Plant Molecular Biology
University of Frankfurt
Frankfurt am Main
Genomics, the whole repertoire of techniques to study the genomes of organisms, has now been developed so that it can be applied to any known genome. With the exception of molecular cytogenetics (e.g. flow sorting of chromosomes, fluorescence in situ hybridisation) and the generation of dominant and codominant molecular DNA markers, most genome technologies have not yet been applied to banana genomics. Therefore, genetic and physical mapping, genes and their expression patterns, EST libraries, expression chips, promoters, and the whole genome sequencing approach are described briefly, and their impact on banana genomics illustrated.
A decade of genomics which has seen major breakthroughs such as sequencing of the whole genomes of, for example, human , fruit fly , Arabidopsis thaliana , and Caenorhabditis elegans  lies behind us. And the genome sequencing euphoria persists: more than 70 prokaryotic genomes are now completely sequenced, a series of lower eukaryotic genomes are either sequenced, or in progress, and relatively complex genomes of lower eukaryotes (yeasts [5,6]) are now known base by base. And the boom of whole genome sequencing projects for higher organisms - regardless of their size - is still unabated. The discovery of novel genes through the exploitation of huge expressed sequence tag (EST) databases, as well as multiple cDNA libraries of varying depths and from almost all tissues of an organ or organism, is routine work these days. Genome-wide expression analysis of virtually all genes has become possible, even though the profiling techniques are technically quite demanding and almost prohibitively expensive (e.g. serial analysis of gene expression, SAGE  and massively parallel signature sequencing, MPSS ). The function(s) of all the genes discovered is now under study: the era of functional genomics has begun. Though major developments in technologies are expected in the future, only protein analysis via two- and three-dimensional polyacrylamide gel electrophoresis, mass spectrometry of peptide fragments, and interactive mapping (i.e. the identification of interacting proteins) with yeast or bacterial two-hybrid systems or protein chips have led to significant progress in the field of proteomics. Highly active laboratories are starting to compile a complete inventory of metabolic activities in normal (resting) and induced (e.g. mutagenised, activated, or stressed) cells, and to make a comprehensive quantification of (preferably) all metabolites simultaneously, thereby covering the largely neglected field of metabolomics.
All these, and many other new developments, have had little effect on banana genomics, i.e. the whole repertoire of techniques to decipher the overall composition and structure, complete sequence, gene content, expression profile, and gross and subtle changes of distinct regions of the banana genome in response to environmental factors (e.g. stress). Therefore, not surprisingly, genome analysis in banana and plantain is definitely lagging behind.
Notwithstanding this deficiency in banana chromosome research, a number of activities have commenced worldwide, aimed at the molecular description of the banana genome at both the chromosomal and sequence levels. The present short overview lists some of these research areas, critically interprets the achievements, and points out some perspectives.
Since the introduction of flow cytometry into banana chromosome research , molecular cytogenetics has contributed much to our understanding of DNA content  and ploidy levels in Musa species  in existing germplasm collections as well as in newly collected accessions. Karyological stability in vitro was confirmed over time, as was the size of the nuclear genome . Moreover, genomic in situ hybridisation has permitted the determination of the gross genomic constitution of hybrids . In situ hybridisation experiments have revealed the physical location of middle repetitive sequences in the Musa genome (ribosomal RNA genes , retrotransposons, e.g. monkey , and integrated banana streak badnavirus sequences ). It is almost certain that the banana genome harbours other repetitive DNA sequences. For example, eight distinct groups of Ty1-copia-like retroelements have been identified in ten different banana varieties, and Ty3-gypsy-like retrotransposons have been isolated. The relatively wide distribution of retroposons and their comparatively fast evolution rate make them potential markers for biodiversity and phylogenetic studies.
Cytogenetics will certainly continue to be the method of choice for studying gross chromosome structure and its changes . However, localization of unique sequences will be necessary, even though the resolution of FISH is far from being satisfactory. Fibre-FISH - once it has been introduced into banana cytogenetics - will expand the resolution beyond today's limits and allow the physical mapping of single-copy target genes. Cytogenetic maps, based on FISH, fibre-FISH, PRINS and chromosome painting, are of great help for sequencing the whole banana genome, which is currently being planned (see below).
