Candidate Gene Mining, Transcriptional Expression Analysis, and Functional Annotation Within Meta-QTL

Among the meta-QTL obtained, the ones considered highly reliable and refined were considered for gene mining. The criteria adopted were as follows: (1) the meta-QTL is generated through the projection of at least two overlapping QTL (otherwise the meta-QTL would simply correspond to a single QTL previously known); (2) it is shorter than 1.0 cM in genetic distance; (3) it is shorter than 20 Mb in physical distance (at Chinese Spring wheat reference genome). For that, the physical positions of markers present at the flanking regions of each highly refined meta-QTL (corresponding to the end of the confidence interval) were searched at the International Wheat Genome Sequencing Consortium reference genome sequence repository at https://wheat-urgi.versailles.inra.fr/Seq-Repository/BLAST through BLAST search of their sequences or either directly through surveying the annotation browser (International Wheat Genome Sequencing Consortium, 2014, 2018). The “IWGSC RefSeq v1.0” version was used. High confidence annotated genes (HighConfidenceGenesv1.1), within each highly refined meta-QTL, were then listed and thereafter called candidate genes.

For the transcriptional expression analysis, the collected candidate genes were analyzed through the expVIP (expression Visualization and Integration Platform)/Wheat Expression Browser online resource (http://www.wheat-expression.com) (Borrill et al., 2016; Ramírez-González et al., 2018). The available data for “Fusarium head blight infected spikelets” (Kugler et al., 2013) was used. This study assayed the transcriptome profile of a set of wheat lines after F. graminearum point inoculation and control conditions (mock) at 30 and 50 h after treatment. The tissue analyzed was spikelets. From this data set, only transcripts from the varieties, CM-82036 (resistant to FHB) and a susceptible near isogenic line, were selected (a cross between CM-82036 and the susceptible variety Remus). Differentially expressed genes (considering disease × mock conditions, only) were collected, based on the standard error. Following the proposed criteria by Wagner et al. (2013), only genes presenting at least 2 transcripts per million (TPM) were considered in this search. Other transcriptomic studies approaching other sources of FHB resistance could not be used in this work as they were not made available in manageable platforms like the one we have used here. Next, differentially expressed genes were further investigated for functional annotation (i.e., protein evidence), also at the International Wheat Genome Sequencing Consortium reference genome sequence repository (https://wheat-urgi.versailles.inra.fr/Seq-Repository/BLAST).