research projects

Sorghum

Bioinformatics

Johnsongrass

Cactus

Cotton

Peanut

Rice

Maize

Millets and Buffelgrass

Papaya

Sugarcane

Brassica

Cross-cutting topics

Sorghum

Sorghum (Sorghum bicolor L. Moench.) is a leading cereal in arid and semi-arid agriculture, ranking fifth in importance among the worlds grain crops (Doggett 1988). Introduced into the US about 150 years ago, sorghum is grown on 10-14 million acres with a farm-gate value of $1.3-1.6 billion/yr. Sorghum is unusually tolerant of low input levels, essential in areas such as the US Southern Plains that receive too little rainfall for other grains. Increased demand for limited fresh water supplies, coupled with global climatic trends, suggest that dryland crops such as sorghum will be of growing importance.

Sorghum is in the same Poaceae tribe (Andropogoneae) as much larger-genome crops such as maize and sugarcane. Sorghum and maize (the leading US crop with a farm-gate value of $15-20 billion/yr) may have diverged from a common ancestor about 24 million years ago (Thomasson 1987) and retain similar chromosome organization (Hulbert et al 1990; Whitkus et al 1992; Paterson et al 1995a). By contrast, rice (the model monocot) and the maize/sorghum lineage may have diverged about 66 million years ago (Linder 1987), and show much more chromosomal rearrangement (cf. Paterson et al 1995a). Sorghum and sugarcane (Saccharum), arguably the worlds most valuable crop at ~$143 billion/yr. (Christou et al 1992) and with a US farm-gate value of ~$1 billion/year, may have shared a common ancestor as recently as 5 million years ago (Sobral et al 1994), retain similar gene order (Ming et al 1998), and even produce viable progeny in some intergeneric crosses (deWet et al 1976; Morrell & Paterson, in prep.).

We have made the worlds most detailed molecular map of the sorghum genome, and applied the map to gaining better understanding of many traits important to domestication and improvement of sorghum. Some present activities include:

Building a physical map of sorghum that is suitable as a framework for sequencing the sorghum genome
Taking advantage of the small and well-characterized sorghum genome to better understand the genomes of more complex and less well-characterized grasses such as sugarcane, bermudagrass, and buffelgrass
Cloning genes that are central to the domestication of sorghum, and indeed to many crops

PGML sorghum links:

Chromatin duplication in the sorghum genome.
The relationship of gene order in the maize and sorghum genomes.
WebFPC interface to our S. propinquum physical map.
Download S. propinquum FPC file.
WebFPC interface to our S. bicolor physical map.
Download S. bicolor FPC file.
BACMan db of our physical mapping data.
PGML reference database.
Cereal Crosslink Genome Center.
Sorghum genetic map information

For more information on sorghum, visit the National Grain Sorghum Producers.

[top]

Bioinformatics

Our bioinformatics activities fall into three general areas:

I. Participation in the development of on-line resources for major plant groups: We have led the development of online genomic resources for sorghum, and are participating in parallel integrated activities for cotton, maize, rice, and the legumes as cited above.

II. Development of application-specific software for data acquisition and project management: In particular, we have focused on development of software for various aspects of BAC hybridization-based physical mapping largely integrated under the BACMan umbrella.

III. Gleaning new messages from on-line resources: There exist growing opportunities to better understand angiosperm evolution based on patterns that are manifested from burgeoning genomic databases. For example, we used in-house software to integrate a phylogenetic approach to relating chromosomal duplications to the tree of life, with a genomic approach to mitigating information lost to diploidization. This showed that a genome-wide duplication postdates divergence of Arabidopsis from most dicots, that an inferred ancestral gene order for Arabidopsis reveals more synteny with other dicots (exemplified by cotton), and that additional, more ancient duplication events affect more distant taxonomic comparisons. By this approach, interpretation of complete genomic sequences benefits from partial sequence data for other taxa, in turn fostering better use of comparative approaches in many lineages to study (for example) evolutionary constraints on genome organization, molecular and phenotypic consequences of polyploidy, and the molecular basis of quantitative traits.

Python code used for dating whole genome duplication events (as described for Arabidopsis whole genome duplication analysis) is available.

