Progress 01/01/09 to 06/30/12
Outputs
OUTPUTS: This project had two major aims, using maize as the model system: 1) develop a methodology that combines association mapping with analysis of allele frequency change to dissect the genetic architecture of response to artificial selection and 2) characterize the genetic basis of quantitative disease resistance to northern leaf blight (caused by the fungal pathogen S. turcica) of maize. Project at NCSU (2007-35301-18133): A population of ~320 families was established based on the proposed genetic design in which multiple families are produced from multiple generations of a previously selected population. Two years of replicated field trials were conducted on the population and data on disease resistance, flowering time and plant and ear height was recorded. The five founder inbred lines of the population were genotyped at >300 SSR loci. A subset of the marker loci was used for preliminary genotyping of the entire population. A two-locus computer simulation model was also developed to study the methodology proposed in this project. Project after transfer to UD (this project: 2007-35301-19859): A comprehensive analysis of the phenotypic dataset was performed using advanced statistical procedures not used previously for the analysis of disease resistance. The two-locus simulation model was extended to a whole genome (mulit-locus) model to study the performance of the genetic mapping framework proposed in this project. Using the computer simulation model aspects of population structure or relatedness statistical power were investigated. A genotyping methodology was developed in concert with other projects to conduct high-density genotyping on the population from this project. Project spanning period from NCSU to UD: RELATED OUTPUT 1: Concerning the second aim of this project, a unique approach for studying the genetic basis of quantitative trait variation (generally applicable) was developed--multivariate association mapping--and used to identify specific genes associated with northern leaf blight resistance, as well as resistance to other diseases. RELATED OUTPUT 2: A new population was developed using the same experimental framework developed in this project to study the genetics of adaptation. PARTICIPANTS: Project at NCSU and UD: PD: Randall Wisser; Co-PDs: Jim Holland (USDA-ARS; NCSU) and Peter Balint-Kurti (USDA-ARS; NCSU); Collaborators: Rebecca Nelson (Cornell Univ.). As well, at Cornell Univ., Judith Kolkman (laboratory manager in R. Nelson's lab) and Jesse Poland (graduate student in R. Nelson's lab) facilitated field trials conducted in NY. Partner Organizations: University of Delaware; USDA-ARS/North Carolina State University; Cornell University, and University of Illinois Urbana-Champaign. Project at UD: Training: One research assistant, Yogasudha Veturi, was supervised on the development of statistical models for data analysis. During the project she transitioned from research assistant to masters student and completed her thesis under the project (thesis title: "Development of a statistical framework for association mapping in recurrently selected populations"). In addition, several undergraduates were trained while assisting in the collection of phenotypic data from field trials. Finally, this project led to collaborations with six other principal investigators (Nick Lauter, Jim Holland, Sherry Flint-Garcia, Natalia De Leon, Seth Murray and Wenwei Xu) and a new NIFA-funded project involving them that extends the ideas and developments of this one. TARGET AUDIENCES: Not relevant to this project. PROJECT MODIFICATIONS: 1) One of the target traits, lesion expansion, was measured in two replicates of only one field trial (NC 2007). This trait exhibited a very low correlation between replications. Because measuring lesion expansion required tremendous effort and because there was such a low correlation between replications we expected there would be little power to detect trait associations. We therefore decided to no longer measure this trait. 2) Concerns about the purity of the original population from which the experimental population of this project was produced (the impurity of the population was discovered during the project), in part, delayed some of the work. It became necessary to determine at which point in population development non-founder alleles were introduced. Fortunately, it was found that this occurred prior to selection of the original population, which was expected not to adversely affect our overall objectives; however, it did cause some delays. 3) Questions arising from our initial testing of the proposed framework prompted investigation by computer simulation. This was not originally planned but turned out to be important for evaluating the proposed framework using high density, genome-wide marker data.
