Three simple questions...

[samja...@gmail.com]

Hi,

I'm a postgraduate student and very unfamiliar with bioinformatics/molecular evolution etc.

I'm afraid I have a couple of stupid questions to ask (I haven't been able to find any concise answers on the internet!):

1- What exactly do 'F3x4' and 'F61' mean? [in the simplest terms possible, please!]
2- How are the M1a/M2a and M7/M8 models generated? And what's the difference between these pairs of models?
3- How do these things come together to tell me whether or not a gene is under positive selection?

Any help will be gratefully received

Thanks in advance!

[Sungmin Lee]

Dear,

For the first question, you can find the answer inside the PAML manual. Simply it is a different way to estimate the codon frequency. 

excerpt: : F1/16 (all codon frequencies are equal), F3X4 (nucleotide frequencies are estimates of the three codon positions and combined), and F61 (codon frequencies are estimated separately for each codon).

-Sung

[James]

Hi SJ,

to answer the second question take a look at:

A Beginners Guide to Estimating the Non-synonymous to Synonymous Rate Ratio of all Protein-Coding Genes in a Genome
Daniel C. Jeffares , Bartłomiej Tomiczek , Victor Sojo , and Mario dos Reis

They have a worked example in which they show how to generate these models and how to read the results. It is easy to follow. :)

James