Transcriptome assembly if adaptors are not well removed

Question

[Cacau]

I am wondering how the removal of adaptor sequences will impact the transcriptome assembly. For example, in 10% of my reads contain adaptors which are not removed. If I try to assemble the transcriptome, will some of my unigene dataset contain adaptors? Besides, will the result be different if I use a genome-guided or de novo asssembly method?

Answer 1

[WouterDeCoster]

I think assembly is one of the things were you definitely want to remove adapters, if not you'll have spurious overlaps creating erroneous contigs/transcripts, just because they share an adapter.

[Asaf]

I've experienced such a catastrophic event. On the other hand you might get cool chimeric proteins.

[WouterDeCoster]

Reminds me of this blog post: http://www.opiniomics.org/we-need-to-stop-making-this-simple-fcking-mistake/