Defense: Improving Variant Discovery using Pedigree-based Graph Alignment

Speaker Name
Charles Markello
Speaker Title
Biomolecular Engineering and Bioinformatics Ph.D. Candidate
Speaker Organization
Biomolecular Engineering and Bioinformatics Ph.D.
Start Time
End Time
Location
Virtual Event / Biomedical Sciences 200

Join us on Zoomhttps://ucsc.zoom.us/j/93119971444?pwd=SkdOMERZYzZtTDZpYlp5emlOUzRWZz09 / Passcode: 1.e4c5

 

Description: Next Generation Sequencing (NGS) methods have become widely adopted in the fields of medical genetics. A majority of these methods currently rely on the use of a linear genome sequence as a reference point for sequence alignment and genetic variation.

Although an effective heuristic, the use of a linear reference genome creates an allele detection bias: alleles in the reference are much easier to find than alleles not in the reference. This bias is acute for structural variants, and for variants in complex regions of the genome with high levels of repetition.

Here I present a workflow for improving the discovery of variants from NGS data within regions of high and low variant composition by leveraging recent advances in graph reference technology with the genomic variant information contained within pedigrees and population data.

In addition, I also present an enhanced workflow that produces a short list of deleterious variants present in an individual. To demonstrate the utility of these methods, I applied these methods to discover candidate deleterious variants that are potentially causal to new and rare genetic disorders of real human pedigree NGS data.

Advisor
Benedict Paten