LSTM Branchpoint Retriever (LaBranchoR )
LaBranchoR predicts RNA splicing branchpoints using a Long Short-Term Memory network. LaBranchoR predicts a correct branchpoint for over 90% of 3' splice sites.
For Competitors: LaBranchoR is trained on the 'high confidence' set of Mercer et. al. 2015 branchpoints on chromosomes 5 to 21, X and Y. Compare to us on chromosomes 1-4! Good luck!
LaBranchoR predictions on UCSC genome browser.
Access LaBranchoR predictions on the UCSC genome browser here.
Download LaBranchoR predictions
All files are for Gencode v19 annotations (hg19 coordinates).
-
Top scoring branchpoint for hg19 3'ss (recommended) 8/25/2017 -
Scores for positions 1-70 upstream of each 3'ss 8/25/2017 -
In silico mutagenesis scores for positions 1-70 upstream of each 3'ss 11/14/2017
LaBranchoR source code
LaBranchoR public git repository can be found here.
How to cite
Paggi, J.M., and Bejerano, G. A sequence-based, deep learning model accurately predicts RNA splicing branchpoints. bioRxiv 185868, 2017. DOI: 10.1101/185868
Contact
Email jpaggi at stanford.edu with any questions.
Licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. Please email 
for commercial licensing inquiries.