Impacts
Project at NCSU (2007-35301-18133): A population was derived for this study from multiple generations of a previously selected population (Carson [2006] Phytopathology 7:910). In the NC field trials but not the NY trials a qualitatively different type of lesion segregating in a subset of the derived families was observed. The phenotype was inferred to be due to Ht1, a major gene conferring race-specific resistance to northern leaf blight. Our observations presented conflicting information from the original report on the population indicating that no major genes were present in its founder lines. Based on preliminary genotyping of the population it was determined that the original population contained alleles not present among the known founder inbred lines; therefore, the population had been partially outcrossed prior to our use of it and it is highly likely that an unknown source contributed the Ht1 and other alleles. This, in part, caused us to proceed cautiously with further investment in the population (particularly genotyping) and alternative methods were pursued to investigate the proposed approach. A computer simulation model was developed and later (at UD) expanded and used to study the proposed approach (see outcomes at UD). Project after transfer to UD (this project: 2007-35301-19859): Contemporary statistical techniques were uniquely applied to study the disease data collected in this project. This led to a deeper understanding of the performance of the population in response to disease development and the publishing of a generalized approach for studying disease resistance. With available (though not genome-wide) marker data preliminary association mapping analysis was performed. This led to questions about the statistical power of the approach, and a computer simulator was developed to model the entire, genome-wide process of population development and the sampling schema of the proposed mapping framework. It was found that association mapping did have sufficient power to detect many, but not all, of the true loci underlying response to selection. While this represents a successful demonstration of the proposed approach, it is yet to be demonstrated with real world data. This dataset is continuing to be used to achieve that outcome. Project spanning the entire period from NCSU to UD: In a separate but related effort, high-resolution association mapping led to the identification of genes associated with resistance to northern leaf blight. At NCSU, we began extending the association analysis framework to handle multiple traits (in our case, resistance to different diseases). This eventually led to the discovery that some diseases are genetically correlated (including northern leaf blight with southern leaf blight and gray leaf spot) as well as the identification of a specific gene for multiple disease resistance. Collaborations established through the use of the framework developed in this project led to a new, funded project on selection response and adaptation in maize involving the PD and six other investigators.
Publications
- Kump, K. L., P. J. Bradbury, R. J. Wisser, E. S. Buckler, A. R. Belcher, M. A. Oropeza-Rosas, J. C. Zwonitzer, S. Kresovich, M. D. McMullen, D. Ware, P. J. Balint-Kurti, and J. B. Holland (2011) Genome-wide association study of quantitative resistance to southern leaf blight in the maize nested association mapping population. Nature Genetics 43:163-168.
- Wisser, R. J., P. J. Balint-Kurti and J. B. Holland (2011) A novel genetic framework for studying response to artificial selection. Plant Genetic Resources: Characterization and Utilization 9(2):281-283.
- Wisser, R. J., J. M. Kolkman, M. E. Patzoldt, J. B. Holland, J. Yu, M. Krakowsky, R. J. Nelson, and P. J. Balint-Kurti (2011) Multivariate analysis of maize disease resistances suggests a pleiotropic genetic basis and implicates a glutathione S-transferase gene. Proc. Natl. Acad. Sci. U. S. A. 108(18):7339-7344.
- Veturi, Y., K. Kump, E. Walsh, O. Ott, J. Poland, J. M. Kolkman, P. J. Balint-Kurti, J. B. Holland, R. J. Wisser (2012) Multivariate mixed linear model analysis of longitudinal data: an information-rich statistical technique for analyzing plant disease resistance. Phytopathology 102(11) 1016-1025.
- Poland, J. A., P. J. Balint-Kurti, R. J. Wisser, R. C. Pratt, and R. J. Nelson (2009) Shades of gray: the world of quantitative disease resistance. Trends Plant Sci. 14: 21-29.
- Wisser, R. J., O. Ott, E. Walsh, J. Poland, J. Kolkman, R. Nelson, K. Kump, P. Balint-Kurti, and J. Holland (2009) Characterizing the selection response by selection and association mapping. 51st Maize Genetics Conference, St. Charles, IL.
