Algorithm 3: Pruning

Input: k-mers derived from a specific sequence in the proteome    Output: groups//groups of continuous k-mers within the input sequence    BEGIN        Sort k-mers by their initial position in the given proteome sequence        total <- LENGTH(k-mers)//total number of k-mers    create the empty set groups whose elements are k-group//groups contain k-group, //every fragment of the proteome covered by consecutive k-mers predicted as AMPs.        j <- 1        k-group.first <- k-mer[0].first //the initial position of the left most k-mer        while j < total do               k-group.last <- k-mer[j − 1].last//the last position of the j − 1 k-mer               if k-mer[j] NOT Embedded(k-group)                      if k-mer[j] Connected(k-group) OR k-mer[j] Intersects(k-group) then                             k-last <- k-mer[j]                      else                             Add k-group to groups                             k-group.first <- k-mer[j].first                             k-group.last <- k-mer[j].last    END