Fig. 3. The use of high-throughput transcriptomic data for discovery of MIA biosynthetic genes. Two non-exclusive strategies may be used: orthology comparison with other species (left panel) and clustering of gene expression (right panel). The former approach requires transcriptomic data from candidate species with specific alkaloid metabolisms (e.g. secologanin (11)-producing (purple inset) and non secologanin (11)-producing (blue inset) species). Such data are available from the MPGR and PhytoMetaSyn projects. A first orthology comparison using the OrthoMCL algorithm (Li et al., 2003) could help identify conserved genes encoding enzymes involved in the biosynthesis of common precursors in MIA-producing species (Camptotheca acuminata, Rauwolfia serpentina, C. roseus) but also specific enzymatic activities dedicated to the synthesis of particular alkaloids, such as the C. roseus specific genes. A second orthology analysis can be further performed on the set of conserved genes found in the MIAproducing species to isolate candidates from the secoiridoid pathway by comparing with transcriptomic data from non secologanin (11)-producing enzymes. The second strategy relies on transcriptome analyses from diverse experimental conditions to construct a reference transcriptome which will be used to estimate the abundance of each transcript in each condition in order to cluster genes according to their expression levels. The more experimental conditions are available, the more the clustering analysis will be efficient to group co-regulated genes. Note that this reference transcriptome could also serve to the orthology analysis depicted in the left panel, and also to verify expression levels of candidate genes resulting from this kind of approach. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)