Recognition sequences are the key to identifying and studying restriction enzymes. Among the four classes of restriction enzymes, the Type I enzymes, although the earliest discovered, have the most complex recognition sequences and this makes them difficult to identify. Traditional experimental methods to find the recognition sequences are complex and computational methods have been hampered by poor algorithm performance.  

Here, We present Enzyme Finder, an efficient computer-aided approach for finding Type I restriction enzyme target sequences. It considers examples of both positive and negative sequence examples, and optimizes for the case of long sequence length and number of positive target sequence examples.  The underlying techniques include the sliding window for obtaining all recognition pattern instances in a sequence and the inverted index for supporting thresholds and reducing memory overheads.  Enzyme Finder provides a user-friendly GUIs to guide users to construct and monitor the sequence-finding process and to highlight potential target recognition sequences.