Detecting and linking fusion transcripts to carcinogenesis

Monday, October 24, 2011 — Poster Session I
Noon – 2:00 p.m.	Natcher Conference Center	CIT	BIOINFO-11

Authors

• X Liu
• W Xiao
• R Schmitz
• S Jhavar
• G Wright
• J Powell
• L Staudt

Abstract

Fusion transcripts, which are encoded by fusion genes or by two different genes by subsequent trans-splicing, may lead to fusion proteins or aberrant expression of oncogenes. Many fusion transcripts have been reported and attributed to various types of cancers. The advancement in Next-generation sequencing technology, especially paired-end RNA-Seq, has offered tremendous advantage in detecting such chimeric events due to its great coverage. With an open-source fusion detection framework, tools developed in-house, together with the powerful supercomputing cluster Biowulf at NIH, we are developing and streamlining a fusion transcript detection process to identify these molecular events. The adapted modular framework is flexible with different sequence alignment tools, can remove spurious candidates through multiple filtration steps, and identify the breakpoint junctions and exact sequences spinning the junctions. The capacity of this pipeline will enable us extract knowledge from the terabytes of NGS sequence data that has been generated from more than 200 samples in lymphoma research. The results, especially the recurrent fusion transcripts, could help expand our understandings in cancer genesis and progression, and further have clinical implications in serving as diagnostic markers and therapeutic targets.

back to top