Chickpea, a Common Source of

Chapter7Chickpea,a Common Source of Proteinand Starch in the Semi-Arid TropicsFred J.Muehlbauer and P.N.RajeshAbstract Chickpea(Cicer arietinum L.)is an important food crop in the semi-arid tropics where it is grown during the cool winter season.Research has concentrated on the development of improved germplasm for resistance to diseases and pests and more recently has focused on the use of genetics and biotechnological tools to enhance the knowledge of the genomics of chickpea.The wild species(8annuals and34perennials)are a potential source of genes for overcoming problems of dis-eases and pests,and work is underway toward overcoming barriers to interspecific hybridization.Bacterial artificial chromosome libraries are available for genomic research in chickpea and a targeted induced local lesion in genomes,also called TILLING,platform is under development that holds promise for identification of important genes and determination of their function.The chickpea plant is described and the tools for further exploitation of the crop are discussed.7.1IntroductionChickpea had its origin in the Near East Arc,where it was domesticated along with other pulses including lentil(Lens culinaris Medik.)and peas(Pisum sativum L.) over7,000years ago(Ladizinsky1975).After domestication,chickpea along with other pulses and cereals formed the basis of early agriculture in the Mediterranean and West Asian regions.Chickpea soon spread south to Ethiopia and east to South Asia where it became an important and popular legume food crop and remains so to the present time.Chickpea apparently was taken to the Americas soon after the dis-covery of the New World and became an important food crop in the Pacific coastal regions of North,Central,and South America.Chickpea has been divided into two broad groupings,“Kabuli”and“Desi,”based on size,shape,and coloration of the seeds.Kabuli types generally have large seeds that weigh in excess of26grams per100seeds,are rounded,and white or cream colored.Often the Kabuli seed type is referred to as“Ramshead.”Plants of Kabuli F.J.MuehlbauerUSDA-ARS,303Johnson Hall,Washington State University,Pullman,WA99164,USAe-mail:muehlbau@P.H.Moore,R.Ming(eds.),Genomics of Tropical Crop Plants171 C Springer2008172 F.J.Muehlbauer,P.N.Rajesh types are tall and are devoid of purple or violet pigmentation of the leaves,stems,flowers,and pods.Desi types generally have seeds that are less than26grams per 100seeds and have a rough,angular appearance.Coloration can vary from light tan to black with all gradations in between.Seeds are often mottled and speckled. Desi types are typically produced in South Asia where they represent the majority of production;while Kabuli types dominate the production in most other regions and especially in the western hemisphere.Even though Kabuli and Desi types are generally distinguished by seed size and seed coloration,there are some Kabuli types smaller than26grams per100seeds and some Desi types that are larger than 26grams per100seeds.Chickpeas are branched,spreading annuals that have considerable variation in form,with some being semi-erect with a main stem and few branches while oth-ers can be relatively prostrate with profuse branching.Germination of the seeds is hypogeal and the plants develop a strong,well-developed tap root with numerous lateral roots that are nodulated under normal conditions.Plant height can range from 30to60cm depending on growing conditions.Leaves can be either pinnately com-pound with3-8leaflet pairs along a central rachis or simple(unifoliolate)with a single leaf lamina.The self pollinatingflowers,usually0.6-1.3cm long,are borne as singles,doubles,or triples on inflorescences that originate from the stem axes. Chickpea is a quantitative“long-day”plant butflowers in all photoperiods(Smith-son et al.1985).Flower color of Kabuli types is usually white;whileflower color of Desi types can be white,pink,purple,or blue.Pod shape is rhomboid ellipsoid and about2.0-5.0cm long and1.5-2.0cm wide with glandular pubescence.Pods have an inflated appearance and usually contain one or two seeds,but rarely more than two.Seedcoat colors vary from white to cream to tan for the Kabuli types,while Desi type seeds can be light tan,green,dark brown or black with many variations of mottling and other forms of pigmentation.The anterior of the seeds of both types are“beaked.”The crop is grown as a cool weather crop in semi-arid regions and as a winter crop in many production regions,particularly in South Asia,the Mediterranean, Australia,and Mexico and other areas where weather permits.Though sensitive to cold,some cultivars can tolerate air temperatures as low as−9.5◦C.Very early maturing cultivars will complete their life cycle in as little as65days,while later maturing cultivars may require in excess of120days.Winter chickpea may require up to180days from planting to