Limits...
TAP score: torsion angle propensity normalization applied to local protein structure evaluation.

Tosatto SC, Battistutta R - BMC Bioinformatics (2007)

Bottom Line: While the readily available criteria are sufficient to detect totally wrong structures, establishing the more subtle differences between plausible structures remains more challenging.It is shown to be more accurate than previous methods at estimating the validity of a protein model in terms of commonly used experimental quality parameters on two test sets representing the full PDB database and a subset of obsolete PDB structures.Highly selective TAP thresholds are derived to recognize over 90% of the top experimental structures in the absence of experimental information.

View Article: PubMed Central - HTML - PubMed

Affiliation: Dept. of Biology and CRIBI Biotechnology Centre, University of Padova, Italy. silvio.tosatto@unipd.it

ABSTRACT

Background: Experimentally determined protein structures may contain errors and require validation. Conformational criteria based on the Ramachandran plot are mainly used to distinguish between distorted and adequately refined models. While the readily available criteria are sufficient to detect totally wrong structures, establishing the more subtle differences between plausible structures remains more challenging.

Results: A new criterion, called TAP score, measuring local sequence to structure fitness based on torsion angle propensities normalized against the global minimum and maximum is introduced. It is shown to be more accurate than previous methods at estimating the validity of a protein model in terms of commonly used experimental quality parameters on two test sets representing the full PDB database and a subset of obsolete PDB structures. Highly selective TAP thresholds are derived to recognize over 90% of the top experimental structures in the absence of experimental information. Both a web server and an executable version of the TAP score are available at http://protein.cribi.unipd.it/tap/.

Conclusion: A novel procedure for energy normalization (TAP) has significantly improved the possibility to recognize the best experimental structures. It will allow the user to more reliably isolate problematic structures in the context of automated experimental structure determination.

Show MeSH

Related in: MedlinePlus

Histogram of correctly recognized improved structures on the obsolete PDB set for TAP and ten previous methods. The number of correct predictions by both methods (blue), TAP only (green) and only the second method (red). The total height corresponds to the performance of combining TAP and the other method. Individual performance of each method (except TAP) is the sum of the first two terms (i.e. red and blue).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1878508&req=5

Figure 2: Histogram of correctly recognized improved structures on the obsolete PDB set for TAP and ten previous methods. The number of correct predictions by both methods (blue), TAP only (green) and only the second method (red). The total height corresponds to the performance of combining TAP and the other method. Individual performance of each method (except TAP) is the sum of the first two terms (i.e. red and blue).

Mentions: A set of 494 pairs of obsolete PDB structures and their replacement was analyzed in order to evaluate the capacity of the TAP score to discriminate better models for the same protein. Figure 2 shows the number of times each method correctly assigned a better score to the newer, and therefore more accurate, model in the obsolete PDB set. The results for each individual methods and the combination with TAP are shown. None of the methods discriminates all 494 structures, and the statistical potentials in particular have difficulty recognizing the improved structures, while TAP has one of the highest single recognition rates. Combining methods improves the overall performance, with TAP typically contributing more unique information than the other method. Figure 3 shows a different analysis of the data in terms of separation Z-score, i.e. the normalized difference between the obsolete and replacement structure. Here it is again apparent that the Ramachandran plot based methods outperform the others, with TAP coming a close second with WC_Chi1&2 after WC_Rama. Taken together, the Ramachandran plot based methods (especially TAP and WC_Rama) seem able to qualitatively discriminate improved from obsolete structures.


TAP score: torsion angle propensity normalization applied to local protein structure evaluation.

Tosatto SC, Battistutta R - BMC Bioinformatics (2007)

Histogram of correctly recognized improved structures on the obsolete PDB set for TAP and ten previous methods. The number of correct predictions by both methods (blue), TAP only (green) and only the second method (red). The total height corresponds to the performance of combining TAP and the other method. Individual performance of each method (except TAP) is the sum of the first two terms (i.e. red and blue).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1878508&req=5

Figure 2: Histogram of correctly recognized improved structures on the obsolete PDB set for TAP and ten previous methods. The number of correct predictions by both methods (blue), TAP only (green) and only the second method (red). The total height corresponds to the performance of combining TAP and the other method. Individual performance of each method (except TAP) is the sum of the first two terms (i.e. red and blue).
Mentions: A set of 494 pairs of obsolete PDB structures and their replacement was analyzed in order to evaluate the capacity of the TAP score to discriminate better models for the same protein. Figure 2 shows the number of times each method correctly assigned a better score to the newer, and therefore more accurate, model in the obsolete PDB set. The results for each individual methods and the combination with TAP are shown. None of the methods discriminates all 494 structures, and the statistical potentials in particular have difficulty recognizing the improved structures, while TAP has one of the highest single recognition rates. Combining methods improves the overall performance, with TAP typically contributing more unique information than the other method. Figure 3 shows a different analysis of the data in terms of separation Z-score, i.e. the normalized difference between the obsolete and replacement structure. Here it is again apparent that the Ramachandran plot based methods outperform the others, with TAP coming a close second with WC_Chi1&2 after WC_Rama. Taken together, the Ramachandran plot based methods (especially TAP and WC_Rama) seem able to qualitatively discriminate improved from obsolete structures.

Bottom Line: While the readily available criteria are sufficient to detect totally wrong structures, establishing the more subtle differences between plausible structures remains more challenging.It is shown to be more accurate than previous methods at estimating the validity of a protein model in terms of commonly used experimental quality parameters on two test sets representing the full PDB database and a subset of obsolete PDB structures.Highly selective TAP thresholds are derived to recognize over 90% of the top experimental structures in the absence of experimental information.

View Article: PubMed Central - HTML - PubMed

Affiliation: Dept. of Biology and CRIBI Biotechnology Centre, University of Padova, Italy. silvio.tosatto@unipd.it

ABSTRACT

Background: Experimentally determined protein structures may contain errors and require validation. Conformational criteria based on the Ramachandran plot are mainly used to distinguish between distorted and adequately refined models. While the readily available criteria are sufficient to detect totally wrong structures, establishing the more subtle differences between plausible structures remains more challenging.

Results: A new criterion, called TAP score, measuring local sequence to structure fitness based on torsion angle propensities normalized against the global minimum and maximum is introduced. It is shown to be more accurate than previous methods at estimating the validity of a protein model in terms of commonly used experimental quality parameters on two test sets representing the full PDB database and a subset of obsolete PDB structures. Highly selective TAP thresholds are derived to recognize over 90% of the top experimental structures in the absence of experimental information. Both a web server and an executable version of the TAP score are available at http://protein.cribi.unipd.it/tap/.

Conclusion: A novel procedure for energy normalization (TAP) has significantly improved the possibility to recognize the best experimental structures. It will allow the user to more reliably isolate problematic structures in the context of automated experimental structure determination.

Show MeSH
Related in: MedlinePlus