Fig. 3. Results of the expansion and mining for QS entries based on the proposed systematic workflow.

a Accuracy, precision, recall, and F1 score of the four classifiers based on DNN, SVM, KNN, and RF algorithms. b Counts of the annotated and uncharacterized positives for the union set of positives from DNN, SVM, KNN, and RF classifiers; c Results of the protein clusters of 534 re-annotated protein entries. d Distribution total of 28,567 redundancy removal entries in 818 gut microbes. Source data are provided as a Source Data file.