Difference between revisions of "Task6 Hemochromatosis Protocol"
Bernhoferm (talk | contribs) |
Bernhoferm (talk | contribs) (→Amino acid features) |
||
Line 1: | Line 1: | ||
== Amino acid features == |
== Amino acid features == |
||
+ | |||
+ | The values were taken from [http://en.wikipedia.org/wiki/Amino_acid Wiki: Amino Acids] and [en.wikipedia.org/wiki/Proteinogenic_amino_acid Wiki: Proteinogenic Amino Acids]. (TODO: find "better" sources). |
||
<br style="clear:both;"> |
<br style="clear:both;"> |
Revision as of 12:03, 17 June 2012
Amino acid features
The values were taken from Wiki: Amino Acids and [en.wikipedia.org/wiki/Proteinogenic_amino_acid Wiki: Proteinogenic Amino Acids]. (TODO: find "better" sources).
BLOSUM62/PAM1/PAM250
PSSM-Matrix and homolog search
The PSSM-Matrix and homolog search was done by using PSI-Blast with the command:
blastpgp -i Q30201.fasta -j 5 -Q outMatrix.txt -o output.txt -d /mnt/project/pracstrucfunc12/data/big/big
To retrieve all homologous proteins, we retained only the results of the fifth iteration and used the following script to extract them:
<source lang="perl">
- !usr/bin/perl
use strict;
- use warnings;
- use autodie;
my $file= shift;
open (FILE, "<$file"); while (<FILE>){ my $currentLine=$_; chomp $currentLine;
if ($currentLine=~m/^..\|(.+)\|.*\d\d\d.*/gi){ print "$1\n"; } } </source>
Because we only wanted mammalian protein sequences, we used Unipot to retrieve the corresponding information to these protein by getting the Flat-formatted file.
To get only mammalian proteins we used the following script on the flat file:
<source lang="perl">
- !usr/bin/perl
use strict;
- use warnings;
- use autodie;
my $file= shift;
my @currentIDs=""; open (FILE, "<$file"); while (<FILE>){ my $currentLine=$_; chomp $currentLine;
if ($currentLine=~m/^AC\s(.*)/gi){ $currentLine=$1;
- print "$currentLine 1\n";
$currentLine=~s/\s//gi;
push(@currentIDs, split(/;/,$currentLine));
#print "$1\n";
} elsif ($currentLine=~m/^SQ\s(.*)/gi){
@currentIDs="";
} elsif ($currentLine=~m/^OC.*Mammalia/gi){
# print "Mammalian found, bitches!\n";
foreach my $entry (@currentIDs){
if ($entry ne ""){
print "$entry\n";
}
}
}
}
</source>
This way we got all mammalian protein IDs that are homologs to our protein and retrieved their sequences through Uniprot.
As some proteins have multiple IDs noted in the flat file we retained our PSIBlast found IDs by using the
uniq -d
command.
After adding our own sequence, we used these sequences in ClustalW and Muscle to retrieve the MSAs.
clustalw -align -infile=Mammals.fasta -outfile=MammalsClustalw.fasta muscle -in Mammals.fasta -out MuscleMSAMammals.fasta