Tuberculosis
Our largest body of work, across the mycobacterial cell wall, fatty acid synthesis, dormancy and the toxin antitoxin system.
Research
Methods proven
in public.
How we validate
How we check
our own results.
0.840 ÅSilicoX plore platform paper
Redock RMSD
The crystallographic ligand of NDM-1 was docked back into its own structure, and the pose it returned sat within 0.840 Å of the crystal pose.
4,247SilicoX plore platform paper
Property matched decoys
Docking was also tested retrospectively, on how well it separates known actives from 4,247 decoys matched to them on molecular properties.
IdenticalOmicsX plore manuscript
Independent re-run
Every job
Full parameter trail
Each job stores its parameters, engine versions, logs and outputs, so a result can be traced back to how it was produced.
Both platformsResearch we run on
our own platform.
Case study · NDM-1From 455 known actives to four nominated moleculesNDM-1 hydrolyses almost the whole β-lactam class and no inhibitor has reached the clinic.
455curated seed actives
5successive filters
4candidates nominated
Flagship · the platform paper
Read the paper
SilicoX plore: An integrated cloud platform coupling machine learning with physics-based modelling for end-to-end drug discovery, applied to the computational prioritisation of putative New Delhi metallo-β-lactamase-1 binders
Journal of Molecular Graphics and Modelling, 149 (2026) 109566
Parth Mangal, Omkar Shinde, Sonali Chikhale, Shovonlal Bhowmick, Md Ataul Islam
In plain termsA cloud platform that takes a drug target from a set of known actives to a short list of candidate molecules in one session, shown here on an enzyme that makes many antibiotics ineffective.
The paper that introduces the platform. Applied to NDM-1, a generative run produced 752,436 molecules from 455 seed actives and the cascade narrowed them to four.
752,436molecules generated
4nominated
31platform modules
Case study · EGFRFrom 36,000 natural products to eight leadsEach stage ran in a named SilicoX plore module, with no file exported between them.
- NPAtlas natural products screened36,000SilicoSAR
- Predicted active, pIC50 above 616,000SilicoSAR
- Passed ADMET filtering2,035PharmK-AI
- Docked at 7.5 kcal/mol or better280LigandLock
- Taken into molecular dynamics20SilicoMD
- Stable over 200 ns of dynamics8SilicoGBSA, SilicoFEP
0.92R² on training data
0.61R² on the external set
8leads carried to free energy
Module paper · SilicoSAR
SilicoSAR: an integrated machine-learning QSAR module within the SilicoX plore platform for QSAR-based virtual screening, demonstrated by the discovery of natural-product EGFR inhibitors
Computational Biology and Chemistry, volume 126, article 109467
Md Ataul Islam, Shovonlal Bhowmick, Sonali Chikhale
In plain terms
SilicoSAR takes a team from a dataset of known actives to a validated prediction model and a screened library in one graphical workflow, with no code and no manual hand-off to the docking and simulation modules that follow. Applied to the cancer target EGFR, it screened 36,000 natural products and the cascade narrowed them to eight leads, two of which were the most promising by free-energy analysis.
MenT3TB
Identification of potential MenT3 inhibitors for Mycobacterium tuberculosis using the generative artificial intelligence and SilicoX plore platform
In plain termsDesigned and screened new candidate molecules, on a computer, against a toxin made by the bacterium that causes tuberculosis.
100,000145
Scientific ReportsOpen access
KasATB
Free energy perturbation and machine learning-assisted identification of novel molecules for the Mycobacterium tuberculosis KasA protein: a fragment-based drug design approach
In plain termsBuilt candidate molecules from small fragments against KasA, an enzyme the tuberculosis bacterium uses to build its cell wall.
5-9.0FEP
Molecular Diversity
VEGFR-2Oncology
Applying artificial intelligence and machine learning framework for de novo design of pyrazole-based VEGFR-2 inhibitors
In plain termsDesigned new molecules from scratch against VEGFR-2, a protein that helps tumours grow the blood vessels they need.
de novoMD
Journal of Molecular Graphics and Modelling
Chk1Oncology
Free energy perturbation-assisted identification of checkpoint kinase 1 targeted small-molecule inhibitors for cancer therapy using AI- and physics-driven molecular design
In plain termsLooked for molecules that block Chk1, a protein that cancer cells rely on to repair damage to their DNA.
FEPML
Computational Biology and Chemistry
GPI anchored wall transfer protein 1Antifungal
Identification of GPI-Anchored Wall Transfer Protein 1 Modulators for Fungal Infections Through Generative AI and Physics-Based Approaches
In plain termsDesigned candidate molecules against a protein that fungi need to build their cell wall.
