further validate the model's generalization ability, we tested it on 90 IRAK4 inhibitors collected in 2023. Subsequently, we applied the model to predict the activity of 13,268 compounds with docking scores less than - 9.503 kcal/mol. These compounds were initially screened from a pool of 1.6 million molecules in the chemdiv database through high-throughput molecular docking. Among these, 259 compounds with predicted pIC50 values greater than or equal to 8.00 were identified. We then performed ADMET predictions on these selected compounds. Finally, through a rigorous screening process, we identified 34 compounds that adhere to the four complementary drug-likeness rules, making them promising candidates for further investigation. Additionally, molecular dynamics simulations confirmed the stable binding of the screened compounds to the IRAK4 protein. Overall, this work presents a machine learning model for accurate prediction of IRAK4 inhibitor activity and offers new insights for subsequent structure-guided design of novel IRAK4 inhibitors.

An in silico approach for identification of lead compound as FtsZ inhibitor.

The remarkable conservation of the FtsZ among Gram-positive and Gram-negative bacteria, a crucial GTPase in bacterial cell division, has emerged as a promising antibacterial drug target to combat antibacterial resistance. There have been several coordinated efforts to develop inhibitors against FtsZ which can also serve as potential candidates for future antibiotics. In the present study, a natural product-like library (≈50,000 compounds) was employed to conduct HTVS against Staphylococcus aureus FtsZ protein (PDB Id: 6KVP). Additionally, molecular docking was carried out in two modes, SP and XP docking, using the Schrödinger suite. The glide scores of ligands obtained by XP docking were further summarized and compared with the control ligands (ZI1- co-crystal and PC190723-a compound undergoing clinical trial). Using the Prime-MM-GBSA approach, BFE calculations were performed on the top XP-scored ligands (≈598 compounds). These hits were also evaluated for ADMET parameters using the Qikprop algorithm, SwissADME, and in silico carcinogenicity testing using Carcinopred-El. Based on the results, ligand 4-FtsZ complex was considered for the 300 ns MDS analysis to get insights into its binding modes within the catalytic pocket of FtsZ protein. The analysis revealed that the amide linkage sandwiched between the triazole and 1-oxa-8-azaspirodecan-8-ium moiety (Val203) as well as the aminoethyl group present at 1st position on the triazole moiety (Leu209, Leu200, Asp210, and Ala202) were responsible for the FtsZ inhibitory activity, owing to their crucial interactions with key amino acid residues. Further, the complex also displayed good protein-ligand stability, ultimately predicting ligand 4 as a potent lead compound for the inhibition of FtsZ. Thus, our in silico findings will serve as a framework for in-depth in-vitro and in-vivo investigations encouraging the development of FtsZ inhibitors as a new generation of antibacterial agents.