| My lab investigates several aspects of genetic diversity in maize, and how levels of genetic diversity affect our understanding of which genes control agronomic traits. Maize was domesticated from teosinte (Zea mays ssp. parviglumis) through a single domestication event in southern Mexico between 6,000 and 9,000 years ago. Artificial selection has impacted maize during its domestication from teosinte to landraces, and during plant breeding from landraces to modern germplasm (i.e. inbred lines). Large-scale sequencing projects indicate that approximately 2% (1,200 genes) of the ~60,000 genes of maize have been selected during domestication and/or plant breeding (improvement). Genes that have experienced artificial selection have greatly reduced genetic diversity in modern germplasm, and therefore cannot contribute to variation for agronomically important traits. Also of significance, selected genes will not be identified through conventional genetic analyses such as QTL analysis and association mapping. |
Because selected genes have greatly reduced genetic variation in modern germplasm, they will not be identified in genetic screens and will not be useful in traditional breeding programs. If selected genes are to be utilized fully, we must reintroduce variation from teosinte and/or landraces. In order to do this, our lab is creating introgression libraries from 10 teosinte and 16 landrace teosinte accessions in the B73 background, resulting in libraries of small genomic fragments of teosinte in a controlled, temperate background. These libraries will be used to 1) identify and define the biological functions of domestication and improvement genes, and thereby identify the traits that were targeted during domestication and plant breeding and 2) determine if reintroducing genetic variation at these loci affects agronomic traits, and in so doing enable a novel approach to crop improvement. One specific research project using this approach is examining the impact of artificial selection on kernel quality traits such as starch, oil, protein, and amino acid content. This novel approach to crop improvement will lead to a better understanding of the genetic bases of quantitative traits in maize.
Unselected (Neutral) Genes:
For those 98% of maize genes that have not experienced selection, there is ample genetic variation remaining in diverse inbred lines for the identification of QTL and crop improvement by plant breeding. One resource developed as part of our NSF project www.panzea.org is the Nested Association Mapping (NAM) population. NAM combines the strengths of linkage-based QTL Mapping and linkage disequilibrium (LD)-based Association Mapping into a high resolution, high powered genome scan for discovery gene-trait associations. NAM was developed by creating 25 linkage populations that would capture a large proportion of maize diversity and be useful for both linkage and association mapping. The NAM population is comprised of 5000 RILs (200 RILs from each cross between the reference parent B73 and 25 diverse inbred lines0 and has been genotyped with 1106 SNPs. NAM has the power to detect 20 QTL per trait, resolved to our LD decay limits (2000bp). Our lab is using NAM to identify genes underlying kernel quality traits such as protein, oil, and starch.