12.

Predict the secondary structure of the given sequence using Chou & Fasman, GOR & Neural Network
methods. (2)
 
Used the following toold for Secondary Structure Prediction: http://cib.cf.ocha.ac.jp/bitool/MIX/
Graph images have been attached in the directory as CF.png for Chou-Fosman, GOB for Garnier-
Osguthorpe-Robsom and NN_1 and NN_2 for Neural networks. The combined prediction is as attached
below.

 
BLAST Results:
The following protein was chosen after blasting the query against the PDB database.
 

13. BLAST the given sequence against PDB and find a suitable template from Sus scrofa<\em>for
homology modelling. Report the query coverage, percent identity, PDB code, resolution of the data and
the chain ID of the chosen template. (2) 
Select seq X-RAY ANALYSES Sus 305 305 83% 8e- 47.56% 326 5PEP_A
pdb|5PEP|A OF ASPARTIC scrofa 102
PROTEASES. II.
 THREE-
DIMENSIONAL
STRUCTURE OF
THE HEXAGONAL
CRYSTAL FORM OF
PORCINE PEPSIN
AT 2.3 ANGSTROMS
RESOLUTION [Sus
scrofa]
 
The above mentioned protein was chosen. Query Coverage = 83%, Percentage Identity= 47.56%,
PDB code = 5PEP , Resolution= 2.45 , Chain ID= PEPSIN A. The PDB page for 5PEP is as follows:
 

14. Ensure that you fix the template for alternate conformers, anisotropic thermal parameters and non-
standard residues, if any. (1)
 
grep -Ei "^atom|^end|^hetatm" 5pep.pdb | awk '{print $4}'| sort | uniq
 
ALA
ARG
ASN
ASP
CYS
GLN
GLU
GLY
HIS
HOH
ILE
LEU
LYS
MET
PHE
PRO
SER
THR
TRP
TYR
VAL
 
No non standard amino acids present.
Cleaned the file to get rid of anisou and alternate conformers using the following command.
 
awk '((substr($0,17,1)=="A" || substr($0,17,1)==" ")) {print substr($0,1,16), substr($0,18,63)} '
clean_noMSE.pdb|  grep -Ei "atom|end|ter" > cleaned.pdb
 
To get rid of any incomplete residues Discovery Studio was used.
Edit --> Preferences --> Protein utilities --> Clean Protein
Saved as cleaned.pdb.
 
15. Construct 3 homology models for the best aligning parts of the given protein sequence using
MODELLER (5).
 
Opening cleaned.pdb in Chimera:
Tools --> Sequence --> Sequence --> Save as FASTA
 
Opening Clustal Omega using MPI toolkit to generate a Alignment between the 5pep.fasta and the query:
The MSA file was downloaded as alignment.fasta.
 
The MSA file was opened using Discovery Studio. The terminal gaps were removed. The MSA was
saved as alignment.pir file. Appropriate header changes were made.
 
The Model Python script was
used to run the command
after editing the model-
default script.
 
 
mod10.2 model-default.py
 
3 models were generated. The
Ramachnadran Plot is
visualised for the least 3
energy models. This is dne by
opening the log file and
checking for the least
energies.
 
16. Based on energy,
Ramachandran outliers
and/or RMSD with template,
identify the best model (1).
 
The log file was opeend to
find the energies:
 
>> Summary of successfully produced models:
Filename                          molpdf
----------------------------------------
query.B99990001.pdb           2016.10815
query.B99990002.pdb           2020.19836
query.B99990003.pdb           2141.05615
  
From the energies
model1 seems to
be the best. Used
COOT to generate
Ramachandran
plots as well.
 
File --> Open
Coordinates
Validate -->
Ramachandran
Plot
 
Model 1 has 6
outliers, Model 2
has 5 and Model 3
has 8. Since the
energy difference
between Model 1
and 2 is very
small, Model 2 is
chosen as it has
less outliers ans
more in favoured
regions percentage.
Both the Model2 and the Cleaned PDB file were opened in PYMOL. Align -> To molecule was used to
superimpose the 2 structures. The reported RMSD value is 0.159 as seen in the above screenshot.
 
17. Generate the cartoon, molecular surface and electrostatic surface of the constructed model (2)
 

Cartoon:
Mouse Clicks: Show -> As -> Cartoon
 
Molecular Surface:
Mouse Clicks: Show -> As -> Surface
 
 
Electrostatic Surface:
 

Mouse Clicks: Actions -> Generate -> Vacuum Electrostatics --> Electrostatic Potential
 
 
 
 
18. Construct a superposition diagram of the structure with PDB code 2ANL and the best model. (2)
 
Model2 and 2ANL were opened in PYMOL. Action -> Align to -> Molecule
Superposition was done. The RMSD 1.148 is mentioned int he below attached screenshot.