2019
DOI: 10.1101/850024
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

GraphProt2: A graph neural network-based method for predicting binding sites of RNA-binding proteins

Abstract: CLIP-seq is the state-of-the-art technique to experimentally determine transcriptome-wide binding sites of RNA-binding proteins (RBPs). However, it relies on gene expression which can be highly variable between conditions, and thus cannot provide a complete picture of the RBP binding landscape. This necessitates the use of computational methods to predict missing binding sites. Here we present GraphProt2, a computational RBP binding site prediction method based on graph convolutional neural networks (GCN). In … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
16
0

Year Published

2020
2020
2023
2023

Publication Types

Select...
5
3

Relationship

3
5

Authors

Journals

citations
Cited by 13 publications
(16 citation statements)
references
References 46 publications
0
16
0
Order By: Relevance
“…Computational prediction tools such as ( 21 , 59 ) promise to yield large amounts of higher-order RNA pairwise interaction data without need for costly crystallography experiments. This opens the door to applying such data in other important biological problems such as RNA binding protein prediction ( 60 ) and ion binding ( 40 ). Furthermore, the promising results obtained from the unsupervised pre-training provide a methodological building block for assisting in supervised learning on complex RNA structures.…”
Section: Discussionmentioning
confidence: 99%
“…Computational prediction tools such as ( 21 , 59 ) promise to yield large amounts of higher-order RNA pairwise interaction data without need for costly crystallography experiments. This opens the door to applying such data in other important biological problems such as RNA binding protein prediction ( 60 ) and ion binding ( 40 ). Furthermore, the promising results obtained from the unsupervised pre-training provide a methodological building block for assisting in supervised learning on complex RNA structures.…”
Section: Discussionmentioning
confidence: 99%
“…Note that this also includes sites at transcript ends, where full extension is only possible in the genomic context case. To assess any effects, three different prediction tools (DeepBind [11], Graph-Prot [12], and GraphProt2 [13]) were run on both context sets, using 10-fold cross validation and no additional features (i.e., only sequence information). Figure 2d shows the performance results as average accuracies over the 6 datasets, for both genomic and transcript context sets (see Table S4 for detailed results).…”
Section: Sequence Context Influences Binding Site Prediction Performamentioning
confidence: 99%
“…Note that this also includes sites at transcript ends, where full extension is only possible in the genomic context case. To assess any effects, three different prediction tools (DeepBind [11], GraphProt [12], and GraphProt2 [13]) were run on both context sets, using 10-fold cross validation and no additional features (i.e., only sequence information). Figure 2d shows the performance results as average accuracies over the 6 datasets, for both genomic and transcript context sets (see Table S4 for detailed results).…”
Section: Sequence Context Influences Binding Site Prediction Performamentioning
confidence: 99%