4fungal
International Journal of Molecular SciencesOpen access
CYP51BAntifungal
Artificial Intelligence-Guided Design and Computational Evaluation of Candidate CYP51B Inhibitors as Potential Antifungal Agents
In plain termsDesigned candidate molecules against CYP51B, an enzyme fungi use to make their membrane sterols and a proven antifungal target.
AI guidedantifungal
ChemistrySelect
Evidence for
OmicsX plore.
Manuscript · under review
Idiopathic pulmonary fibrosis: 39 lung transcriptomes, from reads to enrichment in one guided run
In plain termsThe software took raw sequencing data from 39 lung samples through to an interpretable result in a single guided run, and a second run reproduced the same calls.
Transcriptomes from patients with idiopathic pulmonary fibrosis and control donors were taken from raw reads to functional enrichment. Matrix metalloproteinases, fibrillar collagens and basal keratins led the down regulated set, and extracellular matrix organisation dominated both over representation and preranked gene set enrichment. The optional splicing and fusion modules ran in the same session. A second independent run from the exported counts reproduced identical significance calls.
Battula M, Mangal P, Shinde O, Chikhale S, Bhowmick S, Islam MA. Manuscript under review.
All findings are computational.
39Lung transcriptomes, patients and control donors
89.6%Read pairs kept after trimming
98.2%Retained pairs aligned to hg38
3,464Features differentially expressed, of 10,484 tested
1,714 / 1,750Up and down regulated genes
38,909Differential splice junctions
271Candidate fusion events
IdenticalSignificance calls on an independent re-run
Research by area
Where the work is pointed.
What the work in each of the six therapeutic areas is aimed at.
Oncology
Kinases and regulators central to proliferation and the DNA damage response, from VEGFR-2 and the Aurora and checkpoint kinases to protein protein complexes.
Antifungal
Cell wall biosynthesis and sterol pathway targets, addressing the rising clinical burden of resistant fungal infection.
Neurological
Enzyme targets in Alzheimer's disease and in neurodegenerative and psychiatric disorders.
Antiviral
Structured viral RNA targets, including the hepatitis C internal ribosome entry site.
Antimicrobial resistance
Enzymes that defeat front line antibiotics, including the metallo beta lactamase that resists the carbapenems.
Publications
The complete record.
Every peer reviewed paper from the group, newest first, with the authors and the journal as published. Open any row with the plus sign for a summary and the first page of the article. Filter by area, by technology or by open access. Names that are underlined belong to members of our team.
01Antimicrobial resistanceNDM-1SilicoX plore: An integrated cloud platform coupling machine learning with physics-based modelling for end-to-end drug discovery, applied to the computational prioritisation of putative New Delhi metallo-β-lactamase-1 binders Parth Mangal, Omkar Shinde, Sonali Chikhale, Shovonlal Bhowmick, Md Ataul IslamJournal of Molecular Graphics and Modelling2026SilicoX plore Show summary
First pageSummary
- Journal
- Journal of Molecular Graphics and Modelling
- Published
- 2026 · Volume 149 · 109566
- Therapeutic area
- Antimicrobial resistance
- Target
- NDM-1
- Technology
Run on Silico X plore- Access
- Subscription journal
02OncologyEGFRSilicoSAR: an integrated machine-learning QSAR module within the SilicoX plore platform for QSAR-based virtual screening, demonstrated by the discovery of natural-product EGFR inhibitors Md Ataul Islam, Shovonlal Bhowmick, Sonali ChikhaleComputational Biology and Chemistry2026SilicoX plore Show summary
First pageSummary
- Journal
- Computational Biology and Chemistry
- Published
- 2026 · Volume 126 · 109467
- Therapeutic area
- Oncology
- Target
- EGFR
- Technology
Run on Silico X plore- Access
- Subscription journal
03AntifungalGPI anchored wall transfer protein 1Identification of GPI-Anchored Wall Transfer Protein 1 Modulators for Fungal Infections Through Generative AI and Physics-Based ApproachesIbrahim A. Alsarra, Rupesh Chikhale, Abdullah M. Al-Mohizea, Md Ataul IslamInternational Journal of Molecular Sciences2026SilicoX plore Open accessShow summary
First pageSummary
Screens more than 60,000 AI-generated molecules (REINVENT4) against GWT1, a fungal-specific enzyme in GPI-anchor biosynthesis, through ADMET filtering, docking, synthesisability and pharmacophore selection, molecular dynamics and density functional theory. Four candidates are prioritised as potential leads for invasive fungal infections, pending validation.
