Conceptual Poor Alignment Understanding

[arcolombo698]

Hello.
I am working on some RNAseq and received a few poor reads from our sample set. One possible undersatnding is cDNA contamination. note we aligned untrimmed reads because if there is "junk" in the read, then it will not align to the reference genome and automatically become omitted.

well what other possibilities could this be? My troubleshoot workaround is to trim the reads for quality and omit the adapters. Is it possible for adapter contamination in the prep kit?

if we remove the low quality reads, this might remove the entire read, or will it?

How has your team interpreted poor aligned reads? I will do a quality analysis to understand why the reads were mal-aligned. what are common things to look for?

[Brian Bushnell]

FastQC is a pretty good tool for looking at raw data, and it is often helpful if you have a problematic library to run FastQC and attach the pdf output to your post, so other people can look at it. If the FastQC graphs look normal the problem may be contamination.

I don't recommend aligning non-adapter-trimmed RNA-seq reads, as that will incur bias (against short transcripts, for example). For contamination detection, a good approach is to map the reads to the reference at high sensitivity, then take a subset of the unaligned reads and map them to a broad dataset such as nt to see what they hit.