Identifying "contaminants"?

Hi
I want to identify and know the abundance of the sequences removed as “contaminants”. I removed Chloroplast, Mithochondria, Eukaryota and unclassified, and I’d like to know the taxonomic affiliation of the Mitochondria, Chloroplast and Eukaryota to get a rough idea of the abundance and kind of those detected lineages in the sediment samples.
I know that using get.seqs with the accnos from remove.lineage and the fasta file, I can get the unique sequences, but how do I have the tracking of the abundances of those unique sequences? how do I use the count file together with the accnos and fasta to finally get a shared file and identify the “contaminant” OTUs? or at least identify the sequences in the accnos file but knowing the abundance of those sequences per sample?
Thank you!

Hi,

I am sure that there is a more elegant solution, but why don’t you just re-run the remove.lineage command again, only instead of filtering out the “contaminants” you filter out the Bacteria / Archaea? This way you end up with all the output files that you might want for your “contaminants”. Just remember to rename the output files from your original run so that you do not overwrite them.

Good luck,

René

Rene’s approach is technically good, but don’t expect much info from the chloroplast and mitochondria especially. v4 is too short to get much of a plant/animal signal from their chloroplast or mitochondria. Depending on your primers, you may not want to trust the abundance either because most of the v4 primers have been designed to not hit chloroplasts and mitochondria.

Thank you both! We thought of that way of using remove.lineage, but I wanted to know if there was a way to use the accnos file to get the sequences and their abundance and group distribution and maybe even to build OTUs (also to learn how to do that :wink: ). I know that especially abundances is sth I cannot trust, only the detection of some high lineages among plant/animal, to have an idea, for our record let’s say.
We’ll try what you recommended!

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.