- Wisser, R. J., J. M. Kolkman, M. Patzoldt, J. B. Holland, J. Yu, M. Krakowsky, R. J. Nelson, and P. J. Balint-Kurti (2010) Genes associated with multiple disease resistance identified through a multivariate mixed model association genetic analysis. 2nd International Symposium on Genomics of Plant Genetic Resources, Bologna, Italy.
- Wisser, R. J., Y. Veturi, E. Walsh, J. Poland, J. M. Kolkman, R. J. Nelson, O. Ott, K. Kump, P. J. Balint-Kurti, and J. B. Holland (2010) Development of an efficient genetic framework for studying genomic responses to artificial selection. 2nd International Symposium on Genomics of Plant Genetic Resources, Bologna, Italy.
- Wisser, R. J. (2010) Maize: Durable Resistance Breeding. In D. R. Heldman, D. G. Hoover, and M. B. Wheeler (Eds.), Encyclopedia of Biotechnology in Agriculture and Food. (pp. 375-380) Taylor and Francis Group LLC, New York, NY.
- Wisser, R. J., J. M. Kolkman, M. Patzoldt, J. B. Holland, J. Yu, M. Krakowsky, R. J. Nelson, and P. J. Balint-Kurti (2011) Multivariate genetic association mapping unveils a gene associated with multiple disease resistance in maize. Genetics of Maize Disease Workshop, Raleigh, NC.
|
Progress 01/01/10 to 12/31/10
Outputs
OUTPUTS: Activities: This project aims to develop an approach to identify and characterize genomic loci underlying trait variation responsible for gains from phenotypic selection. The study material is comprised of a set of ~300 families derived from a population improved over multiple generations for quantitative resistance to northern leaf blight (NLB) of maize. This project involves three key elements: (1) experimental field trials for phenotypic evaluation; (2) production of genotypic data for selection mapping and association analysis; and (3) analytical developments. (1) Phenotyping: Completed (see previous reports). (2) Genotyping: Partially completed (see previous report). The remaining genotype data is being produced during the final term of the award. (3) Analytical outputs: Statistical analysis of the phenotype data has been completed. We are now developing the statistical modeling framework that will link phenotype to genotype to complete the main objective of our project. PD Wisser supervised Yogasudha Veturi as a research assistant until she was accepted into the department's M.S. program in August, 2010. Since then, Wisser has mentored Veturi whose thesis topic is centered on statistical problems related to this project. Products: Wisser presented on this project at the 2nd International Symposium on Genomics of Plant Genetic Resources, Bologna, Italy. This led to a publication in a special issue of the journal of Plant Genetic Resources: Utilization and Characterization. Through this project, Wisser was able to establish a network of seven researchers across five institutions. This group of researchers is employing the method developed in this project to study the genetic basis of maize adaptation. In addition to being about the development of a method to study response to selection, this project simultaneously seeks to shed light on the genetic basis of quantitative disease resistance to northern leaf blight of maize. As mentioned in our 2007 report, PD Wisser used another maize population to study northern leaf blight disease resistance, as well as two other important maize diseases. Through that investigation, PD Wisser was able to identify a specific gene associated with resistance to multiple diseases, including northern leaf blight. A manuscript was submitted to the Proceedings of the National Academy of Sciences. PARTICIPANTS: Individuals: PD Wisser has been overseeing the project. Initially, PD Wisser supervised Yogasudha Vetrui while she worked on the project. Wisser is now mentoring Yogasudha Vetrui on her thesis centered on statistical problems related to this project. Co-PD/collaborators Jim Holland (NCSU), Peter Balint-Kurti (NCSU), and Rebecca Nelson (Cornell Univ.) have facilitated the PDs efforts. Partner Organizations: North Carolina State University and Cornell University. TARGET AUDIENCES: Not relevant to this project. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
Most of the outcomes at this stage in the project have to do with developing the appropriate modeling framework to test for trait-marker associations. The population and molecular marker data we have is multi-allelic. In practically all association studies in the literature, bi-allelic markers are used. We were able to figure out a way to analyze our multi-allelic data to test the significance of trait-marker associations and estimate the effects of each allele. This was an important development that will aid us in this project and in future studies that use multi-allelic information. Through our initial association analyses aiming to link phenotype to genotype, it was determined that relatedness or coancestry among the lines in the population must be taken into account when identifying trait-marker associations. This set us on the path of determining how best to estimate coancestry for our population, and for similar types of populations in general. Veturi has recently been focused on addressing this problem. To complete the genotyping phase of our project, we are monopolizing on existing marker and mapping information on maize to develop a custom marker panel that will be used to genotype the population. The marker panel is being developed to span the genome and target previously mapped quantitative trait loci for resistance to northern leaf blight. Collaborator R. J. Nelson has provided us with marker information across ~50 locations in the maize genome where northern leaf blight quantitative trait loci have been mapped. As mentioned in outputs, through this project a network of seven co-investigators was established. Extending from the work of this project, as a group, we have acquired funding to continue developing the approaches initiated in this project and will apply it in unique ways to better understand maize adaptation.
