Areas of Research
Computational Biology | Artificial Intelligence | Biomarker Discovery | Early Cancer Detection | Prognosis and Survival | Recalcitrant Cancers | Immunotherapy | Tumor Microenvironment | Health Disparities
Research Projects:
Identification of novel molecular signatures in cancers:
Lung and Breast cancer are the leading cancers in the U.S. with adverse differences in cancer burden in the medically underrepresented population groups. Genomic differences between lung tumor populations can lead to differences in risk and response to treatment. We have previously reported histological and population differences in tumor biology as a key driver of novel molecular biomarkers and neoantigens. In addition to providing insights into the role of splicing events as potential drivers of cancer, profiling of genomic signatures (gene expression and DNA variants) has also uncovered novel molecular biomarkers for the diagnosis, treatment, and prognosis of cancers. The multifaceted role of tumor antigens has furthered our understanding of population and histological differences in the biology underlying cancer and hold significant potential to aid in development of new therapeutic agents to mitigate cancer and cancer disparities. Our research revolves around understanding that race, ethnicity, genetic ancestry, and social determinants of health can impact biology, and trying to identify such differences is key to advancing health equity.
AI/ML application for biomarker discovery and predictive analysis in cancer:
Studying genetic insight with the application of Artificial intelligence (AI) and machine learning (ML) and state-of-the-art bioinformatics approaches can improve the processes of discovering cancer causing variants and decode genetics of complex cancer heterogeneity enabling improved personalized treatments. AI/ML has advanced in several areas of life science however, it’s progress in the field of multi-omics is less mature because of a cascade of technical issues and challenges that remain unaddressed. We have developed a model with a novel approach combining traditional statistics and a nexus of cutting-edge AI/ML techniques to identify significant biomarkers for our predictive engine by analyzing integrated clinical, demographic, and multi-omic data. Our model combines, knowledge-driven and data-driven approaches to create an AI engine that creates risk profile of each of the individuals that may help predict and possibly prevent cancer. This can facilitate identifying patients at a higher risk of disease. These studies challenge the paradigm that although there are efficient treatments such as correction of risk factors and specific medications, the use of AI is necessary to detect these factors in a longitudinal data for a given subject. The potential use of the prediction models goes beyond this research and can be applied to any disease.