maturity.7.1.1Economic,Agronomic and Societal Importance of Chickpea Chickpea,also known as“Garbanzo,”is the third most important pulse crop in the world after beans(Phaseolus vulgaris L.)and peas(Pisum sativum L.)and is a major source of protein for human food in semi-arid tropical regions(FAO2006).By far the majority of the world’s chickpea crop is grown in South Asia with India being the largest producer with an estimated production of6.0million metric tonnes(mmt)7Chickpea,a Common Source of Protein and Starch in the Semi-Arid Tropics173 annually and accounting for over half of world production.Other major producing countries include Pakistan(0.87mmt),Turkey(0.61mmt),Iran(0.31mmt),Mex-ico(0.24mmt),Myanmar(0.23mmt),Australia(0.19mmt),Ethiopia(0.14mmt), Canada(0.98mmt),Syria(0.55mmt),Morocco(0.42mmt),and numerous coun-tries that produce smaller amounts(FAO2006).Production in the U.S is currently estimated at80,000mt.Of the producing countries,Canada,Australia,Turkey,and the United States are major exporters.Chickpea is an annual legume that is grown from seeds that are either broadcast on the soil surface and incorporated by tillage or planted using grain drills.Seeding rates vary from25to120kilograms per hectare depending on soil conditions and method of planting.Chickpea can be cultivated as a sole crop or in mixtures with numerous other crops(Van der Maesen1972).Originally,it was considered that strains of Rhizobium leguminsarum nodulated the roots of chickpea;however,later evidence indicated that strains of R.ciceri were the active bacteria(Nour et al.1994).Nitrogenfixation through symbiosis can account for a significant portion of the nitrogen needs of the chickpea crop and can be achieved with good nodulation.Estimates of nitrogenfixation by chickpea crops vary considerably with soil type,climatic conditions,and residual soil nitrogen con-tent.It is,therefore,difficult to make predictions for expected nitrogenfixation.In traditional farming systems chickpea is harvested at maturity or near maturity and allowed to dry completely prior to threshing.Chaff is separated from the seed by winnowing.The crop is generally harvested in developed countries by direct combining when the moisture content of the seeds is less than15%.Chickpea is primarily a food crop used in various forms.In India the crop is consumed primarily as dhal,a preparation produced by decorticating the seed and separating the cotyledons.The decorticated and split cotyledons are then used to produce a thick soup that is generally served with rice.Chickpea is also used as a whole pulse and is soaked and boiled.“Chole”is a traditional dish made from whole chickpeas in India.A popular use in the Middle East is as hummus,a dish made from cooked and ground chickpea that is mixed with tahini,olive oil,and various spices and eaten with traditionalflat bread.Dry roasted chickpea seasoned with various spices is a popular snack in most countries of the Middle East and North Africa.In India,the young leaves of chickpea are often harvested green and cooked to make a dish similar to spinach.7.1.2Chickpea as an Experimental OrganismChickpea is a self-pollinating diploid annual with2n=16chromosomes.Natural outcrossing is estimated at less than1%.Procedures for crossing of chickpea are similar to that used for other cool season legumes and require careful handling of theflowers and good sources of pollen for success.The advantage of chickpea as an experimental organism is its relatively short seed-to-seed reproductive cycle,usually less than4months under normalfield conditions or in controlled environments.The174 F.J.Muehlbauer,P.N.Rajesh highly self-pollinating nature of chickpea lends the species to development of re-combinant inbred line(RIL)populations for genetic mapping that can be repeatedly and accurately phenotyped for important traits necessary for precise genetic linkage analysis.The Cicer genus comprises nine annual species and34perennials.The annuals can be placed into three groupings.Thefirst includes C.arietinum,the only cul-tivated species,C.reticulatum,the presumed progenitor,and C.echinospermum. Within this group,C.reticulatum is completely cross compatible with cultivated C. arietinum;while C.echinospermum can be crossed with the cultivated species,but with a high degree of sterility in the hybrids and hybrid progenies.The second group includes C.bijugum,C.pinnatifidum,and C.judaicum.There is no verified evidence that species of this group can be crossed to the cultivated species.There have been attempts to facilitate intercrossing of C.arietinum with C.bijugum through the use of embryo rescue;however,success has yet to be reported and confirmed.A similar approach has been used in attempts to overcome barriers to crossing C.arietinum with C.pinnatifidum.In both cases,weak albino plants have been obtained,but no seeds have been reported from those