Table 1.

Cluster statistics

Protein classa	Cutting levelb	No. protein chainsc	No. building blocksd	No. clusterse	Average sequence lengthf	No. building block sequencesg
α/β	1	1514	2121	515	154.08	464
	2	1514	5257	1286	62.55	1804
	3	1514	8013	1899	38.75	3153
	4	1495	7813	1727	30.86	3137
	5	1174	9944	2124	29.37	3872
	6	595	5663	1402	28.94	2493
α + β	1	1619	2491	379	108.58	457
	2	1616	5445	828	49.45	1212
	3	1558	7454	1199	34.08	1929
	4	1339	7702	1285	28.98	2175
	5	520	4361	1089	28.38	1716
	6	153	1606	628	28.62	878
all-α	1	870	1419	405	85.88	548
	2	870	3181	751	41.81	1422
	3	794	3858	830	30.82	1921
	4	594	2971	727	28.62	1502
	5	263	2662	707	28.06	2131
	6	94	1299	466	29.28	722
all-β	1	1313	1757	428	115.47	514
	2	1311	3814	1049	53.58	1489
	3	1299	5815	1566	36.66	2449
	4	1132	4683	1456	32.23	2200
	5	523	4068	1283	30.46	1898
	6	167	1914	729	30.70	1018

^a The protein class that created the specific database.

^b The cutting level (1 to 6).

^c The number of protein chains participating in that database.

^d The total number of building blocks in that database.

^e The number of clusters resulting from the clustering procedure of that database.

^f The average sequence size at each nonredundant sequence database.

^g The number of sequences in the nonredundant sequence database.

The data here was calculated only from the representatives of the clusters to avoid bias toward large families that contain nearly identical building blocks.