Redundancy vs. Artifacts in de novo transcriptomes

[tboothby]

Hey all,

I was wondering if anyone knows of (or can think of) a way to assess whether isoforms in a de novo transcriptome assembly are real or an artifact of sequencing/assembly.

I was thinking I might make "unigenes" from my transcripts and then map reads against them. I would then load this data into a genome browser and spot check for low coverage SNPs and insertion/deletions... do you think this would be informative?

Would there be a more high throughput way of doing this than just visually inspecting read alignments?

-Thomas

[Melissa]

If you want to differentiate SNPs/indels from sequencing errors, you should map your reads to the genome/transcript to detect SNPs/indels. Say, using BWA and Samtools.