Exploring QTLxE in a network of fields: identification of QTLs associated with scenarios of heat and drought for predicting yields in future climates
Résumé
Genetic analysis of tolerance to climate change is hampered
by climatic variability that generates GxE interactions depending
on the timing and intensity of water or heat stresses. We
have performed a genome-wide analysis of yield in 29 fields
over Europe, with irrigated and rainfed treatments in each location.
Six scenarios of heat and drought were first identified by
clustering observations over 60 European sites x 35 years. A
detailed environmental characterisation was carried out in all
experiments, which were ascribed to environmental scenarios.
A panel of 244 maize hybrids genotyped with 832K polymorphic
SNPs was analysed in these experiments, together with a
characterization of phenology in a phenotyping platform. We
have first identified by GWAS genomic regions (QTLs) associated
with grain yield. Nearly all QTLs had scenario-dependent
allelic effects, with positive, negative or null effects depending
on heat/drought scenarios. Second, by combining information
coming from the field and the platform, we have estimated the
responses of yield components to environmental conditions via
regression analysis over the whole dataset. We have identified
in this way massive effects of temperature and water deficit at
flowering time, and of intercepted light during the vegetative
stage. The slopes of these responses largely differed between
hybrids and were dissected into QTLs. A model combining environmental
conditions with QTL of responses accounted for
72% of the variance of yield, including in eight new experiments.
This work, therefore, allows prediction of genotypes to
cope with climate changes. (Millet et al 2016, Plant Physiology
72:1).
Origine : Fichiers éditeurs autorisés sur une archive ouverte