2.5 │ Generation of histone H1 subtypes and interacting proteins networks
Histone H1-partner proteins interaction networks were created with STRING (Search Tool for the Re­trieval of Interacting Genes/Proteins) platform (Version 11.0). 17 Protein-protein interaction data were predicted based on the active interaction sources including seven available parameters, i.e. co-expression, co-occurrence, databases, experiments, gene fusion, neighborhood and text mining, with medium (0.400) confidence interaction score. A STRING graph networks were further analyzed with Cytoscape (version 3.8.0) plugin app Network Analyzer 18 based on several topological parameters such as average number of neighbors (the average connectivity of a node in the network), network diameter (the largest distance between two nodes), network radius (the minimum among the non-zero eccentricites of the nodes in the network), network heterogeneity (the tendency of a network to contain hub nodes), network clustering coefficient (the average of the clustering coefficients for all nodes in the network) and the number of connected components (the connectivity of a network). An identification of hub objects in the networks was performed with the use of cytoHubba Cytoscape plugin, adopting the maximal clique centrality method (MCM) which was proposed as most effective in the exploration of the hub nodes in the network. 19 To find subnetworks, a Diffusion tool which is an integrated Cytoscape function for performing a network propagation 20 was adopted.