error: edid 158 / 158 patterns and error: edid 184 / 184 patterns

[Gabriel Portwood]

Hello! I am studying the evolutionary origins and ancestral reconstruction of the Transthyretin/HIUase protein family and I have been using PAML 4.10.9 for my analyses. I am studying this protein superfamily that went through a gene duplication event and I was trying to run codeml analyses for the whole tree; however, I have been running into some problems and my supervisor told me to divide my whole tree analysis for now to both a Transthyretin subtree and a HIUase subtree and then run the codeml program for 3x4 and FMutSel, but I am running into the same problems. The following messages are show when I try to run codeml:

BEBing (dim = 4).  This may take several minutes.
Calculating f(x_h|w): 10 categories 20 w sets.
Calculating f(X), the marginal likelihood.
        log(fX) = -8597.920495  S = -8404.984277  -188.955221
Calculating f(w|X), posterior probabilities of site classes.

Error: edid 158 / 158 patterns

This is happening for both subtrees and the whole tree analyses (just different numbers and values showing up, but the same error)

My supervisor and I don't know how to fix those issues and how to go through our analyses again.

I am sending the files used for the analyses. 

Best regards,

Gabe ;)

[Sandra AC]

Hi Gabe,

Thanks for your message! 

The message "did X/X site patterns" lets you know about how many site patterns have been parsed when reading a specific locus in your input sequence file. If you have checked the output file and the run did indeed complete, perhaps the last messages were not printed on the screen (did you use your laptop or did you run the job on a HPC?). I may be wrong, but this issue looks somewhat similar to the one discussed in this google group earlier this year (see thread of messages via this link). You may also check a somewhat related closed issue on the PAML GitHub repository. I have checked your input files and it looks there may be a formatting issue with your input tree files. Firstly, you should add the PHYLIP header following the notation "num_taxa  num_trees". E.g.: if you have 28 taxa, then the PHYLIP header in your input tree file should be "28 1". In addition, branch lengths and another labels should also be removed. You can check the PAML Wiki for more details on data formatting via this link :)

Could you please try to re-run CODEML with the formatted input tree files and see if you do not get this error again? Please let us know if formatting has fixed the issue!

All the best,
Sandy