Vertebrate genomics is heading into an era where the release of hundreds of new reference genomes a year is going to be the norm. If we are to effectively engage with the new biology, we need to be able to annotate genes on these genomes in an accurate and efficient manner. Unfortunately, the current state of gene annotation software requires the use of multiple tools and hours of manual curation. In order to remedy this, I propose a three-part plan: improve the working of the Comparative Annotation Toolkit and use it to annotate genomes for large-scale genome projects, build a complete telomere-to-telomere marmoset genome and annotate it with CAT and provide a well-annotated New World Monkey complete reference genome, build a powerful ab initio gene annotation method for vertebrates taking advantage of recent advances in deep learning and protein language models.

Event Host: Prajna Hebbar, Ph.D. Student, Biomolecular Engineering & Bioinformatics

Advisor: Dr Benedict Paten