2019
DOI: 10.1093/bioinformatics/btz122
|View full text |Cite
|
Sign up to set email alerts
|

Protein model quality assessment using 3D oriented convolutional neural networks

Abstract: Motivation Protein model quality assessment (QA) is a crucial and yet open problem in structural bioinformatics. The current best methods for single-model QA typically combine results from different approaches, each based on different input features constructed by experts in the field. Then, the prediction model is trained using a machine-learning algorithm. Recently, with the development of convolutional neural networks (CNN), the training paradigm has changed. In computer vision, the expert… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
133
0
1

Year Published

2020
2020
2023
2023

Publication Types

Select...
7
1
1

Relationship

1
8

Authors

Journals

citations
Cited by 100 publications
(134 citation statements)
references
References 24 publications
(19 reference statements)
0
133
0
1
Order By: Relevance
“…The authors make use of two 1DCNNs to predict the local and global quality of a decoy. In [40], the authors propose Ornate, a single-model method that applies a deep three dimensional CNN (3DCNN) for model quality estimation. 3DCNN has also been used successfully in [45].…”
Section: Related Workmentioning
confidence: 99%
“…The authors make use of two 1DCNNs to predict the local and global quality of a decoy. In [40], the authors propose Ornate, a single-model method that applies a deep three dimensional CNN (3DCNN) for model quality estimation. 3DCNN has also been used successfully in [45].…”
Section: Related Workmentioning
confidence: 99%
“…They used the long short-term memory (LSTM) to model the dependency tree structure of sentences and achieved a significant improvement than other methods without explicit feature extraction. Other researchers suggested using the models based on the convolutional neural network (CNN) to predict PPI [29]. Hua et al proposed the shortest dependency path based on the CNN (sdpCNN) model to predict PPI [30].…”
Section: Background and Related Woekmentioning
confidence: 99%
“…convolutional neural networks (Derevyanko et al, 2018;Pagès et al, 2019) or graph convolutional neural networks (Baldassarre et al, 2019;Sanyal et al, 2020;Igashov et al, 2020) are used to learn finegrained structure representation, but these QA methods may not generalize well to protein models not extensively refined. One important issue with all these methods is that they do not take full advantage of inter-residue or inter-atom distance information which has greatly improved protein structure prediction recently (Zhu et al, 2018;Xu, 2019;Greener et al, 2019;Senior et al, 2020).…”
Section: Introductionmentioning
confidence: 99%