Evaluation Of Protein Model Based On Homology Modelling
4
4
Entering edit mode
8.4 years ago

Hi. Im Beginner in Bioinformatics. Need an opinion from Professional. I was about to model a 3D structure of hypothetical protein for my MSc study. I have succesfully model a 3D structure by using SWISS-MODEL and go through some refinement and validation protocal such as Ramachandran plot, energy minimization by Using SAVES, GROMACS, etc. Actually, I would like to Publish a paper for MSc Study . The things that was bear in my mind is;

1)To publish any homology based protein models, what kind of structural assessment needs to be done? OR what are the minimum requirements to publish homology based protein models?

2) what is good software for Calculating RMSD (Root Mean Square Deviation)?. Do I need to calculate this ?

Thanks (Sorry for my poor english).

homology modeling • 5.9k views
ADD COMMENT
4
Entering edit mode
8.4 years ago
AsoInfo ▴ 300

Hi Rashid,

There are many validations which you can perform for Predicted Model. As far as I remember, SWISS-PDB already provide you with some basic information like Z-score or Energy plot (I have used it a long time ago so don't recall much). I am listing some of the softwares / tools for the structure validation:

1) PROSA - Z-score and Local Energy

2) Modeller - DOPE score (Compare the energy between the template and target protein).

3) MOLPROBITY - Ramachandran Plot and other validating tools.

4) PyMOL - RMSD (as mentioned by Vijay)

5) SWISS-PDB Viewer - RMSD / Energy Minimization / Superimpose / Ramachandran Plot

Hope it fulfill your requirements....

ADD COMMENT
1
Entering edit mode

Hi asma,

How to determine whether RMSD value good or not? It is low value is better than High one?.. Im a little bit confius for that things . Thanks for ur time..

ADD REPLY
1
Entering edit mode

Lesser the RMSD, more similar is your structure to the template, as RMSD computes the distance between the atoms.

ADD REPLY
1
Entering edit mode

Vijay had explained it well :)

ADD REPLY
0
Entering edit mode

Ok noted Thank a million :)

ADD REPLY
3
Entering edit mode
8.4 years ago
vijay ★ 1.6k

Hi Rashid,

I prefer going with pymol for calculating your RMSD, meanwhile you an use DALI server to calculate RMSD as well. And answering your first question, I would say you should go through the papers that have been published earlier on the lines of homology modelling, to get an idea of what kind of data are expected for publication.

Have provided the link to a sample paper.

http://peds.oxfordjournals.org/content/14/11/875.full.pdf+html

http://peds.oxfordjournals.org/content/14/11/881.full.pdf+html

Hope they are useful to you.

ADD COMMENT
3
Entering edit mode
8.4 years ago
João Rodrigues ★ 2.5k

The best validation you can get is actual biological information that is explained by your model. Quantitative values such as those you are looking for just tell you that the method you used to create the model is good and that you did not do anything outrageous. Nevertheless, you can still have an awful model with good parameters..

As for RMSD, what RMSD are you calculating? I would recommend Profit as a good software that is solely used for fitting and RMS calculation.

ADD COMMENT
1
Entering edit mode

Dear Joao Rodrigues,

actually, Im trying to find my way to calculate RMSD between my target sequence and template from PDB. My target sequence from hypothethical protein with unknown stucture and functions as well. I have choose template from PDB with the highest similarities and good e-value. what are recommended software or tools that can be used to calculate RMSD between unknown protein without structure with template from PDB that revealved exact structure? most of tools that i found only provided the comparison of two structure between the Protein from PDB. Thats not exactly what I want.

Thank a lot...correct me if im wrong

ADD REPLY
1
Entering edit mode

I already pointed you to Profit.

However, it does not make much sense to compare you model and your template... your model is BASED on the template and is therefore bound to look the same.

As I said, look for biological data.. are you modelling an enzyme? Look at the catalytic site. Are you looking at a binding pocket, look at that one. Etc.. these numbers will not give you anything else than the quality of the model from a purely "structural" point of view. No biology there.

ADD REPLY
1
Entering edit mode

Yup. Modelling an enzyme from hypothetical protein with unknown function. I have looking at catalytic site, the result look promising. Ok i will try looking at binding pocket analysis and others. Thanks you sir....:)

ADD REPLY
0
Entering edit mode
8.4 years ago

Ok.. Thanks a lot.. I will go through..Thanks.

ADD COMMENT

Login before adding your answer.

Traffic: 1177 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6