MetaMQAP: A Meta-Server for Protein Model Quality Assessment
Author Information
Author(s): Pawlowski Marcin, Gajda Michal J, Matlak Ryszard, Bujnicki Janusz M
Primary Institution: International Institute of Molecular and Cell Biology, Warsaw, Poland
Hypothesis
The study aims to develop a meta-predictor that improves the accuracy of protein model quality assessment by combining results from multiple existing methods.
Conclusion
MetaMQAP significantly improves the prediction of local model accuracy compared to existing methods, making it a valuable tool for researchers.
Supporting Evidence
- MetaMQAP shows an impressive correlation coefficient of 0.7 with true deviations from native structures.
- The global MetaMQAP score is correlated with model GDT_TS on the level of 0.89.
- MetaMQAP outperformed all methods capable of evaluating just single models.
Takeaway
This study created a new tool that helps scientists check how good their protein models are by using information from several other methods.
Methodology
The study tested eight Model Quality Assessment Programs (MQAPs) on 8251 models and developed a new meta-predictor using multivariate regression.
Potential Biases
The method may be biased towards trivial features that correlate with residue depth and accessibility.
Limitations
The accuracy of MetaMQAP may decrease for models with significant missing residues.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website