Limits...
Improvement in low-homology template-based modeling by employing a model evaluation method with focus on topology.

Dai W, Song T, Wang X, Jin X, Deng L, Wu A, Jiang T - PLoS ONE (2014)

Bottom Line: Our novel method focuses on evaluating the topology by using two novel groups of features.These novel features included secondary structure element (SSE) contact information and 3-dimensional topology information.We further showed that the MEFTop could be a generalized method to improve threading programs for low-homology protein targets.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Protein & Peptide Pharmaceuticals, National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China ; University of the Chinese Academy of Sciences, Beijing, China.

ABSTRACT
Many template-based modeling (TBM) methods have been developed over the recent years that allow for protein structure prediction and for the study of structure-function relationships for proteins. One major problem all TBM algorithms face, however, is their unsatisfactory performance when proteins under consideration are low-homology. To improve the performance of TBM methods for such targets, a novel model evaluation method was developed here, and named MEFTop. Our novel method focuses on evaluating the topology by using two novel groups of features. These novel features included secondary structure element (SSE) contact information and 3-dimensional topology information. By combining MEFTop algorithm with FR-t5, a threading program developed by our group, we found that this modified TBM program, which was named FR-t5-M, exhibited significant improvements in predictive abilities for low-homology protein targets. We further showed that the MEFTop could be a generalized method to improve threading programs for low-homology protein targets. The softwares (FR-t5-M and MEFTop) are available to non-commercial users at our website: http://jianglab.ibp.ac.cn/lims/FRt5M/FRt5M.html.

Show MeSH
Comparison of the performance of MEFTop and FR-t5 in the dawn region of SCOP1.75–500 set.(A) The percentage of native-like Top1 models (Top1%) that selected by MEFTop using P-score and FR-t5 using Z-score. The X-axis is the Z-score cutoff and the Y-axis is the Top1%. The performances of Z-score and P-score are shown as white and black columns, respectively. (B) The TM-score of Top1 models selected according to Z-score and P-score for 63 targets with optimal Z-score <5.0. The X-axis and Y-axis of each point represent the TM-scores of Top1 models selected by Z-score and P-score, respectively.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3935967&req=5

pone-0089935-g001: Comparison of the performance of MEFTop and FR-t5 in the dawn region of SCOP1.75–500 set.(A) The percentage of native-like Top1 models (Top1%) that selected by MEFTop using P-score and FR-t5 using Z-score. The X-axis is the Z-score cutoff and the Y-axis is the Top1%. The performances of Z-score and P-score are shown as white and black columns, respectively. (B) The TM-score of Top1 models selected according to Z-score and P-score for 63 targets with optimal Z-score <5.0. The X-axis and Y-axis of each point represent the TM-scores of Top1 models selected by Z-score and P-score, respectively.

Mentions: To evaluate the robustness of MEFTop, a 5-fold cross-validation was carried out on the training set SCOP1.75-Z6. The average and standard deviation of the percentage of native-like Top1 models (Top1%) (see Methods section for details) was 46.84%±2.55%, which indicates stable performance of MEFTop for targets in the dawn region. The performance of MEFTop was further tested on the data set (SCOP1.75–500) which included 110 proteins in dawn region. The Top1% selected with the P-score of MEFTop was compared to that selected with the Z-score of FR-t5. As shown in Figure 1A, we found that the Top1% selected according to the P-score was higher when the best Z-score cutoff of targets was used as 4.0 or 5.0, but somewhat lower when targets had an optimal Z-score less than 6.0. Furthermore, in order to evaluate the models selected according to P-score and Z-score, we compared the TM-score [25] of Top1 models according to two metrics for targets with an optimal Z-score <5.0 on the SCOP1.75–500 set (Figure 1B). Of 63 targets on the testing set, there were 33 Top1 models with better quality selected according to the P-score, while 22 Top1 models with better quality selected according to Z-score. These results indicate that better performance for protein modeling can be achieved for targets in the dawn region using the P-score of the MEFTop method than using the Z-score of the FR-t5 method.


