Protein-protein interaction extraction with feature selection by evaluating contribution levels of groups consisting of related features

Table 7 Number of positive and negative instances in four corpora: LLL, HPRD50, IEPA, and AIMed

Corpus	LLL	HPRD50	IEPA	AIMed
Positive instances	164	163	335	1000
Negative instances	166	270	482	4834

Four corpora, LLL, HPRD50, IEPA, and AIMed, were converted into a unified XML format with a very simple structure by Pyysalo et al. [17] to make the corpora easily accessible to users. Number of positive instances (interacting protein pairs) and negative instances (non-interacting protein pairs) in each corpus is shown

ISSN: 1471-2105