Each file correspond to one protein. Its file name is protein ID. Its file type is comma-separated ASCII text file. Inside each file, Each row is a possible nsSNP. The first column is sequence position. The second column is variant allele. The third column is deleterious prediction score (DPS). Note that reference allele is from the reference Ensembl proteome. DPS is a weighted vote from 100 decision trees in the random forest model trained on the HumDiv dataset. The higher the DPS, the more deleterious is the nsSNP. If you find the predictions useful, please cite: Wong and Zhang. SNPdryad: Predicting Deleterious Non-synonymous Human SNPs Using Only Orthologous Protein Sequences. Bioinformatics