Genome-wide association analysis (GWAS)
A diversity panel, consisting of 220 B. napus inbred lines was used for the GWAS in this study. The accessions were grown using a randomized complete block design with two replicates. Each accession was grown in plots with four rows. For the phenotypic data about low temperature resistance of rapeseed population showed a nearly skewed distribution. For more details about field experiment design and the distribution of phenotype were described in our previous reports (Xu et al., 2021). More specifically, the population included winter, semi-winter and spring oilseed rape. The accessions originated from Asia, Europe, and North America. The B. napus population was grown at the Farm Station in Yangling, Shaanxi province, China (107°59′–108°08′E; 34°14′–34°20′N), where the local temperature ranged from −15 to 10 °C in winter (https://lishi.tianqi.com/xianyang/202101.html). The reference genome of B. napus “Darmor” (version 4.1) was used to call variants, which was downloaded from BRAD (http://brassicadb.cn/#/) (Chalhoub et al., 2014). Low-temperature resistance was measured in the field in January of 2017 and 2018. The grade of low-temperature resistance was evaluated with criteria including grade 1 (the plants are normal, and the leaves are green), grade 2 (a small part of leaves are necrotic), grade 3 (a half of leaves are necrotic), grade 4 (most of leaves are necrotic), and grade 5 (all leaves are necrotic). The association analysis for low-temperature resistance was carried out using a MLM with the software package TASSEL 5 (https://www.softpedia.com/get/Science-CAD/TASSEL.shtml). The population structure was calculated using PCA. The software PopLDdecay (https://github.com/BGI-shenzhen/PopLDdecay) was used to calculate LD. The parameter r2 =0.2 was read as LD length. The threshold of the GWAS was set to P -value < 10–6 .