Hi,
I got some microbial sequencing data from eukaryotic 18S. I know that 97% similarity is frequently used to define bacterial OTUs based on 16S but the 18S sequence in eukaryotes is often much less variable.
My data contains mostly fungi and "unclassified Eukarya" and I'm aware that in many cases, if not most, one or two nucleotides of difference in the 18S are enough to determine an isolate from a different species or even genus in fungi.
So, anybody knows if there is any "standard" to define eukaryotic OTUs?
I don't know about an entire eukaryotic OTUs threshold (that's a long shot considering how different are...well, all eukaryotes). Instead, it might be helpful to have a range of examples for fungi? In any case,18S in Fungi is used mostly for phylogenetics, and like you said I suspect that its variation is not great at the species level (if that is what you want your OTUs to represent...). Maybe this paper can be useful:
Schoch, C. L., Seifert, K. a., Huhndorf, S., Robert, V., Spouge, J. L., Levesque, C. A., and Chen, W. (2012). Nuclear ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for Fungi. Proceedings of the National Academy of Sciences of the United States of America, 109(16):6241-6.
If you are referring to phylogenetic analysis, a reliable bootstrap value for maximum parsimony analysis is 70%, while for bayesina analysis is 90%
hi,
standard is 97%. This number is governed by our difficulties dealing with sequencing errors, not based on bacterial evolution specificities. The best of course would 100% for bacteria and euk. However, recently developed tools (oligotyping, http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3864673/; Minimum entropy decomposition, http://www.nature.com/ismej/journal/v9/n4/full/ismej2014195a.html) do not need a cut-off and use entropy to partition your data. I recommend you to take a look at the publications. Hope it helps
Given how conservative 18S could be, I suggest 99%, but since 97% has been widely and blindly used in many cases, it would be insightful to use both as the cutoff. Actually, none of the cutoff is right or wrong. It all depends on what kind of question you want to answer. I think in pyrosequencing based studies, pattern is more important than phylogeny.
Tom mentioned oligotyping and MED, which represent the most sensitive way to reveal the microdiversity of closely related phylo-types. And the developer of oligotyping is a very nice guy. Feel free to contact him anytime.
For classification of you sequences, you might be interested in PR2 database. And they offer taxon-based separated fast files, which would help you to improve your classification. 18S rRNA is widely used but not the best barcode to use as it's super conservative and sometimes identical sequences can be shared among distantly related organisms. If there are more fungal sequences in your 18S dataset, it would be better if you can use ITS as the barcode to study the fungal communities.
- Badges
- Science topic
- Similar topics
- Biological Science
- Biota
- Eukaryota
- Fungi