Figure 5
From: CONTRAST: a discriminative, phylogeny-free approach to multiple informant de novogene prediction
Features that score a label based on local sequence. CONTRAST contains three types of features for scoring a label based on local sequence: features based on hexamers in the target sequence (shown in blue), features based on a trimer in the target sequence and a trimer in an informant alignment (shown in red) and features based on a position in the EST sequence (shown in green).