Datasets used in "Ultra-fast Variant Effect Prediction with probNorm"

These files include all the preprocessed datasets (ADASTRA, bQTLs, caQTLs, UDACHA, HAR, HOCOMOCO mono-nucleotide PWMs and di-nucleotide DWMs) used in the study Ultra-fast variant effect prediction using biophysical transcription factor binding models along with an example R code to make motifDiff predictions for the provided variant sets and then fine-tuning the predictions by ElasticNet regression model as described in the paper.