Publications
- Project Publications: Wisser, R. J., Balint-Kurti, P. J., and Holland, J. B. (2011) A novel genetic framework for studying response to artificial selection. Plant Genetic Resources doi:10.1017S1479262111000359
- Project Related Publications: Wisser, R. J. (2010) Maize: Durable Resistance Breeding. In D. R. Heldman, D. G. Hoover, and M. B. Wheeler (Eds.), Encyclopedia of Biotechnology in Agriculture and Food. (pp. 375-380) Taylor and Francis Group LLC, New York, NY.
- Project Meeting Abstracts: Wisser, R. J., Veturi, Y., Walsh, E., Poland, J., Kolkman, J. M., Nelson, R. J., Ott, O., Kump, K., Balint-Kurti, P. J., and Holland, J. B. (2010) Development of an efficient genetic framework for studying genomic responses to artificial selection. 2nd International Symposium on Genomics of Plant Genetic Resources, Bologna, Italy
- Project Related Abstracts: Wisser, R. J., Kolkman, J. M., Patzoldt, M., Holland, J. B., Yu, J., Krakowsky, M., Nelson, R. J., and Balint-Kurti, P. J. (2010) Genes associated with multiple disease resistance identified through a multivariate mixed model association genetic analysis. 2nd International Symposium on Genomics of Plant Genetic Resources, Bologna, Italy
|
Progress 01/01/09 to 12/31/09
Outputs
OUTPUTS: This project was inactive during an approximate six month period (Jan 2009-Jun 2009) during the PIs move from NCSU to Univ. Delaware while the project funds were being transferred and until a suitable candidate to continue the work was identified at UD. The primary aim of this project is to develop an approach that allows for the identification and functional characterization of genomic loci underlying trait variation responsible for gains from phenotypic selection. Our study material is comprised of a set of ~300 families derived from a population improved over multiple generations for quantitative resistance to northern leaf blight (NLB) of maize. This project involves three key developments: (1) experimental field trials for phenotypic evaluation; (2) production of genotypic data for selection mapping and association analysis; and (3) analytical developments. (1) Phenotypic analysis: Completed (see previous report) (2) Genotypic analysis: Genotypic data was produced for the 320 entries that comprise the study material. Based on the initial screening of 300 SSRs in the five founder inbred lines of the populations under study, a subset of 30 randomly distributed polymorphic SSR loci were used to genotype the 320 entries. These data are being used to develop the analytical framework that would be applied to additional markers. (3) Analytical developments: The model system under study is comprised of two populations that were previously selected for two different component phenotypes of quantitative disease resistance to NLB; one population was selected for increased latent period (iLP) and another for decreased lesion expansion rate (dLE). As a precursor to association mapping analysis, we analyzed the phenotypic data on quantitative disease resistance collected across the two separate environments (NY and NC) and two growing seasons (07 and 08). Contemporary statistical techniques were used to analyze the data from each individual environment and across all environments to gain an understanding of the extent to which genotypes were affected by the environments in which they were evaluated. In our analysis, we also compared different statistical approaches for examining disease resistance data: "the standard" approach for studying disease resistance data was compared to a newer statistical technique. Related/additional outputs: Dissemination: A new project following the developments of this proposal has been initiated (currently funded through the PIs UD startup funds and with the help of collaborators at other institutions). For this project, the PI has collaborated with researchers at Univ. Wisconsin, Iowa State Univ., Univ. of Missouri, North Carolina State Univ., and Texas A&M Univ. to conduct field trials across nine US locations (from WI to PR). The same experimental design conceived in this proposal has been applied. Analysis of