Argument-predicate distance as a filter for enhancing precision in extracting predications on the genetic etiology of disease

Table 1 SemGen precision for 2,042 genetic (gene-gene) and etiologic (gene-disease) semantic relations

Semantic relations	Syntactic indicator	Official symbol	Total relations	Correct relations	Incorrect relations	Precision	Standard error
gene-gene	verb	yes	410	172	238	41.95%	2.44%
gene-gene	verb	no	617	299	318	48.46%	2.01%
gene-disease	verb	yes	407	302	105	74.20%	2.17%
gene-disease	verb	no	26	21	5	80.77%	7.73%
gene-disease	preposition	yes	555	357	198	64.32%	2.03%
gene-disease	preposition	no	27	17	10	62.96%	9.29%

"Official symbol" indicates whether all relevant relations include an official gene symbol and Entrez Gene ID for each extracted gene or not.

ISSN: 1471-2105