Biomarker Identification for Prostate Cancer and Lymph Node Metastasis from Microarray Data and Protein Interaction Network Using Gene Prioritization Method

Table 7

Network based gene prioritization methods.

Method	Brief description
Endeavor	Machine learning: using initial known disease genes; then multiple genomic data sources to rank [17]
HITS with priors; Page rank; K-Step markov	310 cm prioritization based on networks using social and web networks analysis [18]
CGI	Combination of protein interaction network and gene expression using markov random field theory [19]
CANDID	Uses publications, protein domain descriptions, cross species conservation measures, gene expression profiles and Protein Interaction Networks [20]
IDEA	Uses the interactome and microarray data [21]
Limitations	Most of these approaches include additional interactions predicted from coexpression, pathway, functional or literature data, but still fail to incorporate weights expressing the confidence on the evidence of the interactions. Another issue is that previous methods start with the given PIN without filtering its edges, to keep more relevant interactions to the disease
GP-MIDAS-VXEF	Our proposed method, integrates protein interaction network with normal and disease microarray data, using this integration we apply all-pairs shortest paths to find the significant networks and calculate the score for the genes. Additionally our method filters interactions, in such way the most relevant interactions are left for analysis