HIGHLIGHTS
SUMMARY
Corpus generation To generate the gene corpus, proteins encoded by the non-redundant set of contigs were scanned using the KO HMM database described above. The authors annotated gene families based on KO database because it is extensive and contains detailed metadata and systemic information on the function of each gene family. For the sequence-based approaches, the authors searched each word against all words in the dataset that were not assigned the same KO as the query and determined the predicted functional category based on the best hit. When testing the classifiers, each . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.