[top]

Johnsongrass

A close relative of cultivated sorghum, Johnson grass (S. halepense) is one of the worlds most aggressive weeds. Johnsongrass (S. halepense) reduces yields of maize, soybean, cotton and other crops by up to 45% (Millhollen 1970; McWhorter & Hartwig 1972), by competing for resources (water, nutrients, sunlight) and providing an overwintering site for insects and diseases (Anderson 1977). Johnsongrass is a polyploid hybrid of SB and SP (cf. Paterson et al 1995b), our mapping parents and BAC library sources. In the same manner that microbial genomics is expediting drug discovery, new data about Sorghum (especially SP) growth and development may lead to new strategies for environmentally-benign plant growth regulation, suppressing weed dispersal and/or stimulating formation of dense stands of desirable forage and turf grasses.

BACMan db of our physical mapping data.

[top]

Cactus

Opuntia cacti are an especially important emerging crop, as they are far more efficient in converting water to dry matter than traditional crops due to a specialized photosynthetic pathway - CAM metabolism. A staple in Latin American and Native American cultures, Opuntia fruits (tunas) and vegetables (nopalitos) are also popular in Mediterranean and Caribbean countries and are attracting growing interest as new products for the USA. For example, the tonnage of nopalitos consumed annually in Mexico approximately equals the tonnage of cauliflower consumed in the USA. Cactus is a valuable crop in (at least) five specific ways:

Nopalitos, the young tender lobes (stems) of spineless cactus varieties, are used for fresh consumption, stir-frying, pickling, and roasting.
Tunas, mature cactus fruits, range from ~100-200 grams in mass, and about 13% in sugar content (among commercial types). A wide range of colors exist -- US consumers prefer deep purple fruit, however other countries prefer orange, green, white, or yellow.
High sugar levels are supplemented by a wide range of flavors, including watermelon, banana, peach, and others.
Both traditional and modern medicine associate cactus consumption with reduced diabetes, representing a new means to prevent diet-related diseases. Nopalitos also contain a high level of complex polysaccharide (fiber), well-established to have many salubrious effects.
An insect parasite of cactus, the cochineal (Dactylopius coccus), produces a deep red pigment (carminic acid) that was the first color-fast red dye used in Europe. Recent interest in natural dyes has driven cochineal prices to $80/kg.
Opuntia has long been a stopgap crop used as livestock forage in times of drought.

Despite 20 years of work in field testing and selecting elite clones of Opuntia, there is a nearly complete lack of knowledge of its genes, and the genetic control of its economically important traits. A comprehensive program is needed to guide germplasm evaluation and genetic improvement of this important botanical model, arid-region staple, and emerging US crop.

We are working to (1) understand the structure of the gene pool and develop strategies for effective collection, characterization and utilization of germplasm; (2) develop and implement genomic tools to accelerate crop improvement, and (3) explore the molecular/physiological basis of novel adaptations of cacti to arid conditions, both toward their further improvement, and also toward enhancing the stress resistance of other crops.

[top]

Cotton

Cotton is the worlds leading textile fiber and second most important oilseed. In recent years, cotton has been grown on about five million hectares in the USA, more than any crop except maize, wheat, or soybean (Anonymous 1999a). Texas, Georgia, California, Mississippi, and Arkansas (in descending order) lead US cotton production. The aggregate value of cotton fiber grown in the USA is typically $6 billion/yr., while cottonseed oil and meal add another $500 million/yr. More than 400,000 domestic jobs are related to cotton production and processing, with an aggregate influence of over $40 billion on the US gross domestic product (Anonymous 1999b). US cotton exports contribute about $2 billion/yr. to our trade balance. The high value of cotton per acre, however, often justifies the heavy use of inputs such as pesticides, fertilizers, water, and energy. Given that these inputs often contribute to negative ecological consequences, future gains in competitiveness of US cotton production must emphasize intrinsic genetic changes (Wilkins et al. 2000), improving productivity at minimal cost to both the grower and the ecosystem. Our research program will contribute in a significant way to a growing understanding of the genetic and genomic context of fiber development, and hence ultimately to agronomic sustainability.

