Towards an automatic classification of protein structural domains based on structural similarity

BMC Bioinformatics

Table 4 Δ-distance*, in 1000s, between partitions within each method and minimum Δ-distance from each method to SCOP

Method	Average Distance◊	Maximum Distance◊	Average Distance to SCOP	Minimum Distance to SCOP
1 VAST	43	56	73	65
2 SHEBA	35	47	72	65
3 DALI	37	53	73	63
4 Size	28	37	66	63
5 Level	48	59	80	70

*Δ-distance as defined in Method, Eq. 8.
◊ Average or Maximum distance among partitions of a particular structural comparison program, or of a tree-cutting strategy. Average distance among SHEBA partitions for example, is obtained by averaging across all SHEBA similarity scores and tree cutting strategies. Average distance among largest size cut partitions for example, is obtained by averaging across all similarity scores of structure comparison programs VAST, SHEBA and DALI.

ISSN: 1471-2105