Hello Mohsen,

I will suggest to use an universal primer (for 16S rRNA gene amplification) which covers the variable regions V1-V8. Subsequent sequence analysis may give an idea about the species. If there is very much similarity among the 16S rRNA gene sequences then the phylogeny analysis using these sequences may be insufficient to confirm the species. In such a case 'multiplex PCR assay' or 'species specific PCR assay' can be followed to confirm the species.

Different target regions will offer better speciation of different taxa.  Ideally, you would look at the entire 16s gene, but that's not really feasible with the MiSeq.

V3-V4 is generally a good target.  Using 2x300 you'll have no trouble assembling the pairs; you can get genus level identification for the vast majority of OTUs and species level identification for a significant portion of them; and it generally works pretty well, experimentally.  However, there are likely to be a handful of OTUs that are interesting or that comprise a significant portion of the microbial population which you wont be able to resolve to the species level with their V3-V4 sequences.  Nevertheless, this a widely used, effective target.

V1-V3 could also be very informative.  You will be able to speciate some of the OTUs that you can't speciate with V3-V4, but there will likely be others that you could speciate with V3-V4 that you can't speciate with V1-V3.  It may be the case that some of the more interesting OTUs are better resolved with V1-V3.  I've found that this target is a little bit harder to work with.  It's a longer target region, so assembling the pairs will be a little bit harder.  In my experience, it also just doesn't work as well experimentally; the library preps don't come out as well and the sequencing data isn't as good, but that may very well be an issue that is limited to our lab.

I don't have any experience with the V5/6-V8 regions.  As far as I know, they're not widely used targets for human oral microbiome profiling with 16s on the MiSeq.

I would recommend V3V4.  It generally works pretty well and allows for reasonably good taxonomic resolution across the board.  V1V3 may allow for somewhat better taxonomic resolution, but it may be more difficult to get it to work well.  If you can try both, I think that would absolutely be worthwhile.

Hi Mohsen,

I totally agree what Alex says. The V3 region is the most informative of the 16S and amplicons from V3-V4 normally results in a good way to determine the microbial communiti inhabiting the environment of your interest. Additionally, be careful to use a DNA extraction kit able to lysate Gram positive bacteria which have a very resistant cell wall. I recomedn to use Mutanolysin+Lysozyme cocktail.

Good luck

Alfonso

Thanks for your comments. It is the first time I am going to do illumina and I do not want to waste money. I have a question:

Why we do sequencing using illumina once it cannot give us identification at the level of species? can  genera or phyla profiling contribute to find significant data to differentiat two different groups of samples?

Hi Mohsen,

The major problem of amplicon sequencing using high-throughput technologies is the read length. Using a configuration of paired-end (i.e. 2x300) you will be able to assembly reads of ~400nt or 500nt as the largest depending of the regions selected. This effective length makes impossible a taxonomic assignation at species  level because your are using only the 25-30% of the genetic information present in the bacterial 16S. To assess identification at species or strain levels you must obtain information of the entire 16S molecule. We are on progress to publish a manuscript  at this concern and I hope this gonna be available in few weeks.

Anyway, data obtained from current approaches can give you a good idea of differential abundance at family and genus level using the OTU-based approaches.

Regards

Hi Mohsen and every one,

I use the following degenerate 27F/519 targeting the V1-V3 region, and they just work fine: 

27F:AGAGTTTGATYMTGGCTCAG 

519R: GWATTACCGCGGCKGCTG

These are supposed to provide maximum coverage of the oral microbiome. The V1-V3 region is the most hypervariable and should thus provide better taxonomic resolution. Species-level classification of the read, however, depends more on the algorithm you use for taxonomic assignment.  Clustering into OTUs and using a Bayesian classifier will assign the majority of reads to the genus level. I don't believe following this approach for  well characterized microbial communities such as the oral microbiome for which well-curated databases of reference 16S rRNA gene sequences (e.g. HOMD) are available is justified. I, in collaboration with Chen Tsute at Forsyth, developed a comprehensive, BLASTN-based taxonomic assignment algorithm for species level classification of NGS reads from oral samples. With this, we managed to  classify around 95% of the reads to the species level. I was just published (link attached). I hope this may be of help to you

Regards

Nezar Al-hebshi

http://www.journaloforalmicrobiology.net/index.php/jom/article/view/28934

Dear Dr. Nazar

Thanks for your answer. Is there a significant difference between V1-V3 and V3-V4 primer in terms of species identification? I can see lots of publications used V3-V4 for MiSeq. I am a bit confuse why people are stil doing sequencing while they are not able to identify at the species level? I guess genera classification cannot give use significant data for using bacteria as biomarkers.

Nezar Al-hebshi Please share the paper and tell me the amplicon size with that primers