Difference between revisions of "Homology Modeling of ARS A"

From Bioinformatikpedia
Line 27: Line 27:
   
 
Our potential tmeplates, identified by the database searches contain all homologs with known structure, regarding to HSSP.
 
Our potential tmeplates, identified by the database searches contain all homologs with known structure, regarding to HSSP.
=== Pairwise alignments ===
 
   
  +
In order to predict the structure, modeller needs pairwise sequence alignments in PIR format. We wrote a python script (''align_all.py'') - using the example python script ''aln_append_model.py'' - to achieve this. Then we wrote one Python script (using the example script ''model-default.py'') per protein to build the model, using the alignments and the pdb structures.
The programs used for TASK 2 are heuristics, which produce local alignments. If we call modeller with only one template structure, we need to globally align the seqeunces.
 
  +
We executed the following commands:
To create pairwise alignments with the Needleman-Wunsch algorithm, we installed the European Molecular Biology Open Software (EMBOSS).
 
 
#<code>
 
#sudo apt-get install emboss
 
#</code>
 
 
Then, we executed the ''needle'' program with the following commands:
 
   
 
<code>
 
<code>
  +
/apps/modeller9.9/bin/mod9.9 align_all.py
needle -gapopen 10 -gapextend 0.5 -outfile 1p49.needle 1AUK.fasta.txt 1P49.fasta.txt <br>
 
  +
/apps/modeller9.9/bin/mod9.9 predict1p49.py
needle -gapopen 10 -gapextend 0.5 -outfile 1fsu.needle 1AUK.fasta.txt 1FSU.fasta.txt <br>
 
  +
/apps/modeller9.9/bin/mod9.9 predict2vqr.py
needle -gapopen 10 -gapextend 0.5 -outfile 2vqr.needle 1AUK.fasta.txt 2VQR.fasta.txt <br>
 
  +
/apps/modeller9.9/bin/mod9.9 predict1fsu.py
needle -gapopen 10 -gapextend 0.5 -outfile 3ed4.needle 1AUK.fasta.txt 3ED4.fasta.txt <br>
 
  +
/apps/modeller9.9/bin/mod9.9 predict3ed4.py
 
</code>
 
</code>
   
A tutorial on the usage of ''needle'' can be found [http://helixweb.nih.gov/emboss/html/needle.html here].
 
 
We executed Modeller with the following commands:
 
 
<code>
 
/apps/modeller9.9/bin/mod9.9 aln_append_model.py
 
/apps/modeller9.9/bin/mod9.9 model-default.py
 
</code>
 
   
We modified the paths and filenames in the scripts ''aln_append_model.py'' and ''model-default.py'' such that it matched our proteins of interest.
+
We modified the paths and filenames in the scripts such that it matched our proteins of interest.

Revision as of 15:36, 7 June 2011

HHpred

We used the webserver and

Modeller

Proteins used as templates

We identified the following proteins (see Alignment TASK) as potential targets for homology modeling:used the following

SeqIdentifier Seq Identity (from TASK 2) source Protein function True homolog (HSSP) Seq Identity (pairw. ali.)
1P49 39.0% Homo Sapiens Steryl-Sulfatase yes 31.9%
1FSU 28.0% Homo Sapiens Arylsulfatase B yes 26.5%
2VQR 20.0% Rhizobium leguminosarum Sulfatase no 20.3%
3ED4 32.0% Escherichia coli Arylsulfatase yes 27.7%

Our potential tmeplates, identified by the database searches contain all homologs with known structure, regarding to HSSP.

In order to predict the structure, modeller needs pairwise sequence alignments in PIR format. We wrote a python script (align_all.py) - using the example python script aln_append_model.py - to achieve this. Then we wrote one Python script (using the example script model-default.py) per protein to build the model, using the alignments and the pdb structures. We executed the following commands:

/apps/modeller9.9/bin/mod9.9 align_all.py /apps/modeller9.9/bin/mod9.9 predict1p49.py /apps/modeller9.9/bin/mod9.9 predict2vqr.py /apps/modeller9.9/bin/mod9.9 predict1fsu.py /apps/modeller9.9/bin/mod9.9 predict3ed4.py


We modified the paths and filenames in the scripts such that it matched our proteins of interest.