- Journal
- International Journal of Molecular Sciences
- Published
- 2026 · Volume 27 · Issue 11 · 4767
- Therapeutic area
- Antifungal
- Target
- GPI anchored wall transfer protein 1
- Technology
Run on Silico X plore- Access
- Open access
04AntifungalCYP51BArtificial Intelligence-Guided Design and Computational Evaluation of Candidate CYP51B Inhibitors as Potential Antifungal AgentsMd Ataul Islam, Mohammad Ajmal Ali, Rupesh Chikhale, Md Lutful Islam, Saikh Mohammad WabaidurChemistrySelect2026SilicoX plore Show summary
First pageSummary
Expands a library of 174 antifungal agents to more than 111,000 structures with REINVENT4, then filters by diversity, ADMET and docking against Aspergillus fumigatus CYP51B. Five candidates were simulated in duplicate for 100 ns, and one, AFR2, is proposed for experimental follow-up. All findings are computational predictions requiring experimental validation.
- Journal
- ChemistrySelect
- Published
- 2026 · Volume 11 · Issue 34 · e74407
- Therapeutic area
- Antifungal
- Target
- CYP51B
- Technology
Run on Silico X plore- Access
- Subscription journal
05AntiviralHCV IRESIntegrated machine-learning methods for identification of Hepatitis C viral IRES sub-domain-IIa inhibitorsOmkar Shinde, Pritee Chunarkar Patil, Gaber E. Eldesoky, Rupesh V. ChikhaleCanadian Journal of Chemistry2026Show summary
First pageSummary
Generates new molecules from FDA-approved drug libraries with REINVENT4 and refines them by similarity, docking, pharmacokinetic and toxicity assessment. Four compounds bind sub-domain IIa of the hepatitis C virus IRES, an RNA target, and 200 ns molecular dynamics examines the stability of each complex.
- Journal
- Canadian Journal of Chemistry
- Published
- 2026 · Volume 104 · Issue 3 · 289-302
- Therapeutic area
- Antiviral
- Target
- HCV IRES
- Access
- Subscription journal
06NeurologicalPDE10AIdentification of phosphodiesterase 10 A modulators for neurodegenerative and psychiatric disorders: Combination of physics-based virtual screening and machine learning approachesVipra Ajay Parekh, Musarat Amina, Md Lutful Islam, Pritee Chunarkar Patil, Mohammad Ajmal Ali, Saikh Mohammad Wabaidur, Md Ataul IslamComputational Biology and Chemistry2026Show summary
First pageSummary
Uses a ligand-guided virtual screen of the ZINCPharmer database, docking with two engines and machine-learning binding-affinity prediction to identify four candidate inhibitors of PDE10A, a target for neurodegenerative and psychiatric disorders. 100 ns molecular dynamics and MM-GBSA support the stability and binding of the complexes.
- Journal
- Computational Biology and Chemistry
- Published
- 2026 · Volume 121 · 108875
- Therapeutic area
- Neurological
- Target
- PDE10A
- Access
- Subscription journal
07OncologyVEGFR-2Applying artificial intelligence and machine learning framework for de novo design of pyrazole-based VEGFR-2 inhibitorsDeepali M. Wanode, Md Ataul Islam, Kumud P. Bhendarkar, Megha P. Ambatkar, Pramod B. Khedekar, Rupesh ChikhaleJournal of Molecular Graphics and Modelling2026SilicoX plore Show summary
First pageSummary
Applies REINVENT4 with ADMET screening, docking, molecular dynamics and MM-GBSA to design pyrazole-based inhibitors of VEGFR-2, a kinase central to tumour angiogenesis. Five lead inhibitors are identified; their biological efficacy has yet to be assessed.
- Journal
- Journal of Molecular Graphics and Modelling
- Published
- 2026 · Volume 146 · 109458
- Therapeutic area
- Oncology
- Target
- VEGFR-2
- Technology
Run on Silico X plore- Access
- Subscription journal
08OncologyChk1Free energy perturbation-assisted identification of checkpoint kinase 1 targeted small-molecule inhibitors for cancer therapy using AI- and physics-driven molecular designMd Ataul Islam, Mohammad Ajmal Ali, Rupesh Chikhale, Md Lutful Islam, Saikh Mohammad WabaidurComputational Biology and Chemistry2026SilicoX plore Show summary
First pageSummary
Combines REINVENT4 and DeLA-Drug to design candidate inhibitors of checkpoint kinase 1, a regulator of the response to DNA damage. Three candidates are assessed by molecular dynamics and MM-GBSA, and the best also by absolute free energy perturbation, providing a basis for further optimisation as anticancer agents.
