File:Fpls-02-00023-g003.png

English: Analysis of primary and secondary cellulose biosynthesis networks in the seven species using Network Comparer. Seven primary cell wall and seven secondary CESA gene vicinity co-expression networks have been analyzed for enrichment of PFAMs. Total presence of at least one gene of the respective PFAM in the fourteen co-expression networks is indicated in the first column. The heatmaps in the second and third column represent enrichment of a gene family in primary (P) or secondary cell walls (S) and monocots (M) or dicots (D) based on the difference of occurrence of genes of the respective PFAM in the 14 co-expression networks. For example, the genes from the NAM family (Transcription factor section) are present in two primary (P) and six secondary (S) co-expression networks (Table S1 in Supplementary Material), which sums up to eight in total (first column), and the difference of P − S is −4 (second column). At the same time, genes from this NAM family are present in five monocot (M) and three dicot (D) co-expression networks resulting in a difference M − D of two (third column). Any PFAM was defined as enriched in primary or secondary cell walls and marked green or orange, if P − S ³ 4 or P − S £ −4, respectively. Enrichment for monocots or dicots, with corresponding PFAMs marked in blue and red, was defined as M − D ³ 3 or M − D £ −5, respectively, because eight dicot and only six monocot networks (corresponding to four dicot and three monocot species) were analyzed. Core components (marked in bold letters) were defined as gene families that were present in at least ten networks, without being enriched in monocots or dicots, nor primary or secondary cell wall biosynthesis.