Number of reference sequences in NCBI 16S database.

[16srdna]

Dear all,

I am currently looking for different databases for assignment of taxonomy to my sequences from V3-V4 Illumina sequencing.

I found a blastdbcmd command to convert the NCBI 16S database files to fasta format. However, the output of the command has only 18k sequences.

This is surprising to me because I found 90k reference sequences from Greengenes database and 731k reference sequences in SILVA database.

Does the NCBI 16S database actually organize the sequences into certain similarity percentage and thus giving a lower number of reference sequences in the database? or there is something wrong with the command (below) I used to convert the database to fasta format?

blastdbcmd -db 16Smicrobial -out 16S_microbial.fasta -outfmt %f -entry "all"

Thanks a lot!

[cliffbeall]

The NCBI database is only from named, cultured, species so it will be missing sequences for the large diversity of uncultured organisms. Its probably not useful for a high throughput survey experiment, more for blasting a few things where you don't want all the top hits to be "uncultivated bacterium".

[osaebo]

Several tools out there

I am sure Illumina has a 16S database and tools as well, but if you want fast 16S taxonomy ID on your Illumina run, you can try the NGS product from Pathogenomix.com.

From FASTQ to report in 5 minutes.