crosses.C.chorassanicum,C.yamashitae,and C.cuneatum comprise the third grouping and are considered to be the most distant of the annual species from cultivated C.arietinum.A very limited amount of genetic polymorphism for molecular markers in chick-pea has been a distinct handicap in the development of genetic linkage maps in these plants.Initially this handicap was circumvented through the use of interspecific hybrid populations.In populations from those crosses,genetic polymorphism for molecular markers was considerably greater than what was found in intraspecific crosses within cultivated chickpea.While the initial genetic maps of chickpea were developed from interspecific cross populations,more recent mapping efforts have concentrated on the use of intraspecific cross populations and the use of sequence tagged microsatellite site(STMS)markers.7.2Genetic Mapping and Tagging in ChickpeaAn initial genetic map of chickpea by Simon and Muehlbauer(1997)comprised a limited number of restriction fragment length polymorphism(RFLP)and random amplified polymorphic DNA(RAPD)markers and showed some similarity to the genetic maps of pea and lentil indicating conservation of linkage ter,a map developed by Santra et al.(2000)comprised mostly RAPD markers and had an estimated size of982cM.More recent genetic maps were developed using STMS markers and interspe-cific crosses that had a greater degree of polymorphism(Huttel et al.1999;Winter et al.2000;Cho et al.2002;Pfaff and Kahl2003;Flandez-Galvez et al.2003). These maps have included important genes for resistance to Fusarium wilt,caused by Fusarium oxysporum f.sp.ciceri.STMS markers have proven useful for mapping within intraspecific crosses and have resulted in more densely populated maps.7Chickpea,a Common Source of Protein and Starch in the Semi-Arid Tropics175 The STMS markers and mapping populations were also used to determine the locations of the quantitative trait loci(QTLs)for resistance to Ascochyta blight, caused by Ascochyta rabiei,considered to be the most devastating disease of chick-pea worldwide.Further refinements of the chickpea map and development of a consensus map of the chickpea genome is currently underway using the numerous genetic maps of chickpea that are available and common markers.At the present time there are more linkage groups than the haploid chromosome number,a prob-lem that will likely be overcome through increased marker density on the chickpea genetic map and through development of a consensus map of the chickpea genome.Gene tagging for eventual use in marker-assisted selection(MAS)has been a long-term goal of chickpea research programs.To date,the genes for resistance to Fusarium wilt(caused by several races of Fusarium oxysporum)have been located on the chickpea linkage map and associated with closely linked molecular markers (Winter et al.2000).QTLs for Ascochyta blight resistance have been located on the chickpea map(Santra et al.2000;Tekeoglu et al.2002)and have formed the basis for further study of resistance mechanisms(Cho and Muehlbauer2004;Cho et al.2004).7.3Sequence and Marker DiversityThe chickpea genome is considered homogeneous based on the minimal polymor-phism for molecular markers.The lack of polymorphism may be a consequence of relatively recent domestication and its self-pollinating reproductive cycle.Hence, the identification of molecular polymorphism is difficult and time consuming.To overcome this problem,researchers have used interspecific crosses that have an in-creased frequency of molecular polymorphism and highly variable simple sequence repeat(SSR)markers.SSR markers in chickpea were identified as being highly polymorphic compared to other types of markers.There are six published linkage maps based on either intra-and interspecific crosses that are populated with mor-phological markers and various types of molecular markers including:isozymes, amplified fragment length polymorphism(AFLP),RAPD,inter simple sequence repeat(ISSR),resistant gene analog(RGA),and STMS markers(Santra et al.2000; Winter et al.2000;Cho et al.2002;Collard et al.2003;Flandez-Galvez et al.2003; Cho et al.2004).In spite of the availability of SSR markers,they are insufficient in number to generate a high density linkage map.However,progress in genomic research in chickpea including bacterial artificial chromosome(BAC)end sequenc-ing and expressed sequence tags(ESTs)will provide additional opportunities for developing sequence-based markers that target specific genomic regions.Single nucleotide polymorphism(SNP)research is relatively new in plant sys-tems but appears to have great potential for marker development.The potential for SNPs was initially shown in Arabidopsis(Jander et al.2002),maize(Ching et al.2002),and soybean(Zhu et al.2003),where extensive analyses for SNPs were performed.Recently,Rajesh and Muehlbauer(2006,unpublished data)estimated176 F.J.Muehlbauer,P.N.Rajesh