Has anyone used BBTools package to remove contaminants

[tamu_anand]

I would like to remove mitochondrial, chloroplast and rDNA contaminants from read data using BBTools - I was wondering if anyone has used BBTools to do this and can share the generic pipeline.

Thanks in advance.

[GenoMax]

There is no pipeline needed. You can provide contaminants you want to remove as fasta sequence in a file.

While you could use `bbduk.sh` you may want to use `bbsplit.sh` in this case to bin the reads into ones you don't want (based on the list above) and the rest. You can find bbsplit thread here.

Ask if you need information/run into issues.

[tamu_anand]

There is no pipeline needed. You can provide contaminants you want to remove as fasta sequence in a file.

While you could use `bbduk.sh` you may want to use `bbsplit.sh` in this case to bin the reads into ones you don't want (based on the list above) and the rest. You can find bbsplit thread here.

Ask if you need information/run into issues.

Thanks @GenoMax for pointing me to the bbsplit thread

To elaborate, I would like to remove chloroplast/rDNA/mito contaminants and I was thinking I would do something like

bbmap.sh in=read_1.fq.gz ref=rRNA_Chlor_Mito.fa maxindel=1 minid=0.95 outu=clean_read_1.fq.gz nodisk

I am basing the above command on a post by Brian here: https://www.biostars.org/p/143019/#210890

The strategy here is to use the rRNA+Mito+Chloroplast file and map the reads using bbmap, then collect the unmapped reads (clean_read_1.fq.gz) for my downstream analysis.

I would appreciate any inputs/suggestions to the above bbmap command line options.

Also, Is it better to use bbsplit instead of bbmap?

[tamu_anand]

Hi all

Any inputs/recommendations on the bbmap command line I am using - Post 3

Thanks in advance

[GenoMax]

You can certainly use Brian's recommendation above. If you wish to find out how many reads actually map to those individual references (e.g. rRNA, mito etc) then using bbsplit.sh would be useful since it will give you that statistics.