Next Publication In:
Days: 00
Hours: 00
Minutes: 00
Seconds: 00

ARTIFICIAL INTELLIGENCE-ASSISTED DRUG DESIGN COMBINED WITH MOLECULAR DOCKING FOR PRIORITIZATION OF POTENTIAL LEAD MOLECULES ACROSS FIVE THERAPEUTIC TARGETS

📘 Volume 11 📄 Issue 8 📅 August 2026

👤 Authors

Harshveer Singh Jaitawat 1 , Dr. C.S. Sharma 1 , Dr. H.S. Udawat 2 , Dr. P.S. Naruka 2
1. Department of pharmaceutical chemistry, B.N. College of Pharmacy, Faculty of Pharmacy, B.N. University, Udaipur, Rajasthan, India.,
2. Department of pharmaceutics, B.N. College of Pharmacy, Faculty of Pharmacy, B.N. University, Udaipur, Rajasthan, India.,

📄 Abstract

Background: Artificial intelligence-assisted drug discovery can reduce the chemical search space by integrating molecular generation, drug-likeness assessment and structure-based docking. However, computational rankings require careful validation and should be interpreted as prioritization rather than evidence of therapeutic efficacy. Objectives: To integrate AI-assisted molecular design, physicochemical/ADMET screening, molecular docking and ligand–protein interaction analysis for identification of potential leads across selected cancer, metabolic and neurodegenerative targets. Methods: A documented computational workflow was applied to five ligand-bound human targets—EGFR (1M17), KRAS G12C (5V71), DPP-4 (1X70), PTP1B (1C83) and AChE (4EY7). The supplied thesis dataset described progressive filtering of a 1350-member virtual library to 14 final candidates, followed by drug-likeness/ADMET assessment, AutoDock Vina docking and redocking validation. Candidate prioritization integrated docking score, target-relevant interactions and computational pharmacokinetic/toxicity characteristics. Results: All five reference systems passed the predefined redocking criterion of <2.0 Å, with RMSD values of 0.65–1.14 Å. The prioritized leads were CAN-AI-01 (EGFR, −9.71 kcal/mol), KRA-AI-01 (KRAS G12C, −9.12 kcal/mol), DIA-AI-01 (DPP-4, −9.35 kcal/mol), PTP-AI-01 (PTP1B, −8.84 kcal/mol) and NEU-AI-01 (AChE, −12.42 kcal/mol). All five showed zero reported Lipinski violations, high predicted gastrointestinal absorption, negative Ames predictions and no predicted hERG inhibition or hepatotoxicity in the supplied screening dataset. Conclusions: The integrated workflow supports computational narrowing of chemical space and target-specific lead prioritization. The identified molecules should be considered computational leads requiring biochemical, cellular, pharmacokinetic and toxicity validation before any therapeutic claim is made.

🏷️ Keywords

Artificial intelligence Drug design Molecular docking AutoDock Vina Virtual screening ADMET Lead prioritization Structure-based drug discovery

🔗 DOI

DOI - ( https://doi.org/10.36713/epra28874)

📚 How to Cite:

Harshveer Singh Jaitawat, Dr. C.S. Sharma, Dr. H.S. Udawat, Dr. P.S. Naruka , ARTIFICIAL INTELLIGENCE-ASSISTED DRUG DESIGN COMBINED WITH MOLECULAR DOCKING FOR PRIORITIZATION OF POTENTIAL LEAD MOLECULES ACROSS FIVE THERAPEUTIC TARGETS , Volume 11 , Issue 8, August 2026, EPRA International Journal of Research & Development (IJRD) , Pages: 91 - 97 , DOI: https://doi.org/10.36713/epra28874

🔗 PDF URL

https://cdn.eprapublishing.org/article/1786980567671-13.EPRA28874.pdf

📄 PDF Preview

Click the button above to load the PDF.