Construction of a core collection of tomato ( Solanum lycopersicum) germplasm based on phenotypic traits and SNP markers

In order to better preserve and utilize tomato germplasm resources, a total of 484 tomato cultivated accessions were used to construct a core collection based on phenotypic traits and single-nucleotide polymorphisms (SNPs). First, 32 phenotypic traits were investigated and the genetic diversity of all accessions were analyzed. The results showed that the entire population has high genetic diversity, and there is correlation between most quantitative traits. According to two genetic distances, six sampling proportions, three sampling methods and eight clustering methods were used to construct core collections. The core collection with each strategy was evaluated by using four genetic parameters: the percentage of mean difference (MD), the percentage of variance difference (VD), the coincidence rate of range difference (CR), and the variation rate of coefficient of variation (VR). The best construction strategy was "Mahalanobis distance + 10 % + Preferred sampling + Weighted pair-group average method". In addition, all accessions were genotyped using 48 pairs of highly polymorphic primers, and population structure analysis showed that all accessions were divided into four subpopulations. The major allele frequency (MAF) and polymorphic information content (PIC) were calculated and compared, and it was found that the core collection constructed at a sampling ratio of 20 % was the most representative core collection. Finally, the core collection was constructed using the Core Hunter package of R 4.3.3. We constructed a core collection of 137 accessions using phenotypic traits and SNP markers, which provided an effective basis for the conservation and utilization of tomato germplasm resources.