RII Track-4: NSF: Extracting Pan Genomic Information from Metagenomic Data: Distributed Algorithms and Scalable Software

University Of Alaska Fairbanks Campus, Fairbanks AK

Investigators

Abstract

The analysis of metagenomes, i.e., genetic data collected directly from environmental samples, has become integral to various research areas, including climate studies, human health, the discovery of rare earth elements, social and environmental resilience planning, and more. Genetic data collected in this manner typically contains a mixed population of multiple microbial communities. Scientists often aim to extract and represent the genomic diversity information of a particular microbial species from such a mixed population, a process known as pan-genomic information representation. However, there is a shortage of theoretically sound and biologically valid algorithms capable of performing metagenomic or pan-genomic analysis. Furthermore, the vast amount of genetic data generated by high-throughput genome sequencing machines necessitates that these algorithms be scalable and distributed in nature. This project will investigate both the distributed algorithmic aspect and its practical implementation to extract pan-genomic information from large-scale metagenomic datasets. This research aligns with at least six different research areas prioritized by Alaska EPSCoR in their latest Science and Technology Plan, including Community Resilience, Resource Extraction, Food-Energy-Water Nexus, Renewable Resources, Environmental Monitoring, and One Health. This RII Track-4: NSF fellowship will enable an Assistant Professor and a graduate student at the University of Alaska Fairbanks (UAF) to collaborate with scientists at North Carolina State University (NCSU) and utilize their resources. The Principal Investigator (PI) will work alongside experts in the field of bioinformatics and algorithms to develop a set of provably correct, scalable, and distributed algorithms with low time complexity for extracting pan-genomic information from large-scale metagenomic datasets. Additionally, utilizing cutting-edge high-performance computing (HPC) resources at NCSU, the PI aims to create a preliminary version of an HPC-compliant software framework implementing these algorithms. The analytic pipeline comprises four distinct stages: 1) metagenomic error correction, 2) metagenomic assembly, 3) binning and annotation of the assembled genome, and 4) creating the pan-genomic profile of the available microbes. Each of these stages presents algorithmic challenges. The diverse coverages of microbiomes in the metagenomic dataset, coupled with instrumental errors, render the process of identifying the actual species and their genetic diversity exceedingly challenging, necessitating extensive research in string matching and graph analysis. The distributed software implementation must address numerous HPC challenges. The research outcomes, including publications and open-source codebases, will support multiple research activities at UAF, focusing on arctic climate change, arctic marine biology, Alaska Native health, among others. The collaboration facilitated by this fellowship will also lay the foundation for an interdisciplinary Ph.D. program at UAF, encompassing computer science, bioinformatics, and indigenous science concentrations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.