K2. Gene Regulatory Network for H1N1 gene expression data

Inferring Gene Regulatory Network (GRN) is important to understand genetic changes occurring in cells. GRNs help in designing drugs and vaccines. H1N1 is the most endemic disease recently and still does not have a known medication. To comprehend alterations in genes due to H1N1 infection, we attempted to build GRN based on gene expressions of H1N1 infected cells. Two approaches were headed to build such GRN. First, synthetic data was used to evaluate six GRN inferring algorithms (Pearson's correlation, mutual information (MI), MRNET, ARCENE, CLR and GENIE3). After excluding correlation as it gave worst results, the other algorithms were yet applied on H1N1 data. Second, ANOVA test was used to determine the most significant genes related to H1N1 infection. These genes were sorted in ascending order and the top 100 genes were used in employing (MI, MRNET, ARCENE, CLR and GIENE3) algorithms to build the adjacency matrix representing each GRN.