- Journal
- Computational Biology and Chemistry
- Published
- 2026 · Volume 124 · 109203
- Therapeutic area
- Oncology
- Target
- Chk1
- Technology
Run on Silico X plore- Access
- Subscription journal
09OncologyERK2Identification of Novel Extracellular-Signal-Regulated Kinase 2 Inhibitors Through Machine Learning-Driven De Novo Design, Molecular Docking, and Free-Energy PerturbationIbrahim A. Alsarra, Mahima Sudhir Kolpe, Md Ataul IslamPharmaceuticals2026Open accessShow summary
First pageSummary
Generates molecules with the machine-learning tool DeLA-Drug, narrows them by similarity search, docking and pharmacokinetic assessment, and refines poses and stability with DiffDock and molecular dynamics. Four candidates are identified, and free energy perturbation supports the affinity of the lead, Ek1, for ERK2, subject to experimental validation.
- Journal
- Pharmaceuticals
- Published
- 2026 · Volume 19 · Issue 2 · 337
- Therapeutic area
- Oncology
- Target
- ERK2
- Access
- Open access
10OncologyCDC34-UBCIdentification of potential allosteric inhibitors-modulators for the heterodimer CDC34-UBC protein-protein complexHemchandra Deka, Heba Taha M. Abdelghani, Aditi Gangopadhyay, Atul Darasing Pawar, Pritee Chunarkar Patil, Shovonlal Bhowmick3 Biotech2026Show summary
First pageSummary
Screens 26,318 allosteric kinase inhibitors from ChemDiv by similarity and docking against the allosteric site of the CDC34-UBC protein-protein complex. Four drug-like compounds are identified, and 100 ns molecular dynamics shows the protein backbone remaining stable with each of them.
- Journal
- 3 Biotech
- Published
- 2026 · Volume 16 · Issue 5 · 160
- Therapeutic area
- Oncology
- Target
- CDC34-UBC
- Access
- Subscription journal
11OncologyMAP3K8Free energy perturbation and machine learning-assisted identification of potential MAP3K8 hit molecules: a comprehensive structure- and ligand-based studiesMd Ataul Islam, Aaherasha Iqbal, Mohammad Ajmal Ali, Rupesh Chikhale, Md Lutful Islam, Mohammad Abul FarahMolecular Diversity2026Show summary
First pageSummary
Integrates ligand- and structure-based screening, starting from a pharmacophore derived from the co-crystal ligand and 20,566 unique molecules, to reach four structurally diverse candidates for MAP3K8, a kinase in inflammatory signalling and cancer. 200 ns molecular dynamics and MM-GBSA rank three of them above the co-crystal ligand.
- Journal
- Molecular Diversity
- Published
- 2026
- Therapeutic area
- Oncology
- Target
- MAP3K8
- Access
- Subscription journal
12TuberculosisMenT3Identification of potential MenT3 inhibitors for Mycobacterium tuberculosis using the generative artificial intelligence and SilicoX plore platform Ibrahim A. Alsarra, Vikramsinh Sardarsinh Suryawanshi, Abdullah M. Al-Mohizea, Pritee Chunarkar Patil, Rupesh Chikhale, Md Ataul IslamScientific Reports2026SilicoX plore Open accessShow summary
First pageSummary
- Journal
- Scientific Reports
- Published
- 2026 · Volume 16 · Issue 1 · 18185
- Therapeutic area
- Tuberculosis
- Target
- MenT3
- Technology
Run on Silico X plore- Access
- Open access
13TuberculosisKasAFree energy perturbation and machine learning-assisted identification of novel molecules for the Mycobacterium tuberculosis KasA protein: a fragment-based drug design approachMd Ataul Islam, Mohammad Ajmal Ali, Rupesh Chikhale, Md Lutful Islam, Mohammad Abul FarahMolecular Diversity2026SilicoX plore Show summary
First pageSummary
Uses fragment-based design: fragments from known KasA inhibitors are recombined into hit compounds and evaluated by multi-tier docking, pharmacokinetics and molecular dynamics. Five candidate molecules score better than the reference ligand TLM in docking, and MM-GBSA and free energy perturbation support their affinity, warranting experimental validation.
