Population Structure And Genetic Diversity Of Foxtail Millet Landraces In China Assessed By Microsatellites

Foxtail millet is one of the most important crops in China, and plays a key role in Chinese agriculture. Assessment of genetic diversity and study of genetic relationships among germplasm are fundamentally important in foxtail millet breeding. On the other hand, there are numerous foxtail millet landraces in China, which exsit abundant genetic variations for favorable alleles in breeding. The genetic variations contained in foxtail millet landraces provided effective potential approach for the germplasm enhancement, improvement, as well as innovation of foxtail millet in China. In this study, we analyzed the genetic diversity and population structure of foxtail millet landraces by SSR marker through auto fluorescence protocol. The main results obtained were as follows:1. Seventy-seven SSR loci distributed equally in the foxtail millet genome were used to evaluate the genetic diversity of the 262 foxtail millet landraces. A total of 1281 alleles were detected, the average of alleles per locus was 24, and the average gene diversity was 0.8648, the average PIC was 0.8508, and the average Shannon’s diversity index was 2.4525. All the Parameters indicated a rather high level of genetic diversity of foxtail millet landraces.2. In this study, two kinds of methods including the UPGMA clustering algorithm and the NJ algorithm based on genetic distance were used to analyze the population structure; two methods obtained basically the same clustering results. Based on the results STRUCTURE software presented, the local varieties of the 262 foxtail millet are divided into six groups. GroupⅠwhich includes 33 samples, mainly from the Heilongjiang in Northeast China Plain. GroupⅡincluding 31 lines mainly from Huainan areas. GroupⅢ, including 47 landraces, mainly from the inland Northwest and Inner Mongolia plateau. GroupⅣincludes only 16 samples, mainly from Tibet and abroad. GroupⅤincluding 53 lines, mostly from the Loess Plateau. The number of lines in GroupⅥis the largest, including 82 local varieties of foxtail millet, mainly from eastern North China Plain, such as Liaoning, Jilin, Hebei, Beijing, Tianjin, Henan, Shandong and Jiangsu provinces, and the North China Plain and the adjacent regions of other provinces.3. Comparisons of genetic diversity among different populations revealed that Pop6 possessed relatively higher level of genetic diversity and have the largest number of alleles per locus, the genetic diversity of Pop3 is similar to that of Pop6; The genetic diversity of Pop4 is the loweat in the six populations.Although the six groups are different in sample size, the average levels of genetic diversity among them are relatively similar.4. By calculating the genetic distance and the similarity between populations we found that Group 3 and Group 5 have the smallest genetic distance, and the Fst between them is also the smallest, indicating that they are the most related groups. While Group 2 and Group 4 have the largest genetic distance, Group 1 and Group 6 have the largest Fst value.5. Analysis of molecular variance (AMOVA) showed that 58.06% of the genetic variation comes from among populations,41.94%of genetic variation comes form within groups. Therefore, the main variation comes among populations.6. In the survey of 12 agronomic traits, the highest genetic diversity index comes from plant height, which is 2.0873; followed by stem length, which is 2.0749; the lowest is ear elastic, which is 1.0033. The distributions of our characristics including leaf sheath color, ear elastic, ear type and length of bristles, are not even; always focus on performance in a feature.7. Through correlation analysis of 12 agronomic traits, we found that there were 30 pairs of traits reach significant level, accounting for 45.5%. Among positive correlation, the ear weight and ear weight have the maximum correlation coefficient,0.900; the smallest correlation coefficient is between spike length and spike type,0, performance no relevant. In the negative, spike tightness and ear diameter have the biggest negative correlation, the correlation coefficient was-0.329. Correlation coefficient which distributed-0.300-0.300 account 83.3% of the total, indicating that although some agronomic traits were significant correlations, but most of the correlation between traits is very low.