SNP frequency at1in94bp in coding sequences and1in74bp in genomic regions in two parental chickpea lines(FLIP84-92C[C.arietinum]and PI599072[C.retic-ulatum])that were previously used to develop an interspecific linkage map.SNP frequency between these two interspecific parental lines was sufficient to allow for additional marker development that was used for increasing marker density in the important QTL1region for Ascochyta blight resistance.SNP frequencies within intraspecific parental lines of chickpea were lower than within interspecific parents but comparable to that found in sugar beet(1in126bp; Schneider et al.2001),rice(1in89bp;Nasu et al.2002),maize(1in60-120bp; Ching et al.,2002),cassava(1in62bp;Lopez et al.2005),and pearl millet(1in 59bp;Bertin et al.2005).Other than genetic mapping,SNP-derived markers can be used to genotype chickpea germplasm collections and to associate marker genotype with agronomi-cally important traits with direct implications for breeding programs.This objective can be achieved because technology is available for large scale SNP genotyping with a capacity of2million genotypes per day(BeadArray technology,Illumina, Inc,USA),but SNP markers still need to be determined on a large scale.Genome-wide SNP determination and mapping by high-throughput technology in combination with genetic map integration with physical maps will improve our understanding of the whole genome and will have broad implications for breeding programs.7.4BAC Cloning and UtilizationChickpea genomic research has progressed rapidly over the pastfive years.Start-ing with the development of thefirst BAC library(Rajesh et al.2002),research has made significant advances in various aspects of genomics including:develop-ment of additional BAC libraries(Lichtenzveig et al.2005),whole BAC sequencing (Rajesh current research),whole genome physical mapping(Zhang personal com-munication),EST development(Buhariwalla2003),targeted induced local lesion in genomes(TILLING)mutants(Rajesh personal communication),established proto-cols for Agrobacterium-mediated transformation(AMT)(Sanyal et al.2003),and SNP discovery(Rajesh et al.2005)7.4.1Development of BAC LibrariesThe use of BAC libraries and other large insert libraries has revolutionized genomic research in plants and animals.Although the insert size in BAC libraries is smaller than in yeast artificial chromosome(YAC)libraries,considering the ease of the cloning,long term maintenance of inserts,and lack of chimerism,BAC libraries have gained importance for plant genomic research.BIBAC(binary BAC)libraries have the added advantage of carrying elements in the vector that are necessary for7Chickpea,a Common Source of Protein and Starch in the Semi-Arid Tropics177 AMT.This approach eliminates the necessity of sub-cloning the insert from a BAC vector to a binary vector once the gene of interest is identified.As is commonly known,several agronomically important genes,especially those for disease resis-tance,may be clustered in the genome(Staskawicz et al.1995).A binary vector that has both a large insert cloning capacity and is amenable to AMT significantly reduces the time requirement when transformation is attempted.The BAC libraries can be used for BAC end sequencing,physical mapping and whole genome sequenc-ing.Since most agronomically important traits are governed by QTLs spread widely across the genome,BAC libraries can enhance the success of map-based cloning by reducing the number of steps in chromosome walking.Thefirst BAC library for chickpea was constructed using Hind III(Rajesh et al. 2002)and has been used to identify clones associated with genes for disease resis-tance(Rajesh et al.2004).The BAC library was constructed in the binary vector pCLD04541(otherwise V41)and had a3.8X coverage of the genome with a95% probability of including a genomic fragment with an individual gene of interest. Additional BAC libraries were constructed using Hind III and BamH I restriction en-zymes and vectors pIndigoBAC and pCLD04541,with average insert sizes of121 and145kb,respectively(Lichtenzveig et al.2005).Two additional Hind III BAC libraries have been developed(Doug Cook and Gunther Kahl personal communi-cations)for a total offive libraries from four different cultivars.These resources are available to the chickpea research community and provide combined genome coverage greater than15X.Since the recognition sites of Hind III(AT rich)and BamH1(GC rich)are complementary to each other and the libraries are constructed using different vectors,these libraries are expected to represent the entire genome. The ease of AMT,combined with an established protocol for chickpea and the abil-ity to transform inserts larger than100kb,enhance direct transformation of large inserts representing agronomically important genomic regions from the BIBAC library without further sub-cloning.These libraries have been