We have made the worlds most detailed molecular map of the cotton genome, and applied the map to gaining better understanding of many traits important to domestication and improvement of cotton. Some present activities include:

Building a physical map of cotton that is suitable as a framework for sequencing the cotton genome (see PGML Cotton Genome Database site).
Better understanding how genetic diversity is arranged in the cotton genome, and how different measures of diversity are interrelated.
Partnering with breeders to bring new diversity into the cotton gene pool from wild relatives, taking advantage of natures solutions to many problems that face cotton producers.
Better understanding the role of polyploidy in the evolution of productive crops.

Physical mapping includes both hybridization of DNA probes to several BAC libraries and fingerprinting of G. raimondii BACs, providing >10 genome equivalent coverage of the genome. A recent update of data from NSF-sponsored fingerprinting of the G. raimondii BAC library is available here.

Please visit the site describing USDA-IFAFS funded cotton genomics activities, entitled "Reducing the genetic vulnerability of Cotton", for scientific results of our research. For information on levels and patterns of chromosomal duplication in cotton, please see our cotton duplications database. For more information on cotton, visit the Georgia Cotton Commission site.

[top]

Peanut

In the early 1900s, George Washington Carver of the Tuskegee Institute in Alabama developed more than 325 uses for cultivated peanut, Arachis hypogaea L. Largely due to his research, peanut has become the second-most important seed legume grown in the US (after soybean), with the US peanut crop having an average farm-gate value of over $1 billion in the 1990s (NASS-USDA, 2000). A valuable source of protein and oil, peanut contains mostly mono-unsaturated fat, which lowers LDL-cholesterol levels in the blood, and resveratrol that leads to improved cardiovascular health. Peanut is also a good source of folic acid, which helps prevent neural tube defects of the fetus during pregnancy. Peanut contains nearly half of the 13 essential vitamins and 35% of the essential minerals. Because of its nutritional value, peanut is being widely investigated as a key food source for astronauts during extended space missions. Georgia typically produces 40-50% of the US peanut crop worth about $500 million at the farm gate, and its production totaled nearly 1.5 billion pounds in 1999 (Ga. Agric. Stat. Svc., 2000). Texas is the second-largest producer, with about 15% of the crop. Florida produces about 5% of the US crop.

The US peanut gene pool has been through at least three major genetic bottlenecks: (1) formation of polyploid peanut perhaps as little as 5,000 years ago from two diploid ancestors; (2) sampling of a small fraction of genotypes found in its center of diversity in South America; and (3) domestication and scientific breeding. As a result, the peanut gene pool is among the most narrow, and genetically vulnerable of any US crop (Gregory et al 1980).

We have made the worlds most detailed molecular map of the peanut genome, and applied the map to gaining better understanding of many traits important to domestication and improvement of peanut. Some present activities include:

Partnering with breeders to bring new diversity into the peanut gene pool from wild relatives, taking advantage of natures solutions to many problems that face cotton producers.
Better understanding of how genes are arranged in peanut in relation to other crops.

For scientific results of our research, visit our peanut BACMan database and the Peanut Genomics Initiative. The Georgia Peanut Commission has more peanut information.

[top]

Rice

Rice (Oryza sativa) has been cultivated for more than 9,000 years and is a major food staple for over 50% of the human population. Extensive structural and functional genomics resources exist for rice including dense genetic and physical maps, EST collections, plant transformation systems, as well as extensive germplasm collections and mutant lines. Rice is considered a model system for plant biology and genetics largely due to its compact genome (430 Mb) and evolutionary relationships with other large genome cereals such as maize (2,500 Mb), barley (5000 Mb) and wheat (15,000Mb). The large size of these genomes make it likely that they will be analyzed by some form of skimming as proposed by Peterson et al (2002) or Rabinowicz et al (1999). As a consequence, such genome data would need to be anchored to the rice genome for context.

We have contributed to genetic and physical mapping of the rice genome, in particular showing that domestication of rice has proceeded in parallel with that of other major cereal crops (Paterson et al 1995; Li et al accepted), and dissecting the complex inheritance of agricultural productivity in rice (Li et al, 1997a, b; 1998; 1999a, b; 2000; 2001). Recently we have participated in the annotation of the emerging rice sequence (Wing et al, submitted), in particular showing parallels to the genome of other as-yet unsequenced grasses, and to the completed Arabidopsis sequence.

