Joseph Russell, PhD

Joseph Russell, PhD

Principal Scientist and Group Leader
Applied Biology & Bioinformatics
MRIGlobal

 

Dr. Joseph A. Russell is a Principal Scientist and Group Leader for the Applied Biology & Bioinformatics capability area within MRIGlobal’s Life Sciences Resource Center. His team is focused on the development of novel computational approaches, platform systems, and workflow improvement for applied microbial ecology, biosurveillance, microbial forensics, genomic epidemiology, and public health. Some examples of this work involve streamlined metagenomics and interpretive bio-analytics for environmental and clinical contexts, genotype-to-phenotype predictive modeling, single-person portable/automated/ruggedized molecular biosurveillance platforms, host-pathogen interactions, arbovirus ecology, and applied synthetic biology. Dr. Russell received his Ph.D. in geomicrobiology from the University of Delaware in 2015, studying the microbial ecology of deep subseafloor sediment and oceanic crust samples. He joined MRIGlobal later that year as a post-doctoral research associate and has worked in the above contexts on various U.S. government programs with DTRA, DoD, NASA, CDC, DARPA, and others.

Applications of Genomic Feature Engineering for Emerging Infectious Disease Biosurveillance

Rapid detection and surveillance of emerging infectious diseases (EIDs) are crucial to global public health. In this study, we explore the benefits of feature engineering of genomic motifs for improving the accuracy and efficiency of EID biosurveillance applications. By extracting relevant genomic features from pathogen sequences and employing advanced machine learning techniques, we aim to develop a robust framework for early identification and monitoring of novel EIDs. Our methodology involves the identification and extraction of informative genomic motifs from a diverse dataset of known pathogens, focusing on RNA viruses. We apply various feature engineering techniques to generate a comprehensive genomic feature set. Subsequently, we employ machine learning algorithms to classify and predict emerging pathogens based on their genomic signatures. Our recent work demonstrates the utility of feature-engineered genomic motifs leading to genotype-to-phenotype characterizations of public health concern in novel pathogens. By utilizing these genomic features, our approach allows for the rapid identification of potential EIDs and supports timely public health interventions. This study underscores the importance of leveraging advanced bioinformatics and machine learning techniques to augment traditional infectious disease surveillance systems and highlights the potential for genomic motif feature engineering to bolster global EID biosurveillance efforts.