HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Search for Related Content Related Collections Vadose Zone Processes and Chemical Transport

SYMPOSIUM ON GENOMICS AND PLANT BREEDING: THE EXPERIENCE OF THE INITIATIVE FOR FUTURE AGRICULTURAL AND FOOD SYSTEMS

SYMPOSIUM DISCUSSION

	INTRODUCTION

TOP INTRODUCTION Specific Genomics Applications... Invest in Genomics? Or... Training and Funding for...

	Specific Genomics Applications and Plant Breeding

TOP INTRODUCTION Specific Genomics Applications... Invest in Genomics? Or... Training and Funding for...

 
Understanding and Managing Linkage Blocks

1. Genomics may allow breeders to better manage favorable or unfavorable gene linkages. Over time, selection builds up linkage disequilibrium and creates both favorable and unfavorable linkage blocks. Previously undetected regions of the genome which seem to have no effect may be linkage blocks of mixed favorable and unfavorable alleles. Genomics information will allow breeders to locate these linkage blocks and possibly provide methods for enhancing recombination in those regions. Once there is recombination, breeders can select against individual unfavorable alleles and begin to accumulate new combinations of favorable alleles.
   
2. Knowledge from molecular markers has already made this possible with some simply inherited traits. For example in Capsicum, tight linkage between an unfavorable allele conferring small fruit weight and a useful allele of disease resistance was broken. To find resistant recombinants without the small fruit gene using conventional plant breeding would have required a larger population size than any program could manage. Using molecular information, an unfavorable linkage that had not been resolved for decades was accomplished in a relatively small population. Another example is the hardness locus in wheat where several duplicated genes of the same family are closely linked. Molecular information allowed the breeders to locate the alleles and identify their individual effects. It is now possible to develop spring wheat germplasm with completely new textural qualities.
   

In Quest of the Perfect Marker

1. Accuracy of selection for desired traits is a central problem in plant breeding. The perfect marker would allow breeders to maximize recombination between desired and undesired genes without fear of losing the linkage between the desired trait and marker, by tracking the alleles of the gene itself. Most examples of such perfect markers available today may represent simply inherited traits, such as pungency in peppers. For some major genes, results from a one-time screening to identify the gene can be used in breeding in many different situations. Such examples are limited, and the widespread use of markers will need greater genomic, i.e., haplotypic information. Some such markers now exist, and are available to the public. For example, EST databases comprised of more than one genotype routinely provide sequence information for SSRs and SNPs. Once these markers are associated with a phenotype, nucleotide sequence divergence (i.e., polymorphism) among different alleles provides a way to track specific alleles within pedigrees and populations. In addition, these polymorphisms may yield insights into gene function. Some SSRs have been experimentally associated with a favorable phenotype
   
2. Public availability of these markers means that speed of transfer of the information to breeding programs will be the only competitive difference among countries.
   

Identification and Cloning of Candidate Genes: Comparative vs. Direct Genomics

1. Growing databases of finished genomic sequence from model species, and partial sequence for important crops, will be used in creative new ways that add information to each. For example, extrapolation of linkage order to more distantly related species may facilitate map-based cloning of genes in new crops, or in "orphan" crops, where there is little or no previous research.
   
2. Candidate genes can be inferred from a closely related model system, through the use of comparative linkage analyses of genes known to affect the trait in the model system. Comparative genomic analysis can identify candidate genes by predicting their location in a crop's genome. Candidate genes for economically important traits in crop plants are potentially useful in plant breeding. Candidate genes can be useful even when a gene's function is unknown, or lacking detailed information about species-specific compounds.
   
3. However, genetic variation in economically important species may not be conditioned by the same loci as in the model species. Moreover, many economically important crop traits are unique to the crop and not present in model species. For these traits, candidate gene prediction from model species will not be possible. Direct analysis of crop genomes will be essential to elucidate unique features such as the fibers of cotton and the underground pegs of peanuts.
   
4. In addition to research on expressed sequences, genomics research will provide information on regulatory elements, introns, matrix attachment regions, and other genomic features that affect gene expression, thus providing information which may be useful in transgenic research as a source for tissue-specific promoters and in plant breeding as polymorphic molecular markers.
   

Understanding and Using Complex Genetic Variation

1. The known horizons of heritable variation have broadened during the last several years on the basis of genomics information from DNA sequence variation. Each individual gene is part of a complex metabolic network comprised of structural genes, transcription factors, and repressors. Regulatory genes are more difficult than structural genes to identify, even with phenotype and sequence data.
   
2. The value of such new variation will vary from one genotype to another, e.g., in hybrids of widely differing yield potential. Like any trait currently manipulated by breeders, traits from genomics research will require re-evaluation in different genetic backgrounds and environments. Genomics-assisted plant breeding, particularly for more complex traits such as yield and adaptation, will require appropriate quantitative analyses that integrates information from genomic sequence to crop phenotype. This will require bioinformatics to integrate all relevant information from genomics to phenotype, such as is being done with The Arabidopsis Information Resource, but also across species, to leverage information from model species to crop species.
   
3. Integration of genomic and phenotypic information requires, in addition, a strong framework of knowledge at the whole plant and system level. For example, it may be revealed that too little is known about whole plant physiology and plant–soil relationships. Also needed are better estimates of phenotypes for traits that underlie yield. The individual contribution of yield components to the overall phenotype—the goal of every breeding program—has not been adequately modeled in a genomics context.
   