Major activities in our lab are to provide information and tools needed to leverage the rice genome and sequence into benefits in understanding the basis of agricultural productivity and quality in other grasses, and to investigate the genic and genomic distribution of diversity in rice with an eye to revealing the footprints of natural or human selection for key attributes that have contributed to the ability of this crop to sustain more humans than any other.

For scientific results of our research, see the the Arizona Genomics Institute's rice links, and our reference database.

BACMan db of our physical mapping data.

[top]

Maize

Maize is among the most important crop plants in the world, and the most important crop (based on farmgate value) in the USA. We have contributed to demonstration of the similarities and differences between the maize genome and those of other grasses at both the structural (Chittenden et al 1994; Bowers et al submitted) and functional levels (Paterson et al 1995; Lin et al 1995), and are presently participating (Draye et al 2001; Cone et al 2002) in the development of an integrated genetic and physical map as a foundation for improving the agronomic characteristics of the plant both by traditional breeding and alternative methods based upon gene isolation and engineering.

The maize genome presents a complex challenge to the development of an integrated genetic and physical map. The genome is large, approximately 2500 megabases; it is laden with numerous families of transposable elements, whose copy numbers can be in the tens of thousands; and public sequence information is limited.

The integrated genetic and physical map for maize that we are contributing to, consists of three core components. The first is a high-resolution genetic map which provides essential genetic anchor points for ordering the physical map, as well for utilizing comparative information from other smaller-genome plants (this latter part being our focus). The physical map component consists of contigs assembled from clones from three deep-coverage genomic libraries. The third core component is a set of informatics tools designed to analyze, search and display the mapping data.

For scientific results of our research, see MaizeGDB and our reference database.

BACMan db of our physical mapping data.
The relationship of gene order in the maize and sorghum genomes.

[top]

Millets & Buffelgrass

The taxa Pennisetum and Cenchrus are closely related and some taxonomists consider them a single genus (Correll and Johnson, 1970). Clayton and Renvoize (1986) placed both genera in the Cenchrinae Dumort, a subtribe of the tribe Paniceae. There are about 150 species in both genera, approximately 125 in Pennisetum and 25 in Cenchrus.

Only a limited number of the species are of economic importance and/or academic interest, including P. ciliare L. Link syn Cenchrus ciliaris L. (buffelgrass; 2n=4x=36 and 5x=45: Bashaw and Hignight 1990), P. glaucum (pearl millet), and several Pennisetum species in the tertiary gene pool of pearl millet, including P. flaccidum Griseb. in Goett., P. massiacum Stapf, P. mezianum Leeke, and P. orientale L.C. Rich. These grasses are native to Africa, the Middle East and central and south central Asia and are adapted to drier climates. Some of them are valuable forage grasses in the semi-arid regions of the world including the southwestern U.S.

Apomixis is prevalent within Pennisetum, but sexuality does exist and genetic improvement has been realized by crossing sexual and apomictic types (Bashaw 1980). Hybrids have been produced between buffelgrass and several of these Pennisetum species, as well as among the different Pennisetum species (Hussey et al., 1993). The most definitive finding regarding species relationships was from a cross between sexual P. ciliare L. Link syn C. ciliaris L. and apomictic C. setigerus Vahl (birdwoodgrass). Their chromosomes were homologous and they were reclassified as botanical varieties of the same species (Read and Bashaw, 1969).

To date, in collaboration with our collaborators at Texas A&M and the associated ARS-USDA unit, we have developed a detailed RFLP map of this cross (Jessup et al, accepted), and shown that apomixes in the cross is largely explained by a single location in the genome (Jessup et al 2002). Curiously, this location also is associated closely with the inheritance of short-day versus day neutral flowering, and plant height in other grasses (Lin et al 1995), and its further dissection is a focus of our ongoing activity.