- Journal
- Molecular Diversity
- Published
- 2026
- Therapeutic area
- Tuberculosis
- Target
- KasA
- Technology
Run on Silico X plore- Access
- Subscription journal
14TuberculosisTMKFree energy perturbation-derived identification of natural compounds targeting Mycobacterium tuberculosis thymidylate kinaseRupesh V. Chikhale, Md Ataul Islam, Vikramsinh Sardarsinh Suryawanshi, Ibrahim A. AlsarraScientific Reports2026Open accessShow summary
First pageSummary
Searches natural-product libraries for inhibitors of Mycobacterium tuberculosis thymidylate kinase using fingerprint search, docking, ADMET prediction and 100 ns molecular dynamics. MM-GBSA and free energy perturbation are used to rank the molecules, and three natural compounds show docking energies better than the co-crystal ligand.
- Journal
- Scientific Reports
- Published
- 2026 · Volume 16 · Issue 1 · 16272
- Therapeutic area
- Tuberculosis
- Target
- TMK
- Access
- Open access
15TuberculosisDNA gyrase BScreening and optimization of potential molecules targeting DNA gyrase B in Mycobacterium tuberculosis: a generative artificial intelligence, free energy perturbation, and de novo design approachMd Ataul Islam, Md Lutful Islam, Aaherasha Iqbal, Mohammad Ajmal Ali, Rupesh Chikhale, Mohammad Abul FarahChemical Papers2026Show summary
First pageSummary
Generates 112,814 de novo molecules from 107 known DNA gyrase B inhibitors, filters them by consensus docking scores and optimises two candidates into analogues. Four candidates are then examined by molecular dynamics, MM-GBSA and free energy perturbation; RI_1 emerges as the most consistent across methods and is proposed for experimental validation.
- Journal
- Chemical Papers
- Published
- 2026
- Therapeutic area
- Tuberculosis
- Target
- DNA gyrase B
- Access
- Subscription journal
16NeurologicalAChEIdentification of Acetylcholinesterase Inhibitors Through a Pharmacophore-Guided Deep Learning Approach for Therapeutic Applications in Alzheimer's DiseaseVikramsinh Sardarsinh Suryawanshi, Md Lutful Islam, Gaber E. Eldesoky, Pritee Chunarkar Patil, Shovonlal Bhowmick, Md Ataul IslamChemistrySelect2025Show summary
First pageSummary
Generates new chemical entities from the pharmacophoric features of known drugs and evaluates them by docking and pharmacokinetic analysis for Alzheimer's disease. Four candidate AChE inhibitors bind more favourably than the co-crystal ligand and donepezil in docking, with molecular dynamics indicating stability at the active site.
- Journal
- ChemistrySelect
- Published
- 2025 · Volume 10 · Issue 25 · e00492
- Therapeutic area
- Neurological
- Target
- AChE
- Access
- Subscription journal
17TuberculosisPks13Pharmacophore guided deep learning approach to identify novel inhibitors targeting mycobacterial polyketide synthase Pks13-TE domainRinku Choudhary, Shovonlal Bhowmick, Heba Taha M. Abdelghani, Pritee Chunarkar Patil, Rupesh V. ChikhaleJournal of Molecular Structure2025Show summary
First pageSummary
Uses the Pharmacophore-Guided Molecular Generation tool to design inhibitors of the Pks13 thioesterase domain, essential for mycolic acid biosynthesis in Mycobacterium tuberculosis. Candidates were filtered by ADMET-AI, docking, absolute binding energy and toxicity assessment and refined by MM-GBSA and trajectory analyses, yielding five hit molecules.
- Journal
- Journal of Molecular Structure
- Published
- 2025 · Volume 1319 · 139360
- Therapeutic area
- Tuberculosis
- Target
- Pks13
- Access
- Subscription journal
18TuberculosisEthRGenerative AI, molecular docking and molecular dynamics simulations assisted identification of novel transcriptional repressor EthR inhibitors to target Mycobacterium tuberculosisRupesh V. Chikhale, Rinku Choudhary, Gaber E. Eldesoky, Mahima Sudhir Kolpe, Omkar Shinde, Dilnawaz HossainHeliyon2025Open accessShow summary
First pageSummary
Uses REINVENT4 with docking and molecular dynamics to design new ligands of the EthR repressor as ethionamide boosters in Mycobacterium tuberculosis. Six promising compounds with improved physicochemical properties are identified as candidates for further development against drug-resistant tuberculosis.
