Analysis of substructural variation in families of enzymatic proteins with applications to protein function prediction

Table 1 Full protein family dataset used for function prediction experiments.

EC class	PDB ID (Chain)	Amino acid number^Labels	EC class size
1.1.1.1	1HET (A)	46^C, 48^S, 67^H, 174^C	82
1.1.1.21	1US0 (A)	43^D, 48^Y, 76^S, 77^K, 110^H	89
1.11.1.7	1ARU (A)	52^RQ, 56^H, 57^D, 93^NR, 184^H	83
1.14.13.39	1DWW (A)	194^C, 346^V, 363^F, 366^W, 367^Y	126
2.5.1.18	2A2R (A)	7^Y, 13^FLR, 47^ACFLM, 108^CFLY	190
2.6.1.1	2QA3 (A)	32^G, 34^G, 183^N, 374^R	105
2.7.4.6	1NHK (R)	51^Y, 117^H, 119^S, 128^K	60
3.1.1.7	1H23 (A)	84^W, 117^G, 130^Y, 279^W, 330^F	110
3.1.3.1	1ANI (A)	51^D, 101^D, 102^S, 331^H, 412^H,	44
3.1.3.48	2CM2 (A)	181^DE, 182^FHMY, 216^S, 221^R, 266^Q	248
3.2.1.1	1HT6 (A)	52^G, 178^R, 180^D, 205^E, 291^D	133
3.5.2.6	1YLJ (A)	70^S, 73^K, 130^S, 132^N	254
4.2.1.1	1HCB (A)	94^H, 96^H, 106^E, 119^H, 199^T	282
5.3.1.1	1YPI (A)	12^K, 95^H, 96^S, 165^A	95
5.3.1.5	1DID (A)	53^H, 56^D, 93^F, 136^W, 182^K	71

For each EC class family, a single PDB structure was used to define an input motif. The list of amino acid numbers are documented functional residues found within the primary PDB http://www.pdb.org reference corresponding to each PDB structure. The superscript labels above each amino acid number are the possible amino acid types that can match at each motif point; further details of alternate amino acid label use can be found here [43]. Where multiple amino acid labels per motif point appear, they were determined using ConSurf [28].

ISSN: 1471-2105