Fully Realizing Pangenomics Alignment

PROJECT ABSTRACT Read aligners first build an index from one or more reference genome(s) and subsequently use it to find and extend matched subsequences between sequence reads and the reference(s). The bottleneck of using these read aligners to index thousands of human reference genomes is the space and time needed for construct

and store the index. Hence, in the case of the human genome, it is common to restrict interest to alignment of the standard reference genome, i.e., GChr38. Yet, the absence of diversity in this single reference genome can cause substandard results in downstream analysis, impacting the ability to identify and study genetic

variation. To address the shortcomings associated with using a single reference genome, the concept of a pange- nomics reference genome has been introduced and adopted. For example, Giraffe, VG, and Moni all aim to index a population of genomes in a manner that enables read alignment. Although, these pangenomics

aligners have been shown to improve on the accuracy over standard read aligners (e.g., BWA and Bowtie) there exists several challenges that prevent these methods from being used in practice for downstream anal- ysis, such as variant calling. The goal of this proposal is to develop algorithms to address these challenges

and fully enable pangenomics alignment. In particular, we will create methods for selecting (from a large population) a subset genomes that will achieve the most accurate alignment results, develop a pangenomics scoring scheme that will enable the alignments from a pangenome to be attained, and disseminate our

methods in a user-friendly manner that enables automated updates.