1st presenter: Elin Sandertun Røed
Title
"Prescription drug use, the gut microbiome and colorectal cancer"
Abstract
Colorectal cancer (CRC) incidence rates have rapidly increased over the last few decades. Screening is recommended to reduce both CRC incidence and mortality, and recent evidence suggests that the gut microbiome could be a source of a non-invasive diagnostic biomarker for CRC. However, associations of prescription drug use with both CRC risk and the gut microbiome complicate the development of such a biomarker. In this hypothesis free study, we investigated associations of prescription drug use with the gut microbiome and screening-detected colorectal lesions in a cohort of 1,034 fecal immunochemical test-positive screening participants aged 55-77 years. By combining gut metagenomic profiles with clinicopathological, demographic, lifestyle, and prescription data, we identified differences in the gut microbiome profile of drug users and non-users, and found that commonly used drugs may influence microbial CRC screening candidates.
2nd presenter: Divyeshkumar Amrubthai Patel
Title
"Computational strategies for quantification of transposable elements expression"
Abstract
Transposable elements (TEs) are mobile, repetitive DNA sequences that constitute nearly half of the human genome and play important roles in development and disease. Accurate quantification of TE expression from RNA seq data remains challenging because the high sequence similarity among TE copies dispersed throughout the genome leads to extensive multi mapping of sequencing reads. In standard RNA seq workflows, these ambiguously mapped reads are often discarded, resulting in systematic underestimation of TE expression.
In this talk, I will first introduce the software for Quantifying Interspersed Repeat Expression (SQuIRE), which implements multi mapping strategies for quantifying TE expression from RNA seq data1. I will then discuss established computational approaches that address uncertainty in read assignment by probabilistically redistributing ambiguously mapped reads to their most likely source transcripts2. Finally, I will present a TE analysis pipeline that we recently applied, based on these existing methods, to identify differentially expressed TEs at both locus specific and subfamily levels.
BiLS chair
Blanca Candelas Galavis