this population will aid in further developing our combined association and selection mapping approach. PARTICIPANTS: Individuals: The PD, Randall Wisser, designed and carried out all experiments. PD Wisser has overseen the data analysis carried out by a new member of the Wisser lab hired for this project, Yogasudha Veturi (M.S. Statistics). Co-PD/collaborators Jim Holland (NCSU), Peter Balint-Kurti (NCSU), and Rebecca Nelson (Cornell Univ.) facilitated the PIs efforts. As well, at Cornell Univ., Judith Kolkman (laboratory manager in R. Nelson's lab) and Jesse Poland (graduate student in R. Nelson's lab) facilitated field trials conducted in NY. Partner Organizations: North Carolina State University and Cornell University. Training or professional development: Yogasudha Vetrui, who holds her M.S. degree in statistics, is being mentored by PI Wisser on this plant genetics project. TARGET AUDIENCES: Not relevant to this project. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
(1) Phenotypic analysis: Completed (2) Genotypic analysis: As previously reported, from a genotypic prescreening of a sample of the experimental population we discovered more than the expected number of five alleles. This could have been due to contamination of the originally selected population used to develop our experimental population. By comparing genotype data from different generations we have determined that the likely point of origin of contamination was prior to selection, and thus our results would not be majorly affected. With current genotyping tools, and resources for maize specifically, we are considering genotyping with a higher density of markers. As previously reported, instead of genotyping DNA pools [to estimate allele frequencies] followed by genotyping the experimental lines, we will directly genotype the experimental lines at >300 loci. The advantage of this approach over the former is that the whole genome will be scanned leading to more informative results. (3) Statistical Analysis: With our improved statistical analysis, we have made several advances. First, we examined whether our results were in agreement with the original publication that reported on the populations we derived our study material from. We found partial agreement; one of the populations (iLP) showed a significant response to selection while the other population (dLE) did not. This may pose limitations to our power to identify genomic regions underlying the selection response. Based on an analysis of our multi-environment trials, we found resistance to NLB disease development was relatively stable across environments. This suggests that inferences regarding NLB resistance that are made in this study have a broad scope. The newer statistical technique we used provided more precise estimates of the level of resistance of each line, which, in turn, will give us more power to detect quantitative differences in our association mapping analysis. It also provided richer information. The statistical approach we have applied allows one to determine the best time to rate or select on a trait scored on a temporal scale, like disease resistance. For NLB disease, we have identified an optimal time window in which disease resistance should be evaluated and are prepared to make a recommendation for this when this work is published. This information is particularly important as phenotyping projects become larger and larger as is now the trend.
Publications
- Journal Publication: Poland, J. A., P. J. Balint-Kurti, R. J. Wisser, R. C. Pratt, and R. J. Nelson (2009) Shades of gray: the world of quantitative disease resistance. Trends Plant Sci. 14: 21-29.
- Book Chapter: Wisser, R. J. (2009) Breeding for durable resistance in maize. Chapter in Encyclopedia of Biotechnology in Agriculture and Food. Taylor and Francis Group LLC, New York, NY. Accepted for publication.
- Conference Abstract: Wisser, R. J., O. Ott, E. Walsh, J. Poland, J. Kolkman, R. Nelson, K. Kump, P. Balint-Kurti, and J. Holland, 2009, Characterizing the selection response by selection and association mapping. 51st Maize Genetics Conference, St. Charles, IL
|
|