- Journal
- Heliyon
- Published
- 2025 · Volume 11 · Issue 4 · e42593
- Therapeutic area
- Tuberculosis
- Target
- EthR
- Access
- Open access
19TuberculosisDosRTargeting Mycobacterial Dormancy Survival Regulator (DosR) With Generative Artificial Intelligence and Omics MethodsMonishka Battula, Samiksha Bhor, Shovonlal Bhowmick, Gaber E. Eldesoky, Rupesh V. ChikhaleChemistry & Biodiversity2025Show summary
First pageSummary
Combines sequence analysis of DosR across mycobacterial species with docking of FDA-approved compounds, de novo generation from the top 100 using REINVENT4, ADMET analysis, molecular dynamics and MM-GBSA. Five compounds are selected as stable binders of DosR, the regulator of mycobacterial dormancy.
- Journal
- Chemistry & Biodiversity
- Published
- 2025 · e02926
- Therapeutic area
- Tuberculosis
- Target
- DosR
- Access
- Subscription journal
20Antimicrobial resistanceRibosomal RNAInvestigating the Ribosomal-RNA: Protein Interactions and AI-Assisted Discovery of Novel InhibitorsMonishka Battula, Shovonlal Bhowmick, Pritee Chunarkar Patil, Gaber E. Eldesoky, Rupesh V. ChikhaleChemistrySelect2024Show summary
First pageSummary
Studies the interaction between ribosomal RNA and protein and uses rDock and REINVENT4 to generate and evaluate new ligands. Five candidate molecules show stronger binding and stability than the peptide in the reference complex, with favourable pharmacokinetic and safety profiles.
- Journal
- ChemistrySelect
- Published
- 2024 · Volume 9 · Issue 43 · e202403459
- Therapeutic area
- Antimicrobial resistance
- Target
- Ribosomal RNA
- Access
- Subscription journal
21OncologyPeregrinIdentification of Peregrin inhibitors-modulators by harnessing the computational prowess of molecular simulation and machine learning algorithmsHemchandra Deka, Atul Darasing Pawar, Monishka Srinivas Battula, Gaber E. Eldesoky, Omkar Dilip Shinde, Pritee Chunarkar Patil, Tapan Kumar Mistri, Hitesh Vishwanath Shahare, Shovonlal BhowmickJournal of Molecular Liquids2024Show summary
First pageSummary
Screens 5,430 bromodomain-directed compounds from ChemDiv by docking with two scoring functions and machine-learning binding-affinity prediction, followed by pharmacokinetics and 100 ns molecular dynamics. Four compounds hold their interaction with Peregrin over the simulation, with K788-9421 the most favourable; extensive experimental validation is needed.
- Journal
- Journal of Molecular Liquids
- Published
- 2024 · Volume 411 · 125782
- Therapeutic area
- Oncology
- Target
- Peregrin
- Access
- Subscription journal
22OncologyHsp90Machine Learning Aided De Novo Design Identifies Novel Benzimidazolone Based Inhibitor-Modulators for Heat Shock Protein 90 (HSP90)Shovonlal Bhowmick, Jagriti Malhotra, Rinku Choudhary, Omkar Dilip Shinde, Ahmed Sayed M. Metwally, Pritee Chunarkar PatilChemistrySelect2024Show summary
First pageSummary
Uses REINVENT4 to generate molecules against known HSP90 inhibitors, then reduces the chemical space by pharmacokinetics, docking and machine-learning absolute binding affinity. Three benzimidazole-based candidates, IM1 to IM3, are examined in 100 ns molecular dynamics and MM-GBSA.
- Journal
- ChemistrySelect
- Published
- 2024 · Volume 9 · Issue 31 · e202402147
- Therapeutic area
- Oncology
- Target
- Hsp90
- Access
- Subscription journal
23OncologyERK2Machine learning-integrated and fingerprint-based similarity search against immuno oncology library for identification of novel ERK2 inhibitorsVikramsinh Sardarsinh Suryawanshi, Surbhi Pravin Pawar, Mahima Sudhir Kolpe, Heba Taha M. Abdelghani, Sonali Chikhale, Pritee Chunarkar Patil, Shovonlal BhowmickStructural Chemistry2024Show summary
First pageSummary
Combines a similarity search of an immuno-oncology library, based on known kinase inhibitors, with docking and AI/ML pharmacokinetic filtering to select ERK2 leads. 300 ns molecular dynamics, with RMSD, radius of gyration, fluctuation, hydrogen-bond and MM-GBSA analyses, confirms the association of the selected complexes.
