Research Projects

Our group is interested in Systems immunology research which employs discrete state and statistical modeling methods along with data curation and mining to gain a comprehensive understanding of the intricate dynamics within the immune system. We are particularly interested in studying host responses during long term HIV infection, HIV vaccination and allergy.

Discrete state modeling involves representing biological systems as discrete entities or states, allowing us to simulate and analyze complex interactions between various immune cells, signaling molecules, and pathogens. This approach enables the investigation of emergent properties and dynamics that arise from interactions between individual components of the immune system. We use discrete-state models for gene-sets analysis using network topologies and to map cells to pathway specific steady states.

Learn more about the discrete-state modeling and use our tool-set BONITA

Statistical modeling methods provide quantitative frameworks to analyze experimental data, identify patterns, and infer relationships between different components of the immune system. By integrating these modeling approaches, we can unravel the complexities of immune responses, identify key regulatory mechanisms, and ultimately develop novel strategies for diagnosing and treating immune-related disorders.

Read about ensemble machine learning model developed using 26 vaccine regimens.

Data curation and mining are essential in the biomedical field for improving data quality, integrating diverse datasets, uncovering insights, advancing personalized medicine and accelerating drug discovery. The data curation and mining drive innovation and enable evidence-based decision-making. We are particularly interested in data curation to improve definitions of immune functions and signaling cascades.

Learn more about data curation using our WikiNetworks tool and data mining using meta-analysis.

Gene-Set enrichment Analysis methods: Gene Set Analysis (GSA) provides a useful approach for quantifying molecular functions and pathways in high-throughput datasets, facilitating the biological interpretation of omics data from humans which is highly heterogenous. We have developed several methods to improve GSA by accounting for extent of overlap across gene-sets, estimating gene-sets at the single cell resolution and incorporating pathway topologies.