Ales Varabyou, Johns Hopkins – “Novel Approaches for Segregating Noise and True Signals in RNA-seq Data for Comprehensive Genome Annotation”
June 12 @ 4:00 pm - 5:00 pm
This talk presents new methods to analyze and annotate the complex and noisy transcriptional landscapes revealed by large-scale RNA-seq studies. It introduces CHESS, a comprehensive catalog of human genes and transcripts derived from deep RNA sequencing, which covers both protein-coding and noncoding genes and captures a high level of transcriptional noise. It shows how transcriptional noise affects gene expression analysis and how to overcome its challenges using novel tools. One of these methods is TieBrush, a software package for efficient processing and summarization of large sequencing datasets, which can be used to design methods to filter out noise and improve transcript quantification. Another method is ORFanage, a system designed to assign ORFs to transcript models using extremely efficient similarity comparison with reference annotation, which can identify novel coding potential and functional domains. It also demonstrates how to integrate protein structure predictions to identify functional isoforms. Finally, it shows how these methods were used to create a comprehensive annotation of the human genome.
Ales Varabyou is a PhD candidate in Computer Science at JHU, working on developing methods to detect transcriptional anomalies in human and viral genomes using RNA-seq data. His advisors are Steven Salzberg and Ela Pertea from JHU BME. He will graduate in 2023 and join JHU as a postdoctoral researcher.
All Welcome! Note this event will take place on Zoom (link to be provided soon) here: TBD
To be added to the ABC seminar email list, please email firstname.lastname@example.org