Empowering Multiomics and Single Cell Based Predictive Analytics
for Biomedical Applications with G2PDeep
Abstract
Accurate phenotype prediction from high-dimensional omics data remains a major challenge in computational biology, particularly in the presence of cellular heterogeneity and demographic variability. Diverse types of bulk multiomics data such as gene expression, microRNA (miRNA) expression, protein expression, DNA methylation, single nucleotide polymorphisms (SNPs), copy number variations (CNVs), as well as single-cell and spatial transcriptomics are regularly generated for any organisms including humans and other organisms for all kinds of research questions. With so much data available there is a growing need for conducting predictive analytics for phenotypic predictions for clinical and translational applications. We have developed G2PDeep, a web-based platform powered by deep learning, for phenotype prediction and markers discovery from multi-omics data. The server provides multiple services for researchers to create deep-learning models through an interactive interface and train these models using an automated hyperparameter tuning algorithm on HPC resources. Users can visualize the results of phenotype, markers predictions, and perform Gene Set Enrichment Analysis for the significant markers, to provide insights into the molecular mechanisms underlying complex diseases. G2PDeep is publicly available at https://g2pdeep.org/. Additionally, our IRnet method, designed for immunotherapy response prediction for cancer patients is also available at https://irnet.missouri.edu
Short Biography
Dr. Trupti Joshi is the Senior Associate Dean for Informatics and Population Analytics and Professor in the Department of Biomedical Sciences at Joan C Edwards School of Medicine at Marshall University. She holds an Adjunct faculty appointment in the Department of Electrical Engineering and Computer Science (EECS) at University of Missouri-Columbia. She is the Principal Investigator of the WV-INBRE (NIGMS / NIH Award P20GM103434) and has previously served as Director of Translation Bioinformatics and core faculty with Department of Plant Sciences and Technology (DPST) and MU Data Science and Informatics Institute (MUIDSI), at University of Missouri-Columbia (UM). She is a renowned expert in the translational bioinformatics field with over 25+ years of experience and leads The Translational Bioinformatics and AI Innovations Lab (TBAiL). She has published more than 175 scientific papers and has developed several multiomics data integration frameworks, computational methods and AI based solutions for clinical and translational research.