Improvement in low-homology template-based modeling by employing a model evaluation method with focus on topology.

Dai W, Song T, Wang X, Jin X, Deng L, Wu A, Jiang T - PLoS ONE (2014)

Comparison of the performance of MEFTop and FR-t5 in the dawn region of SCOP1.75–500 set.(A) The percentage of native-like Top1 models (Top1%) that selected by MEFTop using P-score and FR-t5 using Z-score. The X-axis is the Z-score cutoff and the Y-axis is the Top1%. The performances of Z-score and P-score are shown as white and black columns, respectively. (B) The TM-score of Top1 models selected according to Z-score and P-score for 63 targets with optimal Z-score <5.0. The X-axis and Y-axis of each point represent the TM-scores of Top1 models selected by Z-score and P-score, respectively.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3935967&req=5

pone-0089935-g001: Comparison of the performance of MEFTop and FR-t5 in the dawn region of SCOP1.75–500 set.(A) The percentage of native-like Top1 models (Top1%) that selected by MEFTop using P-score and FR-t5 using Z-score. The X-axis is the Z-score cutoff and the Y-axis is the Top1%. The performances of Z-score and P-score are shown as white and black columns, respectively. (B) The TM-score of Top1 models selected according to Z-score and P-score for 63 targets with optimal Z-score <5.0. The X-axis and Y-axis of each point represent the TM-scores of Top1 models selected by Z-score and P-score, respectively.
Mentions: To evaluate the robustness of MEFTop, a 5-fold cross-validation was carried out on the training set SCOP1.75-Z6. The average and standard deviation of the percentage of native-like Top1 models (Top1%) (see Methods section for details) was 46.84%±2.55%, which indicates stable performance of MEFTop for targets in the dawn region. The performance of MEFTop was further tested on the data set (SCOP1.75–500) which included 110 proteins in dawn region. The Top1% selected with the P-score of MEFTop was compared to that selected with the Z-score of FR-t5. As shown in Figure 1A, we found that the Top1% selected according to the P-score was higher when the best Z-score cutoff of targets was used as 4.0 or 5.0, but somewhat lower when targets had an optimal Z-score less than 6.0. Furthermore, in order to evaluate the models selected according to P-score and Z-score, we compared the TM-score [25] of Top1 models according to two metrics for targets with an optimal Z-score <5.0 on the SCOP1.75–500 set (Figure 1B). Of 63 targets on the testing set, there were 33 Top1 models with better quality selected according to the P-score, while 22 Top1 models with better quality selected according to Z-score. These results indicate that better performance for protein modeling can be achieved for targets in the dawn region using the P-score of the MEFTop method than using the Z-score of the FR-t5 method.

Bottom Line: Our novel method focuses on evaluating the topology by using two novel groups of features.These novel features included secondary structure element (SSE) contact information and 3-dimensional topology information.We further showed that the MEFTop could be a generalized method to improve threading programs for low-homology protein targets.

View Article: PubMed Central - PubMed

Affiliation: Key Laboratory of Protein & Peptide Pharmaceuticals, National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China ; University of the Chinese Academy of Sciences, Beijing, China.

ABSTRACT
Many template-based modeling (TBM) methods have been developed over the recent years that allow for protein structure prediction and for the study of structure-function relationships for proteins. One major problem all TBM algorithms face, however, is their unsatisfactory performance when proteins under consideration are low-homology. To improve the performance of TBM methods for such targets, a novel model evaluation method was developed here, and named MEFTop. Our novel method focuses on evaluating the topology by using two novel groups of features. These novel features included secondary structure element (SSE) contact information and 3-dimensional topology information. By combining MEFTop algorithm with FR-t5, a threading program developed by our group, we found that this modified TBM program, which was named FR-t5-M, exhibited significant improvements in predictive abilities for low-homology protein targets. We further showed that the MEFTop could be a generalized method to improve threading programs for low-homology protein targets. The softwares (FR-t5-M and MEFTop) are available to non-commercial users at our website: http://jianglab.ibp.ac.cn/lims/FRt5M/FRt5M.html.

Show MeSH