The information in the current readme file is not enough to know how to train a new model, i.e. going from an arbitrary FASTA-formatted reference database to a PROTAX model. I think the requirements and specifications of the "training data" and "target data" should be better defined. The original PROTAX paper (https://doi.org/10.1093/bioinformatics/btw346) also talks quite a bit about using the results from other taxonomic classifiers as predictors for PROTAX - this would also be good to have detailed instructions for.
The information in the current readme file is not enough to know how to train a new model, i.e. going from an arbitrary FASTA-formatted reference database to a PROTAX model. I think the requirements and specifications of the "training data" and "target data" should be better defined. The original PROTAX paper (https://doi.org/10.1093/bioinformatics/btw346) also talks quite a bit about using the results from other taxonomic classifiers as predictors for PROTAX - this would also be good to have detailed instructions for.