rarefaction interpretation

[AB]

Hello,

I have a question regarding rarefaction curves.  I am wondering how my rarefaction curve for Shannon diversity appears to plateau, but curves for individual samples does not? Can some help me interpret these?

Thanks

[Colin Brislawn]

Hello AB,

The curves for observed species behaves differently than the Shannon index because these two things are fundamentally different. While observed_species counts the number of OTUs that appears in a sample, shannon serves as an index or metric of the entropy of the sample. Here is some more information about the Shannon diversity index.

http://www.tiem.utk.edu/~gross/bioed/bealsmodules/shannonDI.html

Rarefaction plots let you answer the question 'Would my result change if I had more reads per sample?' or 'Did I sequence deeply enough?' So in your example, observed_species shows that the lines do not plateau at 1000 reads per sample, leading you to conclude that you need more than 1000 reads per sample to consistently capture all OTUs. (Make some plots that go past 1000, say maybe up to 10,000. Let's find out when observed_species does plateau.) 

I find that Shannon diversity is had to interpret in a biological system, partly because this metric is so different from estimated counts of species. For example, your rarefaction plot shows that Shannon diversity plateaus around 8, but what does that mean intuitively? 

http://www.pisces-conservation.com/sdrhelp/index.html?shannonw.htm

Colin