The DeepNovoV1 branch contains a pytorch re-implementation of DeepNovo
The DeepNovoV2 branch contains the implementation of our proposed DeepNovoV2 model.
python >= 3.6
pytorch >= 1.0
dataclasses
The ABRF DDA spectrums (the data for Table 1 in the original paper) and the default knapsack.npy file could be downloaded here. And the 9 species data could be downloaded here: ftp://massive.ucsd.edu/MSV000081382/peak/DeepNovo/HighResolution/.
It is worth noting that in our implementation we represent training samples in a slightly different format (i.e. peptide stored in a csv file and spectrums stored in mgf files). We also include a script for converting the file format (data_format_converter.py in DeepNovoV2 branch).
first build cython modules
make build
train mode:
make train
denovo mode:
make denovo
evaluate denovo result:
make test