Predicting residue contacts using pragmatic correlated mutations method: reducing the false positives
2006

Predicting Residue Contacts Using Correlated Mutations

Sample size: 65 publication Evidence: moderate

Author Information

Author(s): Kundrotas Petras J, Alexov Emil G

Primary Institution: Clemson University

Hypothesis

Can the accuracy of predicting residue contacts in proteins be improved by applying selection rules to the correlated mutations method?

Conclusion

The inclusion of selection rules improved the true positive ratio of predictions by 30%.

Supporting Evidence

  • The true positive ratio improved from 0.08 to 0.14 with the inclusion of filters.
  • The methodology was benchmarked against 65 high-resolution structures.
  • The filters were based on biophysical properties of amino acids.

Takeaway

This study found a way to better predict which parts of proteins are close together by using special rules, making the predictions more accurate.

Methodology

The study optimized a correlated mutations method with filters based on biophysical properties and benchmarked it against high-resolution crystal structures.

Limitations

The method may not perform well on longer protein sequences, particularly those over 600 amino acids.

Digital Object Identifier (DOI)

10.1186/1471-2105-7-503

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication