Algorithms for matching partially labelled sequence graphs

Table 3 Predicted contact numbers assuming one known pair

PDB	Domains 1,2	Domains 1,3		Domains 2,3		Score
	(Known)	T	F	T	F
1aoz	3	3	0	3	0	6/6
1lci	10	0	2	8	1	8/11
1pkm	11	3	7	4	6	7/20
3ctz	7	7	0	6	1	13/14
3vqt	6	1	5	2	4	3/12
4rcn	3	1	2	2	1	3/6

For each protein (“PDB”) the maximum number of correct contacts between domains 1 and 2 (“known”) was used to limit the residue pairs considered between the other domain pairs. In these interactions, the number of true (“T”) and false (“F”) contacts is tabulated and scored by the fraction correct (“score”)

ISSN: 1748-7188