Genomic Variation Of A Diverse Synthetic Population In Maize

The kinds of variation located within the genome largely contributes to phenotypic diversity. Thus it is necessary to expound those variants which connected with various phenotype, the variation itself is indeed a part of field with general interest. In summary, the large-scale identification variations distributed in the genome can not only help us to obtain a deeper understanding of a specie, but also provide abundant molecular markers for both genetics and genomics.With the rapid development of bioinformatic methodology and technology during last decades, next-generation sequencing strategy is becoming frequently-used in detecting molecular markers genome-wide with its obvious advantages of high-throughput, highly reliability, high efficiency and low cost, which provides significant support in genetics improvement of crop molecular breeding. Maize is the most important crop all over the world, also plays as considerable roles in economy and energy, which has been widely planted with its high production. Further more, the intrinsic diversity and complexity of maize genome makes it to be one of the model in genetics and genomic research.In present research, we studied on the maize multi-parent advanced generation inter-cross(MAGIC) population, which consists of 1404 offsprings derived from 24 elite inbreds. A variety of sequencing strategies, including whole genome sequencing(WGS), Genotyping-by-Sequencing(GBS) and high density SNP array, were integrately applied to explore polymorphosms and to finally develop the high-density maize variant map. Because all the parents used are important inbreds being widely employed in many breeding programs of China, a number of novel sequences could be potentially obtained, some of which might be further possible contributes excellent agronomic performance. Thus, the novel sequences compared to the B73 reference genome were assembled to achieve the pan-genome of maize. Facing with such abundant sequencing resources, I have mainly worked on following three parts:A). The first is experimental one, including improving the GBS pipeline, extracting DNA and constructing the sequencing libraries of nearly 1600 lines with high-quality to meet the kinds of requirements. Now, we have constructed the sequencing libraries and identified the genotypes of 279 inbred lines in our MAGIC population. And 653876 SNP locus were obtained.B). SNP calling and cross validation. All the 1404 individuals of the population were sequenced through whole genome re-sequencing in ~1X depth, while the 24 parents were sequenced in ~ 11 X depths. By using a regular workflow, which combining GATK and SAMTools in variant identification, we finally obtained 53.3 M SNPs. Additionally, 258 samples among the population were genotyped by 200 K SNP chip, and 56423 high-resolution SNPs were further gained. Through cross validation of results from re-sequencing to GBS and SNP chip, the concordance rates of genotyping were 99 % and 98 %, respectively, which proved the highly accuracy of SNP calling.C). Maize Pan-genome construction. The novel reads that unmapped to the reference genome for the whole population were extracted to de novo assembly. Totally 367 M and 600 M of novel sequences were obtained from 24 parents and the whole population, respectively. Further specific statistics of the features for novel sequences were evaluated.