BlueSTARR-K562

released
File Set Type
neural network
Summary
BlueSTARR-K562 v0.1.0 neural network predicting regulatory elements
Description
BlueSTARR is a neural network trained on STARR-seq data that takes genomic sequence as input and predicts STARR-seq activity as output. This version of BlueSTARR was trained on a STARR-seq experiment done in K562 cells, and takes a 300 bp sequence as input. Variant predictions are made by running the model twice (once for the reference allele and once for the alternate allele) and computing the predicted log2 fold change in STARR-seq activity for the change from reference to alternate.
Accession
IGVFDS3627LKSF
Model Version
v0.1.0
Prediction Objects
regulatory elements
Aliases
andrew-allen:bluestarr-k562-model
bill-majoros:bluestarr-k562-model

Files

2 items
json
graph structure
graph structure
Bill Majoros, Duke
9.3 KB
validated
hdf5
edge weights
edge weights
Bill Majoros, Duke
63 MB
validated

File Sets Using This Model Set as an Input

1 item
Accession
Type
Summary
Aliases
Lab
PredictionSet
functional effect prediction on scope of loci
  • andrew-allen:bluestarr-prediction-set-v3
Bill Majoros, Duke

Documents

1 item
Page
Description
Type
Download
Preview
Instructions to generate training data for BlueSTARR
model source data

Attribution

Award
Principal Investigator(s)
Contact P.I.
Lab