I am a Graduate Researcher specializing in AI-driven drug discovery, with expertise in Machine Learning (ML), Deep Learning (DL), Generative AI (Gen-AI), RNA-Seq analysis, and molecular docking. I have experience in de novo drug design and drug-target interaction (DTI) prediction modeling.

I am passionate about leveraging computational biology and AI to advance precision medicine and cheminformatics. My aspiration is to lead interdisciplinary research integrating AI and pharmacogenomics for drug discovery.

Doctor of Philosophy (PhD) in Computer Science (Aug 2023 – Present)

• Relevant Coursework: Deep Learning and Neural Networks, Advanced Natural Language Processing, High Performance Distributed Systems, Advanced Algorithms

Master of Technology (MTech) in Biomedical Engineering (Aug 2021 – June 2023)

• GPA: 8.70/10.0
• Relevant Coursework: Application of Machine Learning in Biological Systems, Deep Learning Foundations and Applications, Medical Imaging, Medical Biotechnology, Cancer Biology

Bachelor of Technology (BTech) in Biotechnology (Aug 2017 – July 2021)

• GPA: 9.43/10.0
• Relevant Coursework: Introduction to C Programming Language, Data Structures and Algorithms, Database Management Systems, Biochemistry, Microbiology, rDNA Technology, Molecular Biology, Tissue Culture

(Sep 2023 – Mar 2025)

• Developed a multi-modal deep learning model for drug-target affinity (DTA) prediction by integrating differential gene expression information with drug and protein structural information.
• Demonstrated improved performance upon benchmarking.
• GitHub: debnathk/gramseq

(Feb 2024 – Apr 2024)

• Fine-tuned LLMs, including BERT, ERNIE-2.0, and UnifiedQA, using Hugging Face to evaluate commonsense capabilities on the Com2Sense benchmark.
• Identified UnifiedQA as the best-performing model.
• GitHub: debnathk/commonsense_inference

(Feb 2025 – Apr 2025)

• Generated augmented Electrocardiogram (ECG) images leveraging Neural Style Transfer (NST) using VGG-19 deep learning model.
• Improved model F1 score for disease classification by 2.4 times.
• GitHub: debnathk/nst_ecg

• Programming Languages: C, C++, Python, R, JavaScript
• Frameworks: PyTorch, TensorFlow
• Soft Skills: Research Communication, Adaptability, Team Player, Critical Thinking

Virginia Commonwealth University | Richmond, Virginia, USA (Aug 2023 – Present)

• Developing comprehensive deep-learning pipelines to investigate drug-target interactions (DTI) for exploring novel drugs for biological targets.
• Constructing generative deep-learning models for designing de novo drugs against multiple cancer subtypes.

Indian Institute of Technology, Kharagpur | Kharagpur, India (June 2022 – May 2023)

• Developed robust RNA-Seq analysis pipelines to uncover and examine primary differential gene expressions contributing to Glioblastoma Multiforme (GBM) progression.
• Identified potential protein-protein interactions (PPIs) linked to differentially expressed genes as novel drug targets for treating GBM using classical machine learning algorithms.

Calcutta National Medical College | Kolkata, India (Jan 2020 – Feb 2020)

• Designed experiments to investigate restriction digestion of PCR-amplified G6PD gene to detect the +563 C/T single nucleotide polymorphism (SNP).

Kusal Debnath, Pratip Rana, Preetam Ghosh

• Biomolecules, MDPI (Journal Category: Q1, Impact factor: 4.8) | Mar 2025
• DOI: 10.3390/biom15030405
• Developed an end-to-end deep learning model to analyze and extract features from drug and protein structural data along with gene expression information, demonstrating proficiency in predictive analysis of drug-target binding.

Cydney Dennis, D. Joshua Cohen, Kusal Debnath, Nofrat Schwartz, Brock Lodato, Jonathan Dillon, Tillat Batool, Matthew Halquist, Preetam Ghosh, Zvi Schwartz, and Barbara Boyan

• International Journal of Cancer, Wiley (Journal Category: Q1, Impact factor: 5.7)
• Conducted extensive ligand-protein docking analyses, utilizing advanced computational techniques to predict the binding affinity and interactions between the ligand of interest and numerous target proteins.

Kusal Debnath, Subhasish Dutta

• Cleaner and Circular Bioeconomy, Elsevier (Journal Category: Q2) | Dec 2023
• DOI: 10.1016/j.clcb.2023.100052
• Discussed integrating BESs with electrochemical capacitors, surfactants, biogas generation, 3D electrodes, and thermoelectric regulators, highlighting potential applications and research gaps.