error: edid 317 / 317 site patterns

[Yihang Zhao]

Dear sir,

After a relatively successful launch of paml I have got error message at the very last steps before the analysis finished

BEBing (dim = 6).  This may take many minutes.
Trying to get 44.0M memory in lfunNSsites_ACD

Calculating f(x_h|w) for 120 w's
          120 / 120 sets.  w for branches:  2.8500 2.8500 lnfX = -339757.5318
Time: 48:09
Calculating the marginal likelihood f(X).
        log(fX) = -124829.687134  S = -124592.510060  -230.840122
Calculating f(w_k|X), posterior for site classes for each site.  Slow!
        did 290 / 317 site patterns  48:16
error: edid 317 / 317 site patterns  48:17(paml)

Could you please tell me what this error is?

  paml version: CODONML in paml version 4.10.8, November 2024

Thank you in advance,

Yihang Zhao

[Sandra AC]

Dear Yihan Zhao,

Thanks for your message! Could you please share your input files and control file so that we can further troubleshoot this issue?

All the best,
Sandy

[Yihang Zhao]

Hi

Please see the attached files for my input files and control file

Best

Yihang

[Yihang Zhao]

Hi

I just want to follow-up on this question and see if there's any update on this issue. Also, I think I forgot to mention this but the error message mentioned in this post is found in the terminal when paml finish its run, not found in the output file. And the paml run did produce an output file that seems normal in all aspects. Hope this information help!

Best

Yihang

On Wednesday, February 19, 2025 at 7:12:06 AM UTC-5 sandr...@gmail.com wrote:

[Sandra AC]

Ηi Yihang,

Thanks for your patience and apologies for the delay, busy times! I have checked your input files and the input tree file does not have a header. You should add the PHYLIP header following the notation "num_taxa  num_trees". If you have 462 taxa, then the PHYLIP header in your input tree file should be "462  1".

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 have formatted your input alignment file so that each sequence is in one line (just in case the format has affected the run, not sure what may have exactly caused to prompt this error on the screen). The following are the commands I have used:

# Read your input alignment and generate a

# tmp FASTA file

count=0

while read -r line

do

count=$(( count + 1))

new_line=$(echo $line )

if [[ $line =~ "_" ]]

then

new_line=$( echo $new_line | sed 's/^/\>/' )

fi

if [[ $count -eq 1 ]]

then

echo $new_line > aln.fasta

else

echo $new_line >> aln.fasta

fi

done < Ga_supertree_467_taxa_DNA_clustalo_backtrans.phy

# Get rid of the PHYLIP header and the second

# blank line and carriage characters

sed -i 's/\r//g' aln.fasta

sed -i '1,2d' aln.fasta

# Give first permissions to all my perl scripts

# in case you have not done so

chmod 775 *pl

# Call my perl script to format the FASTA

# file so that there is one line per sequence

./one_line_fasta.pl aln.fasta

# Remove first FASTA file and rename new

# one-line FASTA file

rm aln.fasta

mv aln_one_line.fa aln.fasta

# Get the number of taxa and length of

# sequence and run my script to convert

# the FASTA file into PHYLIP format

aln_name=`ls *.fasta`

a_noext=$( echo $aln_name | sed 's/\.fasta//' )

num=$( grep '>' $aln_name | wc -l )

len=$( sed -n '2,2p' $aln_name | sed 's/\r//' | sed 's/\n//' | wc --m )

perl FASTAtoPHYL.pl $aln_name $num $len

# Remove files that are no longer required

# to save space

rm aln.fasta log_lenseq.txt

The attached compressed file has your input alignment file, my "one_line_fasta.pl" and "FASTAtoPHYL.pl" perl scripts that you see I use in the code snippet above, and the output files that result from running the code snippet above, which you will need to re-run CODEML: "aln.phy" (the formatted input sequence file) and "tree.tree" (the input tree file with the PHYLIP header).

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

All the best,
Sandy

[Yihang Zhao]

Yes, the issue is fixed! Thank you very much!

Best

Yihang