Fig. 6

Systems genetics analysis of LAC/PRX genes associated with SCW biosynthesis in Eucalyptus xylem formation. The network was built from data exported from qtlXplorer and illustrates a systems genetics model for three candidate genes out of the eight previously presented. Cis- and trans-eQTL associations connecting candidate genes (blue nodes) or cell wall-related genes (green nodes) to eQTL peak positions (turquoise triangles) are represented by red and blue edges, respectively. Transcription factors (TFs) underlying eQTL peak positions are represented by grey or green (SCW-related) squares and connected to eQTL peak positions via grey dashed line edges. Gene expression profile correlations are represented by grey solid edges, with thickness proportional to the absolute value of the correlation. Physical overlap of trait QTLs (purple nodes) with candidate gene or eQTL peak positions are represented by green dashed lines. eQTL peak position node size is proportional to the number of genes having eQTLs (cis or trans) mapped at that genomic position. Underlying TF node size is proportional to the number and average value of the correlations of TFs with genes having cis- or trans-eQTLs mapped at that position (underlying TF score; Additional file 1: Table S8). Candidate gene node size is proportional to its score for prioritization (see Additional file 2: Methods S5; Additional file 1: Table S9), taking into account (i) their correlations (number and average value) with SCW genes across the population-wide transcriptomic data, (ii) the number of physical overlaps with candidate gene/SCW-related QTLs, (iii) the number of eQTLs mapped at SCW-enriched eQTL positions, (iv) the number of overlaps of their eQTL positions with SCW-related trait QTLs, and (v) the number of SCW-related TFs in the top 10 best candidate TFs underlying their eQTL positions