Error in bacterial SILVA taxonomy files

mothur bugs
1

Hi all,
This is my first time with Mothur! I’m analyzing some 16S rRNA samples from an anoxic basin and I need some help.
My samples contain a mix of archaeal and bacterial sequences, and I want to classify and split them using the SILVA database as a reference. I tried classify.seq and remove.seq with the archaeal database and all went fine. I tried to do the same with the bacterial one, but the program gave me an error, telling me that it couldn’t match some sequences in the reference database (silva.gold.ng.fasta) with those in the taxonomic reference (silva.bacteria.silva.tax). I checked both files and found that the identifiers in the taxonomy files are different from those in the fasta file:
Taxonomy: AB000389.1 Bacteria;Proteobacteria;Gammaproteobacteria;Alteromonadales;Pseudoalteromonadaceae;Pseudoalteromonas;
FASTA: 7000004128189528 [sequence]
How can I fix this error?
Best regards

Michael Tangherlini

2

Has anybody any solution to this issue?

Michael Tangherlini

3

I believe the silva.gold.ng.fasta file is the non-chimeric database used for chimera checking. If I remember right, the silva.bacteria.*.tax files all refer to the silva.bacteria.fasta alignment database. You can run it through the classify.seqs command as is, because it ignores gaps anyway, or un-align it to save space (degap.seqs).