Heath, H. (BMEB) – Linking Altered mRNA Processing to Translational Regulation in Lung Cancer

In lung cancer, recurrent alterations in RNA processing regulators reshape both the proteins genes encode and the sequences that regulate their production. Understanding how mRNA processing influences protein production requires resolving the coding and regulatory sequences present in individual mRNA isoforms. Transcription start site (TSS) choice, alternative splicing (AS) , and polyadenylation site (PAS) choice determine which coding sequences (CDS) and untranslated regions (UTRs) are expressed together. This allows transcripts from the same gene to differ in the structure and abundance of proteins they generate. Short-read RNA sequencing is only able to identify single splicing changes, failing to link them together, while long-read RNA sequencing (LRS) reads can span across exons allowing for more confident transcriptome assembly. However, accurate transcript end detection is not a priority for most LRS transcript assemblers. As a result, they may miss transcript end changes that could obscure real variations in UTR regulatory sequences. In preliminary benchmarking across multiple cell lines and LRS technologies, we compared validated transcript-end locations to isoform assemblies. Here we found TSSs and PASs are present in long reads but not recovered by current transcriptome assembly methods, in addition to incorrect transcript ends. To address these limitations, I will develop FLAIR-Transcript-End Detection (FLAIR-TED), a computational method to improve identification of accurate alternative transcript ends during transcriptome assembly (Aim 1). I will then use FLAIR-TED on long-read polysome fractionation (LRPF) sequencing to identify splicing and UTR sequence differences associated with ribosome association in lung cancer cell-line models – revealing how splicing and UTR choice work together to impact translation and protein production (Aim 2). Finally in matched lung adenocarcinoma (LUAD) primary tumor and normal tissues, I will examine whether recurrent isoform usage changes favor transcripts with distinct predicted translational efficiencies (Aim 3). Together, these studies will connect disease-associated mRNA processing changes to the coding and regulatory sequences expressed and their potential translational consequences.

Event Host: Harrison Heath, PhD Student, Biomolecular Engineering & Bioinformatics

Advisor: Angela Brooks 

Zoom: https://ucsc.zoom.us/j/92519872137?pwd=mn1j67FWrojyx6bPt8ekwugm0Tcpaw.1

Passcode: 208120