2008
DOI: 10.1186/1471-2105-9-66
|View full text |Cite
|
Sign up to set email alerts
|

Validation of protein models by a neural network approach

Abstract: BackgroundThe development and improvement of reliable computational methods designed to evaluate the quality of protein models is relevant in the context of protein structure refinement, which has been recently identified as one of the bottlenecks limiting the quality and usefulness of protein structure prediction.ResultsIn this contribution, we present a computational method (Artificial Intelligence Decoys Evaluator: AIDE) which is able to consistently discriminate between correct and incorrect protein models… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
31
0

Year Published

2009
2009
2014
2014

Publication Types

Select...
8

Relationship

4
4

Authors

Journals

citations
Cited by 32 publications
(31 citation statements)
references
References 49 publications
0
31
0
Order By: Relevance
“…cgi). 20 The FOX-4 and FOX-4(DGNS) models both have an overall quality considered excellent, with predicted TM scores (which calculate the topological similarity of two protein structures) of 0.74, and respective predicted root mean square deviations (RMSD) of 3.34 and 2.88 Å .…”
Section: Modelling Studiesmentioning
confidence: 99%
“…cgi). 20 The FOX-4 and FOX-4(DGNS) models both have an overall quality considered excellent, with predicted TM scores (which calculate the topological similarity of two protein structures) of 0.74, and respective predicted root mean square deviations (RMSD) of 3.34 and 2.88 Å .…”
Section: Modelling Studiesmentioning
confidence: 99%
“…2). The protein model quality evaluation scores of the SWISS-MODEL and I-TASSER models, based on root mean square deviation (RMSD), TM, LG score, and MaxSub, using ProQ (27) and AIDE (28), pinpointed the best-ranked model generated by I-TASSER as the most suitable candidate to reconstruct the homodimer. Indeed, the I-TASSER model showed a predicted RMSD of 0.69 Å instead of 2.35 Å, a predicted TM score of 0.69 instead of 0.62, a predicted LG score of 4.161 instead of 2.820, and a predicted MaxSub score of 0.396 instead of 0.217 if compared to the model obtained by SWISS-MODEL.…”
Section: )-By Internalizationmentioning
confidence: 99%
“…The top-scored models generated were then ranked and validated by the protein model quality predictors ProQ (27) and AIDE (28).…”
Section: Mccb17mentioning
confidence: 99%
“…In this context, we are presently developing a computational tool in which several structural and physical parameters that can be computed on a protein structure are weighted by a neural network, with the aim of obtaining an empirical energy function suited to discriminate among correct and incorrect protein models [78].…”
Section: Towards the Combined Use Of Molecular Dynamics And Structuramentioning
confidence: 99%