blasting genome contigs against local SILVA 16S RNA database

Question

[robjohn70000]

Hi All,

I'm new to bioinformatics and interested in determining the bacteria strains in a collection of bacteria genome contigs. I wonder if is logical and appropriate to create a local SILVA 16S RNA database and then blast (using blastn) the genome contigs against the 16S RNA database. My main concern is that, will it be logical to blast DNA sequences against RNA sequences in the 16S RNA database.

I welcome some advice. Thanks

Answer 1

[Istvan Albert]

if your RNA sequences are represented in the DNA alphabet and then there shouldn't be any problem - that's standard protocol

[robjohn70000]

Thanks @Istvan. I noticed that the RNA sequences have 'U's in place of 'T's.

[Istvan Albert]

NCBI provides taxonomically annotated 16S databases that can be used directly.

Answer 2

[marina.v.yurieva]

If you haven't done this type of analysis before you might look into developed pipelines: mothur and QIIME. For standard 16S analysis blasting is not very efficient.

[Istvan Albert]

Indeed, for 454 type of data it may work, but for large number of reads such as from MiSeq the process requires a lot of resource

[robjohn70000]

Thank you all very much for your contributions and advice. I will be looking at the the suggested tools.