brianjohnhaas
commented
5 years ago Hi,
responses below
On Sat, Oct 20, 2018 at 10:53 AM MichaelsGITIGIT notifications@github.com wrote:
Two questions regarding contamination in RNA-seq data and its handling in a Trinity workflow:
1.
Do you, ideally, (1) filter contamination on the level of raw reads or do you filter (2) after Trinity assembly? We are talking about viruses/prokaryotes as the source of contamination. Not eukaryotes. So an assembly should probably not mix contamination with RNA from my species of interest. For this reason, I would intuitively go for filtering after assembly.
Some use sortmerna to remove rRNA before doing assembly. As far as other filtering, I'd do it after assembly, but I haven't done any benchmarking or anything to determine what way is best. Others might comment here on their experiences.
1.
Which tools did you use / do you recommend? On the raw read level Kraken2 is a good candidate I guess. What would you recommend when filtering on assembly level?
Kraken or Centrifuge are my go-to tools. I tend to go w/ Centrifuge but generate Kraken-style reports using their included utilities.
1.
Thx!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Trinotate/Trinotate.github.io/issues/10, or mute the thread https://github.com/notifications/unsubscribe-auth/AHMVX5ywW4yPpAp336sqr4-Oms0LicWDks5umzjigaJpZM4Xx9xK .
--
Brian J. Haas The Broad Institute http://broadinstitute.org/~bhaas http://broad.mit.edu/~bhaas
ghost
Two questions regarding contamination in RNA-seq data and its handling in a Trinity workflow:
1) Do you, ideally, (1) filter contamination on the level of raw reads or do you filter (2) after Trinity assembly? We are talking about viruses/prokaryotes as the source of contamination. Not eukaryotes. So an assembly should probably not mix contamination with RNA from my species of interest. For this reason, I would intuitively go for filtering after assembly.
2) Which tools did you use / do you recommend? On the raw read level Kraken2 is a good candidate I guess. What would you recommend when filtering on assembly level?
Thx!