- Journal
- Structural Chemistry
- Published
- 2024 · Volume 36 · Issue 2 · 681-700
- Therapeutic area
- Oncology
- Target
- ERK2
- Access
- Subscription journal
24OncologyCHK2Identification of Potent CHK2 Inhibitors-Modulators for Therapeutic Application in Cancer: A Machine Learning Integrated Fragment-Based Drug Design ApproachMahima Sudhir Kolpe, Vikramsinh Sardarsinh Suryawanshi, Gaber E. Eldesoky, Dilnawaz Hossain, Pritee Chunarkar Patil, Shovonlal BhowmickChemistrySelect2024Show summary
First pageSummary
Applies fragment-based design: fragments from known CHK2 inhibitor scaffolds, generated with MacFrag and screened by similarity, are assembled with Fragmenstein into new molecules, then assessed by docking, pharmacokinetics and molecular dynamics. Five novel compounds are proposed as CHK2 inhibitors.
- Journal
- ChemistrySelect
- Published
- 2024 · Volume 9 · Issue 39 · e202403302
- Therapeutic area
- Oncology
- Target
- CHK2
- Access
- Subscription journal
25OncologyAurora A kinaseIdentification of Aurora A kinase allosteric inhibitors: A comprehensive virtual screening through fingerprint-based similarity search, molecular docking, machine learning and molecular dynamics simulationMahima Sudhir Kolpe, Surbhi Pravin Pawar, Vikramsinh Sardarsinh Suryawanshi, Heba Taha M. Abdelghani, Pritee Chunarkar Patil, Shovonlal BhowmickJournal of Molecular Liquids2024Show summary
First pageSummary
Targets the allosteric site of Aurora A kinase through site recognition, fingerprint similarity search, multi-step docking with AutoDock Vina and PLANTS, and 100 ns molecular dynamics with MM-GBSA analysis, to identify allosteric inhibitors and modulators.
- Journal
- Journal of Molecular Liquids
- Published
- 2024 · Volume 414 · 126115
- Therapeutic area
- Oncology
- Target
- Aurora A kinase
- Access
- Subscription journal
26OncologyAurora A kinaseMolecular simulations and machine learning methods for the identification of novel aurora A kinase inhibitorsSurbhi Pravin Pawar, Mahima Sudhir Kolpe, Vikramsinh Sardarsinh Suryawanshi, Sonali Chikhale, Ammar M. Tighezza, Pritee Chunarkar Patil, Shovonlal BhowmickJournal of Biomolecular Structure and Dynamics2024Show summary
First pageSummary
Screens a ChemDiv kinase-inhibitor library against Aurora A kinase, a regulator of mitosis implicated in cancer, using multi-step docking with AutoDock Vina and PLANTS, absolute binding energy estimation and molecular dynamics. Three hit compounds bind the active site in a mode similar to the reference ligand VX-680.
- Journal
- Journal of Biomolecular Structure and Dynamics
- Published
- 2024 · Volume 43 · Issue 14 · 7797-7810
- Therapeutic area
- Oncology
- Target
- Aurora A kinase
- Access
- Subscription journal
27TuberculosisTMKIdentification of mycobacterial Thymidylate kinase inhibitors: a comprehensive pharmacophore, machine learning, molecular docking, and molecular dynamics simulation studiesRupesh V. Chikhale, Surbhi Pravin Pawar, Mahima Sudhir Kolpe, Omkar Dilip Shinde, Kholood A. Dahlous, Saikh Mohammad, Pritee Chunarkar Patil, Shovonlal BhowmickMolecular Diversity2024Open accessShow summary
First pageSummary
Builds a statistically validated pharmacophore model from known mycobacterial thymidylate kinase inhibitors and screens the Enamine database, reducing the chemical space by docking, pharmacokinetics and machine-learning binding energy. Molecular dynamics and MM-GBSA support strong, stable binding, subject to in vitro and in vivo validation.
- Journal
- Molecular Diversity
- Published
- 2024 · Volume 28 · Issue 4 · 1947-1964
- Therapeutic area
- Tuberculosis
- Target
- TMK
- Access
- Open access
28TuberculosisPknBIdentification and Design of Novel Potential Antimicrobial Peptides Targeting Mycobacterial Protein Kinase PknBHemchandra Deka, Atul Pawar, Monishka Battula, Ayman A. Ghfar, Mohamed E. Assal, Rupesh V. ChikhaleThe Protein Journal2024Open accessShow summary
First pageSummary
Docks a library of 5,626 antimicrobial peptides to the PknB kinase domain, selects binders by length and affinity, and examines five in molecular dynamics. A 15-residue peptide binds best, and residue substitution and elastic network analysis are used to refine it; in vitro analysis is needed before any clinical application.
