Deep Learning Approaches for Predicting Virus-Host Interactions and Drug Response [virtual]
Virus infection is commonly observed in nature. Recently, SARS-CoV-2 has caused a global pandemic which has infected over 200 million individuals (as of August 2021). An effective and efficient detection of viruses in the host genomes, and tracking how they interact with the host genomes, are currently a main challenge. In this talk, I will first introduce our computational approaches to detect viruses and their integration sites in the host genomes from next generation sequencing data. Then, based on our recently developed virus integration site database (VISDB), we developed a deep learning method, DeepVISP, for virus site integration prediction and motif discovery. To study COVID-19, we developed a deep learning method, DrivAER: Identification of Driving transcriptional programs with AutoEncoder derived Relevance scores from single cell RNA sequencing (scRNA-seq) data. We applied DrivAER to COVID-19 scRNA-seq data, as well as integrative analysis of COVID-19 genome-wide association studies (GWAS) and transcriptome-wide association studies (TWAS). Our investigation identified several genes, regulatory factors and cellular trajectories that might be involved in COVID-19 disease severity. Finally, I will present a deep generative neural network for accurate drug response imputation. In this work, we developed a deep variational autoencoder (VAE) model to compress thousands of genes into latent vectors in a low-dimensional space. We demonstrated that these encoded latent vectors could accurately impute drug response, outperform standard signature-gene based approaches, and appropriately control the overfitting problem.
Dr. Zhongming Zhao holds Chair Professor for Precision Health and is the founding director of the Center for Precision Health, the University of Texas Health Science Center at Houston (UTHealth), where he previously held Dr. Doris L. Ross Professor of Biomedical Informatics. Before he joined UTHealth in 2016, he was Ingram Endowed Professor of Cancer Research, Professor (with tenure) in the Departments of Biomedical Informatics, Psychiatry, and Cancer Biology at Vanderbilt University Medical Center, Chief Bioinformatics Officer of the Vanderbilt-Ingram Cancer Center (VICC), and the Associate Director of the Vanderbilt Center for Quantitative Sciences. Dr. Zhao has unique, interdisciplinary training: he received his master’s degrees in Genetics (1996), Biomathematics (1998), Computer Science (2002), Ph.D. degree in Human and Molecular Genetics (2000), and Keck’s Postdoctoral Fellowship in Bioinformatics (2001-2003). Dr. Zhao has broad interest in bioinformatics, genomics, precision medicine, and machine learning and has co-authored over 400 publications in these areas (cited by >16,000 times, H-index = 67). Dr. Zhao has served as the Editor-in-Chief, Associate Editor, or editorial board member of 22 journals. Dr. Zhao is the founding president of The International Association for Intelligent Biology and Medicine (IAIBM, 2018). Dr. Zhao has received several awards, including the Keck Foundation Post-doctoral Fellowship (twice: 2002, 2003), the NARSAD Young Investigator Award (twice: 2005, 2008), a NIH-funded VPSD Career Development Award in GI Cancer (2009).