Development of tools to enable high-throughput elemental analysis and their application to soybean mutant identification and genome wide association studies

Gregory R Ziegler, Purdue University


We have developed a high-throughput ionomics pipeline that can quantify the concentrations of 20 elements in more than 1700 samples a week. Requirements for high-quality, noise-free data for downstream analyses necessitated the development of automated programs to collate, organize, and visualize datasets so that inconsistencies could be found in a time-sensitive manner. We demonstrate the power of the pipeline with two case studies. First, we analyze a mutagenized population of field-grown soybean samples. We show that 1) we can control for field variation, 2) we can identify ionomic mutants by visual inspection of z-score plots, and 3) we can computationally detect ionomic mutants. Second, to broaden our understanding of how genetic and environmental components affect the ionome, we analyze a diverse set of more than 1600 soybean lines, divided into 14 independent populations grown in three locations over the course of a decade. Coupled with a high-resolution genetic map, we perform a genome wide association study (GWAS) using a multi-locus mixed model procedure. To analyze the GWAS results, we develop an interactive browser that allows for the fast comparison and analysis of the 384 GWAS experiments performed. We detect 9 significant associations in two or more locations and, using the browser we developed, are able to quickly find genes known to be involved in metal transport for 4 of the 9 loci.




Gribskov, Purdue University.

Subject Area

Genetics|Computer science

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server