An integrative approach to identifying cancer chemoresistance-associated pathways

BMC Medical Genomics

Table 2 Lists of # of genes and relations in integrated database

*Statistics information of integrated databases*
Database	Organism	# of gene	# of relation
PID+KEGG+TRANSFAC	Homo sapiens	8173	9308
Reactome	Homo sapiens	538	31240
*Statistics information on each of the three databases*
Database	# of TFs	# of target gene parsed	# of pairing regulate relation parsed
TRANSFAC	157	825	529625
Database	# of pathways	# of gene, protein, enzyme parsed	# of relation parsed
PID + KEGG	197	18937	8880
PID	60
KEGG	137

We integrated the PID (the date of version, July 15, 2008), KEGG (release 47.0, July 1, 2008) and TRANSFAC public databases (version 7.0), and further eliminated duplicated reactions and elements. Accordingly, 8173 genes and 9308 interactions were remained. To assess the importance of genes within each filtered pathway, we also implemented the betweenness centrality and degree centrality for each node. The degree and betweenness centrality of genes were calculated using the Reactome database [31] as a base to cross validate our experimental results. Pathways downloaded from PID and KEGG were parsed by batch processing. A gene (or protein) may be involved in several pathways, which means some genes were repeated. Therefore, the number of parsed entity (including genes, proteins, and enzymes) was 18937. Moreover, one gene may be regulated by several TFs, or one TF may regulate numerous target genes. As a result, the total number of pairing regulate relation parsed was 529625.

ISSN: 1755-8794