Module Construction

network modules were constructed using the louvain method (version 1 networks) or the nested leiden approach (version 2 networks), including the genes and edges post edge filtering and prioritization ( see version specific variations docid\ ma6qmd9m7kotqydtwakjw ) the modules has been labeled based on their respective top enriched go\ bp terms and their connection in the go hierarchy using deltacon, as described under biological relevance docid\ m3ywqj3arjsy4zaf7wthc , the module genes were further divided into conserved and context specific subcategories the conserved/context specific gene lists are included in discovery as mavatar curated gene lists for graph generation and functional enrichment analysis (only the context specific gene lists for version 1 networks) the conserved gene lists are further used for dina network similarity analysis extensive validation analyses have been made on the biological relevance docid\ m3ywqj3arjsy4zaf7wthc of the networks based on louvain communities, as well as on the gene list values for dina network similarity validations docid\ uhclpee8 fosx5lplvkqt further evaluations on the improvements on module quality and robustness were made on the updates for version 2 networks, using both structural metrics and biological validation criteria