Skip to main content
. Author manuscript; available in PMC: 2009 Jun 8.
Published in final edited form as: J Mol Biol. 2005 Jan 16;346(4):1173–1188. doi: 10.1016/j.jmb.2004.12.032

Table 3.

Difficult structure pairs with good alignments found just by one method

Categorya Total SSAP STRUCTAL DALI LSQMAN CE SSM
All α–All α (same CAT)b 369(8) 0 103 (5) 2 (0) 212 (2) 0 52 (1)
All α–All β 61 0 57 0 2 0 2
All α–Mixed α/β 610 0 275 0 243 0 92
All α–Few Secondary Structure 13 0 7 1 4 0 1
All β–All β (same CAT)b 37 (5) 1 (0) 24 (4) 1 (0) 10 (1) 0 1 (0)
All β–Mixed α/β 318 0 260 1 39 0 18
All β–Few Secondary Structure 0 0 0 0 0 0 0
Mixed α/β–Mixed α/β (same CAT)b 292 (8) 0 130 (4) 1 (0) 111 (1) 0 50 (3)
Mixed α/β–Few Secondary Structure 4 0 2 0 2 0 0
Few Secondary Structure–Few Secondary Structure 0 0 0 0 0 0 0
Total 1704 1 858 6 623 0 216

Good alignments are defined as those where one of the methods finds a good alignment (SAS<4 Å and matches more than 35 residues), while all other methods find bad alignments (SAS>6 Å for any length of match). We also tested other definitions of difficult alignments (e.g. longer matches, or less stringent SAS values) and found a similar distribution amongst the different methods (data not shown).

a

We categorize the pairs of aligned structures by their class (C) classifier.

b

We list in parenthesis the number of difficult alignments found with structures of the same CAT classification.