Résumé
The rubber tree (Hevea spp.), primary plant used for natural rubber production, have a long breeding cycle. However, the advent of molecular genetic markers favored early assessment methods, allowing the characterization of genotypes, genetic diversity analysis, QTL analysis and identifying genes of interest. In this context, this work aimed at the development and genotyping of new SNP markers presents in regulatory regions in two populations of genetic mapping of rubber tree. The raw data for each pool of samples (GT1 and RRIM701 genotypes) were separately trimmed and de novo assembled in a unique file using the CLC Genomics Workbench software (Version 6.0.1, CLC Bio, Denmark). Next, BWA aligned short reads back to assembled contigs, and Freebayes was used to perform variant calling and SNP detection. Using the varFilter command of vcfutils script, SNPs were filtered only for positions with a minimal mapping quality and coverage of 30 and 10 respectively. Unique and shared SNPs among the two cultivars were extracted with the VCFtools software. The selected SNPs annotation was made in two databases of cis regions: PLACE and ALGEEN-PROMO. Identification of SNP allelic forms was performed using KASPAR chemistry, either with PCR final point analysis on rtPCR equipment (LC480) for testing the polymorphism of the parents, or with Fluidigm technology for progeny genotyping. We obtained an average of 76% success rate in genotyping. The results will be used for saturation of molecular genetic map of H. brasiliensis, the location of these cis regions mutated on the map. (Résumé d'auteur)