SILVA mapping file

User Questions
1

Hi, I have a question regarding the SILVA’s mapping files. I read on SILVA’s web page that they provide accession to taxonomy mapping files MEGAN compatible (see, for example, the following file at line 78: https://www.arb-silva.de/fileadmin/silva_databases/release_128/Exports/README.txt ). What’s the difference between SILVA’s tax_slv_ssu_128.acc_taxid file and the SSURef_NR99_128_tax_silva_to_NCBI_synonyms.map.gz you provide? They both consist on a column with the SILVA accessions, and another one that I supposed it would be the ncbi taxonomy ids, but these integers don’t match between them, so the SILVA accessions are mapping different things.
Thanks,
Cecilia

2

Dear Cecilia

SILVA provides two sets of three files, one for SSU and the other for LSU sequences.
These can be used to view SILVA assignments to the SILVA taxonomy:

tax_TAXNAME_[ls]su_VERSION.{map,tre}
. . . . . . . . . . .
SILVA taxonomy in the Newick tree format and the corespoding numerical id to
taxonomic path mapping file (MEGAN compatible).

tax_TAXNAME_[ls]su_VERSION.acc_taxid
. . . . . . . . . . .
Mapping of ‘SILVA’ sequence IDs (..)
used in FASTA files to the numeric SILVA taxid (MEGAN compatible).

For example, if you have 16S sequences that you have aligned against the SILVA database and you want to view the analysis using the SILVA SSU taxonomy, then you must do two things:

  1. Supply the two files tax_TAXNAME_ssu_VERSION.tre and tax_TAXNAME_ssu_VERSION.map to MEGAN as “alternative taxonomies” using the appropriate menu item in Edit->Preferences.
  2. Use the file tax_TAXNAME_ssu_VERSION.acc_taxid as a synonyms file when importing your alignments into MEGAN.

If you have aligned your sequences against the SILVA database and want to assign the sequences to the NCBI taxonomy, then use must use the mapping file provided on the MEGAN webpage.