Energy minimization for protein model refinement

Question

[utpal.bioinfo]

Hi. I'm Beginner in Bioinformatics. Need an opinion from Professional. Recently I am doing my project work on in silico protein model of a protein. I have completed my protein model by modeller 9 v14 and I want to store my model in PMDB . But the PMDB showing the stereochemical error of my model. so, how can I remove this error and if you have any video tutorial link on energy minimization by swiss pdb viewer please give me the link

Answer 1

[João Rodrigues]

Hi,

First of all, if you are a beginner in bioinformatics, I would advise against storing one of your first models on a public database.

Regarding the modelling. I would take a look at your alignment. Did you use a multiple sequence alignment or a pairwise alignment? What happens if you use HHpred for example, would you get the same template suggestions? What about the model they produce? Having small deviations from 'normalcy' is not a problem, even crystal structures have those. Also, with MODELLER, did you optimize the model? With which settings? Did you do a good model selection?

Without further information on the error itself and on the modelling process, it's quite hard to gauge the problem and give you a better answer...

[utpal.bioinfo]

Thanks for your advice.Actually i have understood that i have huge lacking to model a protein by using modeller software. I have selected a model on the basis of lowest DOPE score in MODELLER soft. but i don't know how to optimize the model with proper settings. I am really interested to learn about Modeller soft. with optimization. so, i will be greatful to you if you give me a outline to make a protein model from first to last step by insilico approach. Here is my email adress- [e-mail]. please help me..

[João Rodrigues]

Keeping the discussion public will help others.

The outline is simple: find a template, get an alignment, build the model, refine the model. What did YOU do with this particular model? What is the template identity? How did you get the alignment? What options in modeller did you use to refine it (paste the python code f example)?

[utpal.bioinfo]

The steps that i have used- - find out the template by using Blast with settings the parameters PDB and protein sequence. - Download the template in pdb format with 88% identity but did not complete any alignment. -install the Modeller 9.14 with python 2.3 . - I made my protein model by using the basic modelling method from the modeller website. - I made a 'test' folder which contain templatein pdb format, 5script file in PY format, query sequence in .ALI format. - After making folder i paste the folder in program files(in Bin folder). -after that gave the command in modeller soft and wait for the result.

The python file i used- like as below build_profile.py, compare.py, align2d.py,model-single.py, evaluate_model.py

I don't know how to refine the model. please give some instruction.