- Journal
- The Protein Journal
- Published
- 2024 · Volume 43 · Issue 4 · 858-868
- Therapeutic area
- Tuberculosis
- Target
- PknB
- Access
- Open access
29TuberculosisPks13Identification of novel hit molecules targeting M. tuberculosis polyketide synthase 13 by combining generative AI and physics-based methodsRupesh V. Chikhale, Rinku Choudhary, Jagriti Malhotra, Gaber E. Eldesoky, Parth Mangal, Pritee Chunarkar PatilComputers in Biology and Medicine2024Show summary
First pageSummary
Combines AI-based tools (REINVENT4, pkCSM, KDeep, SwissADME) with physics-based methods (docking, molecular dynamics, MM-GBSA) to design inhibitors of the Pks13 thioesterase domain from known inhibitors. Six ligands, Mt1 to Mt6, show favourable predicted pharmacokinetic and toxicity profiles and good synthesisability.
- Journal
- Computers in Biology and Medicine
- Published
- 2024 · Volume 176 · 108573
- Therapeutic area
- Tuberculosis
- Target
- Pks13
- Access
- Subscription journal
30TuberculosisMmpL3Gen-AI Methods, Molecular Docking and Molecular Dynamics Simulations for Identification of Novel Inhibitors of MmPL3 Transporter of Mycobacterium tuberculosisAtul Pawar, Tahani Mazyad Almutairi, Omkar Shinde, Rupesh ChikhaleJournal of Computational Biophysics and Chemistry2024Show summary
First pageSummary
Starts from 85 known MmpL3 compounds and uses DeLA-Drug to generate about 15,000 molecules, which are docked at the SQ109 binding site and filtered by pharmacokinetics and toxicity. Five ligands show stable binding and drug-like properties in 100 ns molecular dynamics.
- Journal
- Journal of Computational Biophysics and Chemistry
- Published
- 2024 · Volume 24 · Issue 04 · 471-489
- Therapeutic area
- Tuberculosis
- Target
- MmpL3
- Access
- Subscription journal
31TuberculosisInhAMachine learning assisted methods for the identification of low toxicity inhibitors of Enoyl-Acyl Carrier Protein Reductase (InhA)Rupesh V. Chikhale, Heba Taha M. Abdelghani, Hemchandra Deka, Atul Darasing Pawar, Pritee Chunarkar Patil, Shovonlal BhowmickComputational Biology and Chemistry2024Show summary
First pageSummary
Screens about 119,000 molecules from the SelleckChem database for InhA inhibitors using drug-likeness, docking against a standard-drug threshold, and pharmacokinetic and toxicity filters. Five molecules are assessed as non-toxic and readily synthesisable by machine learning, and molecular dynamics and MM-GBSA support stable, high-affinity binding.
- Journal
- Computational Biology and Chemistry
- Published
- 2024 · Volume 110 · 108034
- Therapeutic area
- Tuberculosis
- Target
- InhA
- Access
- Subscription journal
32TuberculosisEthRIdentification of Mycobacterium tuberculosis transcriptional repressor EthR inhibitors: Shape-based search and machine learning studiesRupesh V. Chikhale, Gaber E. Eldesoky, Mahima Sudhir Kolpe, Vikramsinh Sardarsinh Suryawanshi, Pritee Chunarkar Patil, Shovonlal BhowmickHeliyon2024Open accessShow summary
First pageSummary
Identifies three molecules from a drug repurposing library as potential EthR inhibitors that could act as ethionamide boosters in tuberculosis, after physicochemical, toxicological and safety filtering, docking, molecular dynamics and binding energy studies. Machine learning suggests they are safe and easy to synthesise, and their known activities may allow faster repurposing as adjuvant therapy.
- Journal
- Heliyon
- Published
- 2024 · Volume 10 · Issue 5 · e26802
- Therapeutic area
- Tuberculosis
- Target
- EthR
- Access
- Open access
Open access means the publisher's own page is free to read under an open licence. Status as recorded by OpenAlex on 12 October 2026.
Take the record with you
All 32 papers with their authors, journals, volumes, DOIs and open access status, for a reference manager or for due diligence. Titles, authors and pages come from Crossref, the publishers' own record.
Check it yourself
Every title links to its DOI at the publisher. For an independent view of the record, see the profile of our Head of Science, Dr. Md Ataul Islam, on Google Scholar and ORCID.
Work with us
Put the methods to work
on your target.
The same technologies and the same scientists behind these papers are available to your programme, whether you license the software, run a campaign with us, or train your team.