used for various applications in chickpea such as iden-tification of clones containing resistance genes(Rajesh et al.2004),SNP discovery, whole BAC sequencing(Rajesh and Muehlbauer unpublished data),and develop-ing SSR markers(Lichtenzveig et al.2005).A limited number of sequence based markers have been placed on the chickpea linkage map and the current number of linkage groups is greater than the haploid number of chickpea chromosomes. BAC end sequencing has enormous potential for developing markers to increase marker density in the linkage maps.Currently,these large insert libraries are in use for whole genome physical mapping by BACfingerprinting(HB Zhang personal communication).7.4.2BAC Physical MappingAlthough development of dense linkage maps facilitates identification of mark-ers tightly linked to the desired trait,it requires establishing physical and genetic178 F.J.Muehlbauer,P.N.Rajesh correlation to effectively apply the markers in MAS.Hence,developing markersfrom the genomic region that are physically closer to the gene will have a long term application in breeding programs.Development of genome-wide integrated physical and genetic maps will allow easier chromosome walking,offers an effi-cient and economical approach for map-based cloning,and will serve to identify tightly linked markers.Physical mapping using BAC clones can be accomplished byfingerprint analysis,fluorescent in situ hybridization,optical mapping,iterativehybridization,and chromosome walking(Ren et al.2005).Of these procedures,fin-gerprint analysis is the most commonly used method.Successful physical mappingusing BAC libraries is enhanced by large insert sizes that minimize the number of clones needed to cover the genome and reduces the number of clones required to span a given genomic region with minimal tiling path.Whole genome representa-tion is improved by using complementary restriction enzymes that target AT-rich and GC-rich nucleotides while making the library(Ren et al.2005) The chickpea BAC library of Rajesh et al.(2002)was used to develop markersfrom the ends of the BACs spanning QTL1for Ascochyta blight resistance and to physically map the QTL1genomic region.Sequencing of BACs from these con-tigs identified several candidate disease resistance genes.High resolution mapping was performed using theflanking markers(Rajesh et al.unpublished data).Work is in progress to develop and integrate physical and genetic maps byfingerprinting ∼50,000(∼10X)BACs from BAC libraries and assemble them into a genome-wide contig map of chickpea.The ultimate goal is to anchor the map contigs to the chick-pea genetic maps to whichflowering time and Ascochyta blight resistance QTLwere mapped.The approach should provide additional markers that can be used to fine map these important QTLs(HB Zhang personal communication)and be used for MAS in breeding programs.7.4.3Whole-BAC SequencingHigh throughput technology has made it feasible to sequence entire genomes of different organisms.Genome sequencing of higher organisms is difficult because of their complex nature and large size.However,large scale DNA sequencing is under-way and is useful for studying genome organization,comparing genomes of related species,and SNP discovery.Currently,most genome sequencing projects in higher organisms are performed using BAC clones because of their advantages such as containing specific gene-rich genomic regions(euchromatin)that can be sequenced rather than the entire genome.Also,BAC clones allow easy distribution of specific genomic regions to collaborating laboratories.By sequencing BACs representing only euchromatic genomic regions,the Medicago Genome Sequencing Consortium predicts that90%of the gene content will be sequenced by sequencing just50%of the500Mb Medicago truncatula genome(Town2005).In chickpea,although the number of sequences being deposited in the database has gradually increased,whole-BAC sequencing was not attempted until20037Chickpea,a Common Source of Protein and Starch in the Semi-Arid Tropics179 (Rajesh2006unpublished data).Currently,there are758,388base pairs that show particular genome organization and composition in chickpea.Based on an estimated gene density of one gene per9.2kb and genome size of740Mb,approximately 80,000genes are estimated to be present in chickpea genome.Considering the av-erage gene size and the total number of genes,it is estimated that200Mb of the genome(27%)consist of genes.These estimates were based on11BAC clones that represented different genomic regions,three of which are located in a genomic region associated with resistance to Ascochyta blight(QTL-1).Interestingly,repeti-tive elements were relatively infrequent(8.68%)in these BAC clones,while in other plants repetitive elements account for more than50%of the genome.Sequencing of additional BAC clones is underway and should help clarify the organization of the chickpea