I need to provide a colleague with the maximum % nucleotide and amino acid diversity in my set of sequences (overall and within subpopulations). everywhere I look I see mean measures! and I feel goofy for not being able to find this!

thanks in advance,
Rachel

Hi Rachel,

F_ST (a standard under Compartmentalization) is probably your best bet. This will compute mean pairwise diversity (no phylogenetic correction).

Sergei

thanks, Sergei. I'm after maximum measures of diversity, not mean, in order to compare to past published data. at this point I'm considering just a crude pairwise matrix, but the number of taxa involved makes me cringe with sore eyes just to imagine it. I'll keep looking.

Rachel

Hi Rachel,

If your sequences are aligned, I have a really fast command line utility that can compute pairwise TN93 distances on 50,000 sequences in <5 minutes and output the maximum as well as all distances above a fixed threshold. Let me know if you are interested.

Sergei

that would be super, Sergei! thank you!

Rachel

Hi Rachel,

Here you go. Do the usual Linux source install:

./configure
make
make install

You should now have the TN93dist binary which takes a .fasta file and a number of arguments (hopefully the help line is somewhat meaningful) and runs the pairwise distance calculation. Maximum distance is spooled to the console; only those distances exceeding the threshold you put on the command line are output to a file.

Sergei

---

Hi Rachel,

A more documented version:

Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login

Sergei