Table 3.

Sequence length distribution. The minimum length is 31 residues and the maximum length is 40 residues.

Residues	Training set	Test set	Total
31	1	0	1
32	0	0	0
33	2	2	4
34	18	22	40
35	210	189	399
36	1142	1162	2304
37	30	31	61
38	11	10	21
39	11	8	19
40	0	1	1
Total	1425	1425	2850