In addition, the mapping population segregates for several other traits that are important to evolution and productivity of forage and turf grasses., that are amenable to QTL mapping in the 233-plant progeny array that is a focus of work on this cross. Specifically, progeny of this hybrid segregate for rhizomatousness, sexual vs. apomictic reproduction, and cold tolerance, hence this segregating population provides excellent germplasm for comparative mapping. The RFLP map is comprised largely of heterologous DNA markers that foster alignment to the genomes of sorghum, maize, rice, and other well-mapped taxa.

For scientific results of our research, see our reference database.

[top]

Papaya

Much of biological diversity is of tropical origin, yet the genomes of tropical trees remain an especially under-explored node of the angiosperms. In addition to their importance to fill gaps in basic knowledge, production of tropical fruit crops is one of the most rapidly growing segments of U.S. agriculture. Fruit production now ranks number one in market value, over $130 million annually, in the State of Hawaii, exceeding sugarcane (Hawaii Agriculture Statistical Service, 2000). The genomic composition of tropical fruit crops has remained underexplored in part due to narrow genetic diversity among tropical fruit germplasm collections, a serious obstacle to generating genetic linkage maps. Recent improvements in physical mapping technologies provide a viable alternative approach to genomic research on fruit crops, and the availability of burgeoning data from botanical models provides the means to focus such efforts on building a framework for comparative genomics.

An especially promising system in which to nucleate exploration of tropical tree genomes is the papaya, (Carica papaya L.). As a genomic model, papaya is appealing for a long list of reasons, including:

A small genome of 372 mbp (Arumuganathan and Earle, 1991), about 10% smaller than the rice genome, which has been largely sequenced.
Diploid inheritance, with 9 chromosomes in its gametes.
A detailed genetic map of 1,283 AFLP loci (based on 54 F2 plants) and availability of 829 recombinant F2 plants to use for fine-scale mapping as needed.
An existing 13.7 X papaya BAC library comprised of about 40,000 clones with average insert size of 132kb (Ming et al 2001).
The Caricaceae family (which includes papaya) is the closest family among flowering plants to the family Brassicaceae (which includes Arabidopsis) (Bremer et al., 1998), and the genome of Arabidopsis is completely sequenced (The Arabidopsis Genome Initiative, 2000).
A well-established transformation system (Fitch et al., 1992).

Building on prior contributions (Ming et al 2001), we seek to use comparative genomic approaches to better link tropical trees to other nodes of the angiosperms, and to foster study of parallel or convergent evolutionary processes in temperate and tropical trees, using the small genome of papaya.

For scientific results of our research, see our reference database.

BACMan db of our physical mapping data.

[top]

Sugarcane

Among the worlds leading crops with an annual production at a projected record 97 million metric tons in 1999/2000 (FAS 1999), sugarcane is a classical example of a complex autopolyploid. Cultivated sugarcane varieties have about 80 - 140 chromosomes, comprising 8 - 18 copies of a basic x=8 or x=10. Most chromosomes of cultivated sugarcane appear to be largely derived from Saccharum officinarum, however, in situ hybridization data suggest that about 10% may be derived from S. spontaneum. S. officinarum commonly has high sucrose content, low fiber content, thick stalks, little pubescence, rare flowering, and limited tillering. S. spontaneum does not accumulate sucrose, and is fibrous, thin-stalked, pubescent, profusely flowering, and abundantly tillering.

Like other vegetatively propagated plant species, cultivated sugarcane (Saccharum spp. hybrids) and its wild relatives are highly heterozygous. Pure inbred lines do not exist due to the difficulty of self pollination and the random pairing of multiple homologous chromosomes. The segregating populations used in genetic studies are the progenies (first generation) derived from crosses between two cultivated varieties or cultivated varieties and wild species. Chromosome transmission is normal for most crosses, yielding n + n progeny, but 2n + n transmission predominates in S. officinarum (2n=80) x S. spontaneum F1 and BC1 crosses, a phenomenon known as female restitution.

We have used a genetic mapping approach based on single dose restriction fragments (SDRFs: Wu et al. 1992; Da Silva et al. 1995; Guimarães et al. 1997; Ming et al. 1998) to build the most detailed (to date) genomic map available for sugarcane, to show the similarities and differences in genomic organization of sugarcane and its nearest cultivated relative (sorghum: Ming et al. 1998), and to begin to dissect the genetic control of key traits related to quality and productivity of the worlds elite sugarcane genotypes (Ming et al 2001, 2002a,b; in press, in preparation).

