I have a couple of questions. Previously, I have used both SSU and LSU SILVA databases with our data using the old “silva2ncbi.map” mapping file and got slightly different results for each. I noticed that an updated mapping file is available for the latest SSU database, but there is no mapping file for the LSU database. My questions are, firstly, is it worth downloading the latest LSU database and, if so, what mapping file can you use with it? Is the old mapping file correct to use, or should you use the one supplied for SSU, or is there another one available from somewhere else? Secondly, for SSU I have downloaded the non-redundant file. Full alignment, “standard” and truncated versions are available. Which is the preferable database file to download? Thanks for any assistance.
Dear David, sorry for the late reply.
My understanding is the SSU is usually used for taxonomic analysis and so that is why I only provided a SSU mapping file for download here: SSURef_Nr99_132_tax_silva_to_NCBI_synonyms.map.gz.
However, I have just generated a mapping file for LSU sequences and will put it here:
LSURef_132_tax_silva_to_NCBI_synonyms.map.gz.
BTW, after generating the LSU synonyms file I also regenerated the SSU file, using a more recent version of the NCBI taxonomy, so probably best to re-download that file, too, so that both files refer to the same recent version of the NCBI taxonomy.
I believe that it is best to use the non-redundant SSU database:
SILVA_132_SSURef_Nr99_tax_silva.fasta.gz.
I am not sure about the LSU, I would probably use:
SILVA_132_LSURef_tax_silva.fasta.gz.
These are the two files that I used to create the two synonyms files.