Construction of co-expression network based on natural expression variation of xylogenesis-related transcripts in Eucalyptus tereticornis.

Natural genetic variation is randomly distributed and gene expression patterns vary widely in natural populations. These variations are an effect of multifactorial genetic perturbations resulting in different phenotypes. Genome-wide analysis can be used to comprehend the genetic basis governing this naturally occurring developmental variation. Secondary growth is a highly complex trait and systems genetics models are presently being applied to understand the molecular architecture of wood formation. In the present study, the natural variation in expression patterns of 18,987 transcripts expressed in the developing xylem tissues were documented across four phenotypes of Eucalyptus tereticornis with distinct holocellulose/klason lignin content. The differentially expressed genes across all the phenotypes were used to construct co-expression networks and sub-network 2 with 380 nodes and 17,711 edges was determined as the network of relevance, including 30 major cell wall biogenesis related transcripts with 2394 interactions and 10 families of transcription factors with 3360 interactions. EYE [EMBRYO YELLOW] was identified as major hub transcript with 173 degrees which interacted with known cell wall biogenesis genes. K-mean clustering was also performed for differentially expressed transcripts and two clusters discriminated the phenotypes based on their holocellulose/klason lignin content. The cluster based networks were enriched with GOs related to cell wall biogenesis and sugar metabolism. The networks developed in the present study enabled identification of critical regulators and novel transcripts whose expression variation could presumably govern the phenotypic variation in wood properties across E. tereticornis.