Fig. 1: Pathway coessentiality mapping reveals a unique association between complex II and nucleotides.
a, Development of pathway mapping model. Gene sets representing ~15,000 genes were used to select essentiality data across all cancer cell lines in the Cancer Dependency Map. For each gene set (depicted as ‘pathway’), the first four PCs were extracted from the gene dependency matrix and CCA was used to determine correlations between pathways. CC here refers to canonical correlation. b, Heat map of the canonical correlations between each ETC complex and 2,827 pathway gene sets. k-nearest-neighbours hierarchical clustering was used to group pathways by their coessentiality patterns (side). The colour of each cell represents the relative correlation strength for each pathway–complex comparison. c, Network diagram showing the connectedness of ETC complexes with GO pathways, with each ETC complex or pathway represented as a node and each connection as an edge. Notably, a larger sub-network includes complexes I, III, IV and V and related pathways, whereas a smaller sub-network includes only complex II and its unique pathways. d, The association between ETC complexes (consisting of both core and accessory subunits; Extended Data Fig. 1). The colour of each cell and the associated correlation score depict the strength of the complex–complex association. NA, not applicable. e, Top pathway associations for each ETC complex, with multiple shared pathways shown for complexes I, III, IV and V and distinct pathways for complex II. The correlation strength between each pathway and complex is shown at the bottom. rRNA, ribosomal RNA. For c and e, the colour of nodes or bars represents a GO pathway type as indicated. Panel a created with BioRender.com.
