Minakshi Sahoo

Dynamic Research Scholar with a proven track record at CV Raman Global University, excelling in computational modeling and sustainable chemistry. Expert in HPLC and data analysis, I foster collaboration and innovation, mentoring peers while driving impactful research. Passionate about leveraging AI tools for enhanced scientific communication and drug discovery processes.

Predictive Catalyst Selection Framework: Developed a novel computational tool using Python and Scikit-Learn to model and predict catalyst performance. This framework integrates data on synthesis parameters and characterization results to forecast catalyst sustainability and suitability for specific bioprocesses, enabling rapid, data-driven selection of optimal candidates.

Drug Stability & Solubility Prediction Model: Engineered a machine learning model using Python (Scikit-Learn, TensorFlow) to predict the aqueous solubility and degradation pathways for small molecules. The model leverages chemical descriptors and experimental data to provide critical, early-stage insights into a compound's potential stability and shelf-life, accelerating lead optimization.

In-Silico Bioactivity & Safety Prediction Pipeline: Built a custom workflow integrating PASS Online, SwissADME, and ProTox-II to rapidly screen novel molecules. This pipeline predicts biological activity and evaluates ADMET properties, significantly accelerating the initial stages of hit-to-lead identification by flagging potentially toxic or nonviable compounds.

Catalyst-Ligand Interaction Modeling: Created a project-specific tool using AutoDock Vina and Discovery Studio to simulate and visualize binding affinity between synthesized catalysts and target molecules. This process, analogous to structure-based drug design, provided critical insights into reaction mechanisms and enabled data-driven optimization of catalyst design for enhanced efficiency and reusability.

Curriculum & Training Contribution: Developed and delivered training materials for junior researchers on the practical application of molecular docking and ADMET prediction tools, bridging the gap between theoretical chemistry and applied computational drug discovery.

Data Analytics and Machine Learning with Python, ONLEI Technologies, 11/01/23, 05/01/24

• Das, S., Roy, S., Manohar, E. M., Bandyopadhyay, S., Singh, M., Sahoo, M., Giri, S., Lande, S.). Transforming CO₂ into methanol: a critical review focusing on the noble metal-based catalytic conversion through heterogeneous and electrochemical processes Progress in Energy and Combustion Science, 85, 100905
• Dash, B., Digal, L., Sahoo, M., Mohanty, J. (2024). Potential use of dry fallen bamboo leaves for sustainable bioethanol production Department of Chemistry, C. V. Raman Global University.
• Sahoo, M., Mohanty, J., Singha, D. (2024). Bioethanol from Residual Biomass: A Sustainable Approach for Industrial Pipeline Corrosion Prevention. CV Raman Global University, Bhubaneswar, Odisha, India. (Abstract Accepted).

Conferences & Presentations

• AI Drug Discovery & Development Summit, Boston, MA, USA (November 2025)
• Gordon Research Conference (GRC) on Computer-Aided Drug Design, Portland, ME, USA (July 2025)
• XXXI Symposium on Bioinformatics and Computer-Aided Drug Discovery (BCADD-2025), Virtual (October 2025)
• Drug Discovery Chemistry: AI/ML for Early Drug Discovery, San Diego, CA, USA (April 2025)
• International Conference on Artificial Intelligence in Healthcare and Wellness (AIHW 2025), Hyderabad, Telangana, India (November 2025)
• Webinar on Recent Trends in Inorganic Chemistry (RTIC-2022), Bhubaneswar, Odisha, India (April 2022)
• International Webinar on Recent Trends in Materials and Chemical Science (RPMCS-2022), Bhubaneswar, Odisha, India, March 2022
• National Conference on Current Trends in Minerals and Materials Technology, Bhubaneswar, Odisha, India (February 2019)

• Python (Pandas, SeaborThe listed skills cover a range of tools and techniques for data analysis, machine learning, and drug design, including: Python libraries (Matplotlib, Scikit-Learn), R, KNIME, Orange Data Mining, TensorFlow, Keras, PyTorch, generative models, NLP for bioactivity data mining, Google AutoML, DataRobot, RDKit, Open Babel, AutoDock Vina, PyRx, Schrodinger Suite, MOE, ChemAxon, SwissDock, Discovery Studio, ChemDraw, ligand- and structure-based drug design, QSAR, molecular docking, virtual screening, ADMET prediction, lead identification and optimization, pharmacophore modeling, bioactivity and toxicity prediction, handling chemical libraries (ChEMBL, PubChem), similarity searching, clustering, database management (SQL, MongoDB), web application frameworks (Dash, Plotly, Flask, Streamlit), Google Colab, OriginPro, Empower, LabSolutions, SciNote ELN, LabTwin, LabVoice, and analytical chemistry techniques (HPLC, GC, XRD, FTIR, SEM, BET Analysis).