IntSplice predicts a splicing consequence of a single nucleotide variation (SNV) at intronic positions -50 to -3 close to the 3' end of an intron of the human genome. Compared to our previous ver. 1.x that was modeled by Support Vector Machine (SVM), ver. 2.0 used a newly built training dataset that included 54% more pathogenic and common SNVs, and was modeled by Gradient Boosting (GB). Pre-processed genome-wide scores of IntSplice ver. 2.0 are available under the “Download” tab.