	Invest in Genomics? Or Invest in Plant Breeding?

TOP INTRODUCTION Specific Genomics Applications... Invest in Genomics? Or... Training and Funding for...

 

1. Would plant breeding be further advanced today if the same amount of money had been spent on population improvement as on hybrid development in maize? Will the future be asking, "If the same amount of money had been spent on traditional plant breeding instead of genomics, would we be ahead of where we are now?" Particularly in developing countries, where needs are urgent, relatively less systematic breeding has been done and small investments in breeding can provide rapid gains in even a few years. It is a special concern, therefore, if international research centers, whose mandate is to assist the poorest countries, invest in genomics at the expense of plant breeding.
   
2. In 40 years, entire genome sequences may exist for most economically important crops. However, sequences per se do not improve yield per acre and resistance to major crop pests. To date, genetic improvements in domesticated crops have used empirically obtained phenotypic data, pedigree information, and selection. Prediction methods based on data routinely collected by plant breeding programs have enhanced the power of selection, and genetic gains from application of statistical models based on phenotypic and pedigree data are far from exhausted. In comparison to plant genomic projects, statistical approaches may provide quick and cost-effective advances in plant breeding methodology.
   
3. The products based on molecular techniques now available, such as transgenic varieties having a Bt gene, are simple improvements of high performance genotypes. The Bt lines were developed in 2 to 3 yr of backcrossing laboratory lines to elite parental lines. Each of these elite lines had been de-veloped though decades of traditional plant breeding. With additional basic research, plant genomics could allow breeders to move beyond such simple improvements, to the manipulation of fundamental gene networks within high performance genotypes. So applied, plant genomic research could in theory extend the biological limits of traditional plant breeding. This will be a long-term effort—one participant estimated at least 40 yr. It may become increasingly important if transgenic approaches become unavailable because of lack of public acceptance.
   
4. The hybrid maize trade-off may not be an accurate representation of history. Population improvement schemes for maize were developed about 20 yr after hybrid breeding, and in fact, much of the effort to develop today's excellent theory and practice of population improvement was inspired by the early work and difficulties in breeding hybrid maize. Applying genomics to plant breeding may be similarly heuristic. Breeding hybrid corn once looked as weird, strange, and exciting as genomics does now.
   

	Training and Funding for Genomics-Assisted Plant Breeding

TOP INTRODUCTION Specific Genomics Applications... Invest in Genomics? Or... Training and Funding for...

 
Training

1. Connection between laboratory and field is essential for integration of genomics and breeding. Field observation and breeding experience may be the only way to identify the right questions, even with a plethora of genomic information at hand. However, extensive panelist experience reading research proposals presented in recent years to the National Science Foundation and the CSREES USDA National Research Initiative reveal that lab-to-field connections have not generally been in place in these proposals. Knowledge of basic agronomy and biology, population genetics, statistics, and experimental design are critical for analyzing field data; incorporating data provided by molecular markers and genomics; and then relating all to a plant phenotype. Plant breeding students are required to study statistics, plant physiology, molecular biology, and other disciplines; however, it is difficult to find a reciprocal: a molecular genetics student with enough statistical training to study quantitative plant breeding.
   
2. A costly outcome of reductions to programs at public universities is the loss of an educational environment that provides hands-on breeding experience. Only a handful of public universities exist with the knowledge depth and the operational field programs required to educate future plant breeders. Ironically, this plant breeding experience is in high demand in the current job market. Plant breeding students are often hired before they graduate.
   

Funding

1. Compared with genomics, plant breeding is inexpensive. Operational funds of $40000 to $50000 will sustain a medium-sized program for a year. However, in most universities today, a plant breeding program is more likely to be billed than funded. Many universities cannot fill a research position unless there are opportunities for the researcher to obtain research grants. For plant breeding, there is "nowhere to go" for grant funds. Some plant breeding is "piggybacked" on genomics research grants. Traditional resources for classical plant breeding are continuously eroded because of depressed domestic commodity prices and declining political support as the number of growers declines. These circumstances will result in reduced numbers of field breeders. Without them as cooperators, workers in genomics will be unable to apply their new data and knowledge to plant breeding.
   
2. To add value, genomics must help create varieties that will be recognized and popular with consumers. Given that novelty is in demand, if public plant breeders had pursued the Baye-Dohl Act of 1980, would the resulting intellectual property rights on varieties (mostly Plant Variety Protection certificates) have provided an income stream for public plant breeding today?
   
3. However, because of their diverse needs, large numbers, and the uncertainty of continuing markets for any one specialty crop, how much effort can breeders, private or public, put into breeding for high-value-niche crops with small markets but relatively high margin for farmers? As fashions change, demand for specialty crops may come and go. Creation of novel types of traditional commodities, such as new classes of wheat for new products, may be a more feasible investment.
   
4. With regard to breeding cultivars for crops and environments that do not generate sufficient return on investment for the private sector (the public goods versus private goods conundrum): Who is in responsible for identifying the crops that are public goods and then providing public monies to support their breeding? Financial support from legislatures and granting agencies reflects priorities established by input from both scientific communities and the public. By competing for the attention of decision-makers, rather than presenting a coordinated message, the scientific community may bear part of the responsibility for not bringing money into plant breeding.
   

This Article Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Search for Related Content Related Collections Vadose Zone Processes and Chemical Transport