Protein structure validation

Question

[shuvo.bgenstu]

Using I-TASSER, 3D structure of a protein was predicted. The validation of the structure was checked in PROCHECK and only 61% residues were found in most favoured region, 4% were in disallowed region. Is the predicted model still acceptable? Noted that, In SAVES v6.0 - ERRAT quality factor was 80,

Answer 1

[Mensur Dlakic]

Is the predicted model still acceptable?

It depends on what exactly you mean by this, or ultimately for what purpose you want to use the model. Let me put it this way: it is definitely acceptable to show the model at a lab meeting, but it may not be good enough for docking studies. It may be good enough as an overall fold assignment, but not good enough as a molecular replacement solution.

If you are asking whether the model is correct, it would seem unlikely, but it is difficult to tell with certainty because PROCECK displays only stereochemical statistics. I didn't know what you meant by SAVES quality factor until I looked it up - it seems that is a successor to Verify3D. Well, did Verify3D give you a PASS score? You may want to check other servers that do more extensive energy evaluations and stereochemistry of predicted models.

• https://prosa.services.came.sbg.ac.at/prosa.php
• http://www.reading.ac.uk/bioinf/ModFOLD/
• https://swissmodel.expasy.org/qmean/
• http://molprobity.biochem.duke.edu/

I think the functionality of the last two servers can be accessed together using this server:

• https://swissmodel.expasy.org/assess

Answer 2

[shuvo.bgenstu]

Verify 3D and other mentioned tools showed that the 3D structure was not valid. If the predicted structure is not valid then, is it meaningful to include the structure in a research article where different bioinformatics analysis (nsSNPs prediction) has been performed. Thank you so much for your response.

[Mensur Dlakic]

Not sure why you are asking a question for which you already know the answer. What would be the purpose of showing an incorrect model?