Classification up to species level

[jitendra keshri]

Hi,
I wonder if we can classify the 16s rRNA data up to species level. I was going through the EzBioCloud 16S database and there was suggestions that we can classify 16S ngs data up to species level. They have curated formats are compatible with QIIME and MOTHUR pipelines. 

https://www.ezbiocloud.net/resources/16s_download

I tried my sequences with their database, the results up to genus level is more or less similar to Silva, but it further classifies sequences to several species. I have doubt on validity of the results, whether we can present data of species level classification based on only 460 bp of 16S?

Bet Regards,

Jitendra

[Colin Brislawn]

Hello Jitendra,

This is a tricky question. If we zoom about about, we can rephrase this question as "based on the information in my amplicon, how much resolution do I have to detect differences between microbes?" Some species will have very different 16s genes, so you can totally classify them to the species level. Other species will have the exact same DNA in the region of the 16s gene you sequenced, so they all look the same based only on their 16s amplicon. This can be different for different microbes, so you may have family level classifications and genus level classifications in the exact same data set!

This also depends on the referance database you use. If your 16s amplicons are different, but you don't have exact matches in your database, the database may say 'this microbe is inside this family but I'm not sure which species it is."

I hope this helps. Taxonomy classification is hard to do well, and these are some of the limitations you will face working with amplicon marker genes.

Colin

[jitendra keshri]

Thanks for your informative reply!

Regards,

Jitendra