Co-expression network of species

To construct co-expression networks for each species, we downloaded raw reads from SRA NCBI (see Data section), and quantified the gene expressions across samples. Then, we constructed the networks for each condition using Rapid Conditional Fused Graphical Lasso (RCFGL) method (paper and GitHub). Precision matrices were calculated and then based on the covariance the correlation matrix were calculated. The resulting correlation matrix contains the Pearson correlation coefficients, which range from -1 to 1 and indicate the strength and direction of the linear relationship between pairs of variables. Then, we used a truncation cut-off (0.05) and made the adjacency matrix. These adjacency matrices were used as inputs of igraph to perform network analysis and visualization. For enrichment analysis, we used clusterProfiler package with all genes that were used for the network construction as background gene and the following inputs: pvaluecutoff = 0.05, qvaluecutoff = 0.05. The networks are visualized using visNetwork.

Union, intersection, difference operations for graphs

To compare networks with each other, we can use some functions such as union (aggregation of all edges in input networks), intersection (conserved edges among all networks), and difference (A-B; The set of all edges that are present in A and are not present in B). Of course, for the difference operation the order of inputs are important. If you use more than one graph, we would calculate the union of them to further calculate the difference between A and B. Putting the results into biological context could be difficult in some comparisons.

Update the input and click the button

If you want to change the input or comparison, please do not forget to click on the button at the end of input list. There will be a notification on the bottom right of your screen mention when the update started and when it is ready.