Uniprot datasets for training taxonomic classification

The data is based on the UniProt-Swiss-Prot release 2020-04 dataset and contains data derived from amino acid sequences of human, bacterial and viral origin. From each original sequence we created multiple patches of length 100 using a sliding window. The data is stored in the FASTA format according to<br> <br> &gt;{ID}_{patch index}|{class marker}<br> sequence<br> <br> with<br> <br> ID - denotes the ID of the original sequence in the UniProt-Swiss-Prot dataset<br> sequence - patch of length 100 of an amino acid sequence<br> patch_index - denotes the starting index of the given patch within the original sequence<br> class marker - indicates the taxonomic class<br> 0 - virus<br> 1 - bacteria<br> 2 - human / mammal<br> <br> The data is split into test, training and validation set which contain the following number of patches per class:<br> <br> <br> - train: 4.891.278<br> - test: 611.602<br> - val: 611.602