We are presently focusing on linking the complex sugarcane genome to those of more tractable small-genome grasses such as sorghum and rice, through combined physical mapping and informatic approaches, toward the establishment of tools and resources that foster comparative analysis of similarities and differences in the structure, function, and diversity of the worlds Poaceae crops.

For scientific results of our research, see our reference database.

BACMan db of our physical mapping data.

[top]

Brassica

Brassicas are widely used in the cuisine of many cultures, recognized as a valuable source of dietary fiber, vitamin C, and other possible salubrious factors such as anti-cancer compounds. The Brassica crops can be loosely categorized into vegetables (cole crops), oilseeds, and condiments. The genus has a Mediterranean center of diversity, is tolerant of cool temperatures but not of excessive heat, and is often grown as a winter crop in moderate climates.

B. oleracea and B. campestris comprise a large variety of vegetables in our daily diet, such as the enlarged inflorescence of cauliflower (B. oleracea subsp. botrytis) and broccoli (subsp. italica); enlarged stem of kohlrabi (subsp. gongylodes) and marrow stem kale (subsp. medullosa); enlarged root of turnip (B. rapa subsp. rapifera); enlarged and twisted leaves of Pak-choi (subsp. chinesis), collards, and Chinese cabbage (subsp. pekinesis); and enlarged apical buds of cabbage or lateral buds of Brussels sprouts (B. oleracea subsp. gemmifera) (Kalloo and Bergh 1993).

B. campestris, B. juncea, B. napus and B. carinata provide about 12% of edible vegetable oil world-wide (Labana and Gupta 1993), and generate more than $8 billion market value in the US and Europe. Canola oils have become accepted in recent years by the US consumer, however the vast majority of canola oil is produced in Canada and Europe. This is largely due to earlier acceptance of canola in these countries, and more advanced breeding efforts. Winter types of B. napus (requiring vernalization) are adaptable to the southern temperate zones of the US, and spring types (with no vernalization requirement) to the north. While improvement programs are underway in the US, much more progress will be needed to compete with other regions that pursued this crop earlier. Finally, B. nigra is primarily utilized as a condiment, mustard seed. The estimated crop value of Brassicas in the USA is about $1 billion per year. Such estimates are conservative, in that some Brassicas such as collards are cultivated primarily for local or home use, but are nonetheless a dietary mainstay and economical source of vitamins and nutrients, especially in low-income communities where out-of-season vegetables can be prohibitively expensive.

We have made several genetic maps of Brassica oleracea (see citations), and used the maps to gain better understanding of genome organization and also to identify the locations of agriculturally important traits. Our present focus is on the assembly of robust physical maps for Brassica oleracea, based on three BAC libraries and using the completed sequence of Arabidopsis thaliana as a reference point.

BACMan db of our physical mapping data.

[top]

Cross-cutting topics

Our interests in a diverse set of angiosperm taxa often permit elucidation of principles and processes that distinguish the angiosperms from many other groups of organisms.

A few of these include:

Shared ancestral gene arrangements: We have shown that similar gene orders in some genomic regions have persisted over long periods of angiosperm evolution and earlier, enhancing the reach of comparative biology and setting the stage for exploring possible fitness consequences of gene arrangement.

Parallel or convergent evolution of phenotypes: We have shown that some angiosperm lineages are remarkably conservative, most notably the Poaceae in which mutations corresponding to many of the same genes or genomic locations have been independently selected during domestication of diverse crops on different continents by different cultures.

Genetic and breeding strategies: We have contributed to comparative genomic strategies, and fine-scale genetic mapping strategies that play important roles in many applications of genetics to study of biotic diversity.

Consequences of polyploidy: We have shown that virtually all angiosperms are paleo-polyploids, and begun the process of unraveling the course of angiosperm genome evolution through multiple duplication/diploidization events.

Consequences of genome size: We have contributed to molecular dissection of the elements that are responsible for the roughly 1000-fold variation in angiosperm genome size, and also developed methods that foster the efficient discovery of angiosperm genomic diversity by fractionating genomes into components based on the degree of sequence iteration.

[top]