Research Projects

Type 2 diabetes mellitus (T2DM) is a rapidly growing public health crisis in India and globally. As per WHO, India ranks among the top countries with the highest diabetes burden, affecting over 77 million people, a figure projected to reach 134 million by 2045. T2DM is frequently associated with cancers such as colorectal, breast, liver, and pancreatic, significantly increasing patient morbidity and mortality. Peroxisome proliferator-activated receptor gamma (PPAR-γ), a nuclear receptor involved in glucose and lipid metabolism, adipocyte differentiation, and anti-inflammatory pathways, has emerged as a key target for treating both T2DM and cancer. Current pharmacological interventions targeting PPAR-γ are dominated by synthetic thiazolidinediones, including pioglitazone and rosiglitazone. While effective in improving insulin sensitivity, these drugs are linked to serious side effects like fluid retention, weight gain, cardiotoxicity, and hepatotoxicity. Even newer investigational drugs, such as AMG-131 and Muraglitazar, face safety concerns and remain costly, limiting their accessibility, especially in resource-limited settings like India. In this context, natural compounds (NCs) offer a promising alternative for safer, more affordable PPAR-γ modulation due to their structural diversity, ethnomedicinal relevance, and generally favorable toxicity. It is hypothesized that robust, OECD-compliant QSAR models, trained on high-quality bioactivity data and enriched with molecular descriptors, can reliably predict PPAR-γ inhibition. Such models can aid in identifying safe and potent natural modulators of PPAR-γ with translational potential in metabolic and cancer therapy. Thus, this work aims to develop OECD-compliant QSAR models using AI and ML techniques to predict the PPAR-γ inhibitory activity of small molecules. These models will be trained on experimentally validated datasets and characterized using a wide range of descriptors (2D/3D). Following rigorous validation, via internal/external testing, Y-randomization, and applicability domain analysis, the models will be used to screen a curated library of compounds for potential PPAR-γ inhibition. Contingent upon the commercial availability of shortlisted compounds, experimental validation may be carried out through in vitro or in vivo assays to confirm biological relevance. This integrative strategy is expected to yield a predictive, regulatory-acceptable platform for early identification of natural PPAR-γ modulators with dual anti-diabetic and anti-cancer properties. If successful, the project will advance both the fundamental understanding of structure-activity relationships in bioactive natural products and the development of accessible drug discovery pipelines. The outcomes hold strong translational value, particularly in developing countries where safe, effective, and affordable therapeutics are urgently needed to manage metabolic diseases and associated cancers.