Binder for Pitt PAML/codeML lesson
Two comparisons in tutorial:
(1) Low dN/dS: Compare P. aeruginosa clinical isolate P32_8's wspF gene to homolog in a distantly-related P. aeruginosa strain (PA7)
(2) High dN/dS: Compare a different clinical isolate's wspF gene to PA7
(1) P. aeruginosa P32_8's wspF versus PA7's wspF (PSPA7_1435)
Enter the first direcotry
View the FASTA amino acid file (.faa)
Align the FASTA amino acid file using MUSCLE. THe .fa ouutput is the peptide alignment.
muscle -in P32_8_wspF_vs_PA7.faa -out P32_8_wspF_vs_PA7.fa
Use pal2nal to make a codon alignment from the peptide alignment and nucleotide sequence.
pal2nal.pl P32_8_wspF_vs_PA7.fa P32_8_wspF_vs_PA7.ffn -output fasta > P32_8_wspF_vs_PA7.codonalign.fa
Check that the correct input/output files are in the codeml.ctl file that I already made.
Run codeml using the .ctl file
Examine output and find the dN/dS among other interesting details.
less -S P32_8_wspF_vs_PA7.codemlOutput.txt
The dN/dS value is low (0.0299), suggesting that this gene has evolved under purifying selection in P32_8.
(2) Compare a different P. aeruginosa clinical isolate (P24_18)'s wspF gene to the same homolog in PA7
Open the second directory
cd .. cd P24_18_wspF_vs_PA7
To save time, I have already run codeML for this comparison. Examine the files that are in the directory and find the output file.
ls less -S P24_18_wspF_vs_PA7.codemlOutput.txt
In P24_18, wspF has a higher dN/dS (0.162), but still not likely to indicate that this gene evolved under neutral or positive selection. A closer examination of the peptide and codon alignments revealed that a deletion occurred in P24_18's wspF, leading to erroneous amino acids at the C-terminus of the protein.