TransportTP: A two-phase classification approach for membrane transporter prediction and characterization

BMC Bioinformatics

Table 2 The cross-validation results yielded by all model organisms.

Organism	Num of proteins	Predictions by TransportTP	Annotations in TransportDB	Matches	Mismatches	TransportDB unique	Text mining validated	Recall (%)	Precision (%)	Balanced accuracy (%)
P. torridus	1535	169	171	134	1	36	16	78.4	79.3	78.8
P. profundum	5489	603	580	472	4	104	41	81.4	78.3	79.8
D. psychrophila	3234	327	305	246	1	58	37	80.7	75.2	77.9
A. fumigatus	9923	705	619	556	1	62	52	89.8	78.9	84.0
Average on non-model proteomes								82.6	78.0	80.2

The proteomes of seven model organisms were used for training and proteomes of four non-model organisms were used for testing at the e-value threshold of 0.1. TransportDB was chosen for the benchmark transporter database. The definition of the columns can be referred to Table 1.

ISSN: 1471-2105