Automated workflow-based exploitation of pathway databases provides new insights into genetic associations of metabolite profiles

Table 2 Performance of the database:interrogation schemes in GWAS dataset analysis

Database: interrogation scheme	Size of gene set¹	Top hits from Illig et al. study identified by the method²	Sensitivity³
BioCyc pathway	399	ACADL, ACADM, ACSL1, CPS1, FADS1, PHGDH, SCD, SPTLC3	0.53
BioCyc reaction	806	ACADM, ACADS, ACSL1, CPS1, FADS1, SCD, SPTLC3	0.47
KEGG pathway	703	ACADL, ACADM, ACADS, ACSL1, CPS1, ELOVL2, FADS1, PHGDH, SCD, SPTLC3	0.67
KEGG reaction	768	ACADL, ACADM, ACADS, ACSL1, CPS1, PHGDH, SCD, SPTLC3	0.53
Pooled set	1246	ACADL, ACADM, ACADS, ACSL1, CPS1, ELOVL2, FADS1, PHGDH, SCD, SPTLC3	0.67

Snapshot of the matches between our method and the association data from the Illig et al. 2010 study for each of the database:interrogation scheme. ¹corresponds to the unique set of genes generated for all the metabolites for the given database:interrogation scheme. ²corresponds to the top hits in the Illig et al. publication that were present in the gene set for the given database:interrogation scheme. ³Sensitivity is a measure of the actual positives that have been captured by our method and is equal to the ratio of the number of top hits identified by the method over the total number of top hits in the Illig et al. publication which is 15.

ISSN: 1471-2164