2021
DOI: 10.1126/science.abj8754
|View full text |Cite
|
Sign up to set email alerts
|

Accurate prediction of protein structures and interactions using a three-track neural network

Abstract: DeepMind presented remarkably accurate predictions at the recent CASP14 protein structure prediction assessment conference. We explored network architectures incorporating related ideas and obtained the best performance with a three-track network in which information at the 1D sequence level, the 2D distance map level, and the 3D coordinate level is successively transformed and integrated. The three-track network produces structure predictions with accuracies approaching those of DeepMind in CASP14, enables th… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

18
2,734
0
1

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
5

Relationship

2
8

Authors

Journals

citations
Cited by 3,607 publications
(3,096 citation statements)
references
References 76 publications
18
2,734
0
1
Order By: Relevance
“…23a ). Of note, our MymK model shares high similarity with that predicted by AlphaFold2 (AF-A6NI61-F1-model_v1) 39 and RoseTTAFold 40 , two other neural network-based methods.…”
Section: Resultsmentioning
confidence: 62%
“…23a ). Of note, our MymK model shares high similarity with that predicted by AlphaFold2 (AF-A6NI61-F1-model_v1) 39 and RoseTTAFold 40 , two other neural network-based methods.…”
Section: Resultsmentioning
confidence: 62%
“…In the most recent 14th Critical Assessment of Structure Prediction (CASP14), AlphaFold2 achieved a median score of 87.0 GDT for the free modeling category, which is a great improvement. RoseTTAFold 8 was developed using a similar end-to-end network architecture and achieved structure prediction accuracy close to that of AlphaFold2. The high accuracy of AlphaFlold2 make it possible to offer meaningful predicted structures to biologists, which is applied to human genome 9 .…”
Section: Protocol To Identify New Foldsmentioning
confidence: 99%
“…Therefore, we used a molecular modeling approach to explore whether AgMESH extracellular domains could interact with chitin. The overall domain architecture of the AgMESH polypeptide was modelled using the RoseTTAFold algorithm (Figure S3) (53). Individual domains within the A. gambiae MESH sequence and tertiary structural folds were predicted for several putative chitin-interacting or stabilizing domains using homology modeling.…”
Section: Mesh Promotes Integrity Of the Peritrophic Matrixmentioning
confidence: 99%