Table 1 The selected numbers and sizes of the top 5 most frequent feature types from the selected feature subset using S212 and NT-S160.

From: Improved prediction of anti-angiogenic peptides based on machine learning models and comprehensive features from peptide sequences