Markup | Genome Research

Table 3.

Performance of Neural Network Recognizers for Various Domain Groups

A. Protein Domain Groups	Training set		Test set		Total

	tp	fp		fn	tn	tp	fp	fn	tn	tp	fp	fn	tn

Immunoglobulin domain	1721	24		15	2428	829	18	19	1155	2552	71	32	3554

Zinc finger, C2H2 type	330	9		9	540	147	3	6	245	477	12	15	785

Protein kinase domain	1120	0		0	264	522	0	0	119	1642	0	0	383

EF hand	409	14		1	116	186	3	1	62	594	17	3	178

EGF-like domain	291	2		0	332	145	3	0	137	436	3	0	471

WD repeat	186	2		4	168	76	0	1	92	262	2	5	260

Fibronectin type III domain	237	1		0	507	113	0	0	228	350	2	0	734

ABC transporters	530	3		0	90	239	2	0	43	769	1	0	137

Globin	595	2		2	200	269	1	0	92	866	3	0	292

Homeobox domain	584	0		1	58	260	0	0	31	845	0	0	89

ANK repeat	99	1		3	56	54	0	1	22	155	2	2	77

Sushi domain (SCR repeat)	106	0		0	77	59	0	0	32	165	0	0	109

RNA recognition motif (=RRM, RBD, or RNP domain)	215	1		2	50	99	1	1	26	315	2	2	76

Tetratricopeptide repeat	98	0		0	152	45	0	0	74	143	0	0	226

Short chain dehydrogenase	316	0		0	55	142	0	0	29	458	0	0	84

Trypsin	296	0		1	215	146	0	0	97	442	0	1	312

Dead/H box helicase domain	277	1		4	499	117	0	2	238	397	2	3	736

Kazal-type serine protease inhibitor domain	211	0		0	84	99	0	0	38	310	0	0	122

Response regulator receiver domain	242	0		0	20	112	0	0	11	354	0	0	31

Spectrin repeat	21	0		1	293	21	0	1	124	43	0	1	417

4FE-4S ferredoxins and rel. iron-sulf. clust.bind.domains	183	0		0	51	86	0	0	24	269	0	0	75

RAS family	203	0		0	32	96	0	0	14	299	0	0	46

SH3 domain	185	0		0	27	94	0	0	6	279	0	0	33

APOA/APOE-repeat	33	0		0	498	16	5	2	229	51	12	0	720

Zinc-binding dehydrogenases	152	0		0	74	80	0	1	26	233	0	0	100

Cytochrome C	142	1		6	73	71	0	1	33	213	1	7	106

ATPases associated with various cell. activites (AAA)	122	0		0	102	69	0	0	38	191	0	0	140

Helix-loop-helix DNA-binding domain	153	0		1	21	74	0	0	13	227	0	1	34

Alpha amylase	135	0		0	118	69	0	0	51	204	0	0	169

HSP70 protein	130	1		0	247	63	0	0	117	193	0	0	365

Ligand-binding domain of nuclear hormone receptors	176	1		0	13	81	0	0	10	257	0	0	14

Protein-tyrosine-phosphatase domain	110	0		0	42	56	0	0	20	166	0	0	62

Lectin in C-type domain	129	0		0	23	59	0	0	15	188	0	0	38

Intermediate filament proteins	127	2		0	192	61	2	0	86	188	2	0	280

Zinc-binding metalloprotease domain	148	1		0	4	75	0	0	1	223	1	0	5

ACYL carrier protein domain	102	0		0	8	54	0	0	2	156	0	0	10

PAS domain	77	0		0	52	46	0	0	15	123	0	0	67

FOS/JUN DNA-binding domain	108	0		0	69	64	0	0	23	172	0	0	92

Snake toxin	102	0		1	31	51	0	0	12	153	0	1	43



B. Functional Groups	Training set		Test set		Total

	tp	fp		fn	tn	tp	fp	fn	tn	tp	fp	fn	tn

Permeases	289	22		12	593	139	7	4	270	428	29	16	863

Sensory transduction histidine kinases	177	2		3	167	87	2	1	75	264	2	4	244

Glycosyltransferases I	204	1		1	103	95	0	0	46	300	2	0	148

Thiol-disulfide isomerase and thioredoxins	139	0		4	65	77	1	4	17	218	1	6	82

Serine/threonine protein kinases	1107	1		10	264	515	1	2	121	1622	1	12	386