MutationTaster2025 Supplementary Data

Random Forest models

Model	Features	Performance scores	Grid search	Random Forest model
without_aae	.txt	.txt	.csv	.pkl	.json
simple_aae	.txt	.txt	.csv	.pkl	.json
complex_aae	.txt	.txt	.csv	.pkl	.json
5utr	.txt	.txt	.csv	.pkl	.json
3utr	.txt	.txt	.csv	.pkl	.json

Training data

For gnomAD and ClinVar, training and test sets have the format:

chr   pos   ref   alt   transcript

Training and test sets were generated at transcript level. However the split into training and test data was done so that no variant, even on different transcripts appears in the training and test data at the same time.

The full test and training set will be available as soon as possible.

Updated

20 December 2024