genome.7.5Molecular CytogeneticsKaryotype analysis has been used to describe the relative size and arm lengths of the eight chromosomes of C.arietinum and several of the wild relatives;however, the small size and sticky nature of the chromosomes does not lend itself to more detailed analysis(Ahmad2000).The relative size and chromosome arm lengths of the chromosomes has made it possible to place some of the species into groups. For example,it has been possible to place the cultivated species,C.arietinum,with C.reticulatum,C.echinospermum,and the perennial species,C.anatolicum and C.songaricum into the same grouping based on chromosome morphology(Ah-mad2000;Croser et al.2003).Morphological examination indicates that only one chromosome has a nucleolar organizer region(Ocampo et al.1992;Ahmad2000).Classification of chickpea chromosomes byflow cytometry(Vláˇcilováet al.2002) was successful in isolating chromosomes corresponding tofive of the eight chickpea chromosomes and associating STMS markers from linkage group8to the small-est chromosome.These promising results complement genetic linkage and physical mapping of the chickpea genome by associating specific linkage groups to particular chromosomes.7.6Functional GenomicsWith the availability of complete genome sequences(Arabidopsis and rice)and near complete whole genome DNA sequences of model plants,25,000genes have been predicted in Arabidopsis and from32,000to50,000have been predicted in rice. Since the predicted number of genes exceeds the number of genes with known function,it is challenging to determine the function of each gene.Full length cDNA synthesis is a major aspect of functional genomics not only for annotating,but for in vitro and in vivo characterization of each gene.Correlation of functional change by alteration of a gene at the molecular level in mutants will facilitate the determination180 F.J.Muehlbauer,P.N.Rajesh of the gene function.Emphasis has also been placed on how well these genes are spatially and temporally separated or coordinated by transcriptional profiling of dif-ferent tissue and at different time intervals.Overall,the study of functional genomics of chickpea is in the very early stages.7.6.1EST DevelopmentFull length cDNA synthesis of all the genes is technically laborious and costly; therefore,partial sequencing of the expressed genes,known as expressed sequence tags(ESTs),promulgates nearly as much information as full length cDNA and is less time consuming.Development of an EST database for each crop is necessary for transcriptome profiling and also for gene discovery.Differentially expressed genes during the Ascochyta rabiei-chickpea interaction have been studied using various techniques such as differential cDNA hybridization (Ichinose et al.2000),differential display reverse transcription(DDRT)(Rajesh et al.2003),and cDNA-AFLP(Cho et al.2004).However,global transcriptome analysis representing the entire chickpea genome has not been possible because of the small number of ESTs.Thefirst report of large scale EST development was published in2003with488entries in Genbank(Buhariwalla et al.2003).Recently, Coram and Pang(2005)increased this number by depositing an additional566ESTs in Genbank and developed a microarray platform to study the Ascochyta rabiei–chickpea interaction.At the time of writing this chapter,there were1,302chickpea EST entries in Genbank.ESTs can be used as genetic markers as well.Buhariwalla et al.(2003)showed the utility of ESTs as markers in genetic diversity analysis among chickpea wild species.EST mapping is worthwhile because gene-rich regions in the genome can be defined and gene discovery is possible for traits of interest.SNP discovery in ESTs across different accessions will help in developing SNP markers that can be used in high throughput genotyping.Microsatellites present in ESTs(EST-SSRs) are variable and are likely to be naturally polymorphic.EST-SSR markers designed from Medicago truncatula has been attempted in chickpea,lentil and pea.Although the number of polymorphic markers was low,it showed the potential for transfer-ability of markers across the species(Muehlbauer unpublished data).However,these markers were70%polymorphic among six Medicago species(Eujayl et al.2004).A separate study by Gutierrez et al.(2005)indicated significant transferability of M.truncatula microsatellites to three pulse crops(40%to faba bean,36.3%to chickpea and37.6%to pea)and that the SSR motifs were variable,although the flanking regions were conserved.It is,therefore,possible to design markers with good transferability and representative of the same genomic regions.This may be achieved by incorporating phylogenetic relationships with molecular analyses to determine genome conservation.Comparative EST sequence analysis involving two major clades including cool season legumes and warm season legumes identifies the conserved exon sequences。