Sequence-based mutation analysis TSD Journal

From Bioinformatikpedia

Back to results.

To improve on the readability of the journal, only the basic steps and program calls are outlined here, while the full source code of self-written scripts is linked to like this.

Substitution matrices

Quantiles can be easily calculated in R, using <source lang="bash"> quantile(m) #Where m is a matrix </source>

To create the PSSM, PSI-Blast was called as follows, using the parameters from Task 2, if not already given by the Task description (number of iterations): <source lang="bash"> wget PAT=`pwd` blastpgp -m 8 -Q $PAT/blastpgp_pssm -d /mnt/project/pracstrucfunc12/data/big/big -i $PAT/P06865.fasta -v 3800 -b 3800 -j 5 > $PAT/lastpgp.out </source>


To create the MSA, a psiblast query with P06865 was performed, using the NCBI's webserver with default values (2 iterations, database 'nr'). The resulting sequences were downloaded and aligned using the EBI's clustalw webserver. The resulting MSA was downloaded and converted to FASTA format, using Jalview. Finally the PSSM can be created with the following commandline: <source lang="bash"> psiblast -subject ./query.fasta -in_msa clustalw.fasta -out_ascii_pssm pssm.txt </source>

Structural visualisation

Pymol, see Mutagenesis.



PolyPhen 2 predictions were done using the webserver's batch mode. All settings were left at default values. Here are batch file and query sequence.


SIFT predictions were performed using the webserver at default values (database: UniRef90 2011 Apr). Here is the input mutation file.


snap2 -i P06865.fasta -o snap.out -m all --tolerate

or respectively

snapfun -i P06865.fasta -o snap.out -m mutation file

egrep "(M1|L39|C58|L127|R170|R178|S210|D258|L451|E482)[A-Z]+" snap.out > filteredsnap.out
egrep "M1V|L39R|C58Y|L127R|R170W|R178H|S210F|D258H|L451V|E482K" snap.out > onlyRealSNPsnap.out