Run PAML in Parallel

[hayot sepnos]

Hi everyone, I am trying to run PAML in parallel where i have each node on my computer compute a specific model. My question is:

 (1) how can i change the control file so that it only compute a given model (Let's say M8 and not the other 5 models). 

(2) I am also trying to get the BEB analysis to print the BEB for every amino acid in the alignment (not simply the ones with values close to 0.95). How can i edit the control file to see this result?

Below is an example of my control file

eqfile = jonk.fasta  * sequence data filename

outfile = jonk.txt   * main result file name

treefile = jonk.nwk

noisy = 0      * 0,1,2,3,9: how much rubbish on the screen

verbose = 0      * 1:detailed output

runmode = 0     * -2:pairwise

seqtype = 1      * 1:codons

CodonFreq = 2      * 0:equal, 1:F1X4, 2:F3X4, 3:F61

model = 0      

NSsites = 0 1 2 3 4 7 8      *

icode = 0      * 0:universal code

fix_kappa = 1      * 1:kappa fixed, 0:kappa to be estimated

kappa = 1      * initial or fixed kappa

fix_omega = 0      * 1:omega fixed, 0:omega to be estimated

omega = 0.5    * initial omega value

*ndata = 1

[Janet Young]

Hi Hayot,

(1)  you can simply specify one model at a time on the NSsites line, e.g.  for M8:

NSsites = 8      *

Make sure you run each model in a separate directory - if you don't, the output files will overwrite each other

(2) I don't think you can do that by editing the control file, but you can find what you want in the 'rst' output file.  The first half of the 'rst' file gives the NEB results, the second half gives the BEB results. You'll see a space-delimited table, and the number in parentheses is the most likely dN/dS class. The number immediately before that is the BEB posterior probability that each site is in the positively selected class (only relevant if there is support for positive selection according to the likelihood ratio test).

Janet

[hayot sepnos]

Thank You! This was so helpful !