3. MOLECULAR MARKERS
The pioneering era for molecular marker technologies is almost over: many different marker techniques are available and are increasingly being exploited to generate DNA markers of various kinds for various purposes. Only a minority of the 30 or more molecular marker techniques have been tested on banana nuclear, mitochondrial and chloroplast DNAs, and some were found to be useful, all of them PCR-based. For example, random amplified polymorphic DNA (RAPD), arbitrarily primed PCR (AP-PCR), amplified fragment length polymorphism (AFLP), inter-simple sequence repeat assay (ISSR), methylation-sensitive amplified polymorphism (MSAP) and mini- and microsatellite-based markers have all been applied to solve various problems in banana genetics [e.g. 18-20]. One type of marker that has turned out to be most promising is the simple sequence repeats (SSRs) or sequence-tagged microsatellite sites (STMSs), which are locus- and allele-specific codominant markers . Microsatellites, mono-, di-, tri-, tetra- or pentanucleotide repeats with copy numbers from 2 to 60 at a particular locus, are ubiquitous components of all eukaryotic genomes (including Musa ) and are highly variable, since they undergo slipped-strand mispairing-induced expansion or contraction. Although the application of these valuable markers is simple, their production is still prohibitively expensive. Genomic DNA has to be isolated, restricted and cloned, clones with microsatellites have to be selected by hybridisation with synthetic oligonucleotide repeats, the microsatellite islands together with their genomic flanks must be sequenced, and flank-specific primers have to be designed. Notwithstanding these difficulties, a series of STMS markers have been generated  and used extensively for the identification of some of the 1400 clones from the International Musa Germplasm Collection at the Catholic University of Leuven (Belgium), identification of species (some STMS are specific for Australimusa, M. schizocarpa and M. balbisiana, respectively), and identification of interspecific hybrids . STMS markers also allow determination of the degree of biodiversity generally, characterisation of wild species, detection of sequence diversity in the A versus B genomes, help management of gene banks, and help to fingerprint varieties, to identify introgressed genes, to serve as markers for the establishment of genetic maps (see below) and marker-assisted selection (MAS) in breeding (once STMSs are identified that cosegregate with the desirable trait, they can be exploited to accelerate the selection of good-yielding cultivars for fruit quality, post-harvest fruit stability, abiotic stress tolerance, pest and disease resistance and higher yield), and last but not least support evolutionary, taxonomic and population genetics. For example, they are unsurpassed for following the secondary diversification of the few plantain varieties introduced into Africa some 3000 years ago, the effect of the appearance of pathogens (previously absent) on the genome of these plantains, and the co-evolution of host plants and pathogens at the genomic level. In essence, molecular markers such as STMS will be used increasingly by numerous laboratories around the world for the various applications listed above, and will soon become an indispensable routine tool in banana research. Moreover, single nucleotide polymorphisms (SNPs) have now to be added to the repertoire of useful markers, especially when they appear in coding regions and hence in the messages (so called expressed SNPs or eSNPs, or cSNPs for SNPs as part of a cDNA), since only then can they be expected to have an impact on protein function, and consequently on the phenotype of the organism . To date, not one report has appeared on detection of SNPs in the banana genome.
4. GENETIC AND PHYSICAL MAPPING OF THE BANANA GENOME
Basically, mapping aims at identifying molecular markers genetically (genetic maps) or physically (physical maps) linked to major or minor genes (generally loci) contributing to the expression of a particular trait or continuously varying character (e.g. a QTL). Linked markers can then be exploited to isolate the gene(s) underlying the trait. The isolated genes in turn are used to improve selected genotypes via direct or Agrobacterium-mediated gene transfer or, alternatively, the linked markers may serve to select segregants of a cross that carry a desirable trait (marker-assisted selection, MAS). For these reasons genetic and (in a more advanced state) physical maps have now been established for almost all the important crop plants. Most of these maps are integrated maps, i.e. they contain a series of different molecular markers, preferably in a framework of STMS (skeleton map). Genetic mapping in Musa is not very far advanced, though a first low-density map of M. acuminata was published as early as 1993 , based on a cross between SF265 (AA) × Banksii (AA) segregating for parthenocarpy, and established using isozyme, RFLP, and RAPD markers. Although a series of crosses segregating for other traits (e.g. black Sigatoka resistance, bunch position, chromosome rearrangements, etc.) has been developed, and mapping projects have been announced (e.g. by CIRAD), no high-density linkage map is yet available. Nor are the mapping populations themselves available, despite the fact that several activities are aimed at developing suitable segregating populations (e.g. at IITA, CIRAD, INIBAP, CARBAP, and EMBRAPA), all of them based on the A genome. No cross exists which involves the B genome, but BB segregating populations would be of high value. This situation hampers progress in banana genomics severely. INIBAP has recently been engaged in the development of crosses with large populations that segregate for a series of desirable traits.
Although genetic mapping of the Musa genome obviously lags behind other crop plants of comparable market value, several bacterial artificial chromosome (BAC) libraries of the A genome have already been established (B genome libraries are under construction), which will allow the physical mapping of the banana genome. Physical mapping aims at defining the location of a particular gene (or DNA sequence) on a cloned genomic sequence of a size of 100-200 kb. It also allows one to relate genetic distances (cM) between two (or more) markers to physical distances (kb), to align syntenic (and also non-syntenic) regions of two or more genomes from related or non-related organisms to search for homologous, orthologous or paralogous sequences, and to build contigs of specific genomic regions to pinpoint a target gene and to isolate it using map-based cloning approaches [e.g. 27-29]. The physical mapping methodology involves FISH or fish-FISH to locate sequences on chromosomal preparations [30, 31], or the production of yeast artificial chromosomes, or more efficiently, BAC libraries with large inserts, or transformation-competent artificial chromosomes (TACs). Global physical mapping comprising the entire genome has been achieved for only a few plant species (e.g. for Arabidopsis thaliana  and Oryza sativa ), but is in progress for many other crop plants.
Although several BAC libraries of different Musa species (M. acuminata, A library; M. balbisiana, B library) have been produced, the clones have not generally been ordered, nor has a tiling path been constructed around interesting regions. However, it is to be expected that the growing awareness of the scientific banana community will catalyse the process of physical mapping, which is the path to the isolation of agronomically interesting genes.
5. GENES AND GENE EXPRESSION
Although several expressed sequence tag (EST) and cDNA libraries have been established (e.g. for M. acuminata ssp. malaccensis), no Musa EST database can yet be tapped for information about expressed genes. Also, the depth of these libraries is not known (or has not been disclosed). Yet full-length cDNA libraries, normalized and representative, would be needed from a whole series of tissues and states (e.g. normal vs. diseased; susceptible vs. resistant; different developmental states). First experiments on the expression of, for example, elicited genes, revealed interesting cDNAs. For example, mRNAs from chitinase- and phenylalanine ammonium lyase-encoding genes appear after a simulated attack on M. acuminata leaves by an elicitor preparation from Mycosphaerella fijiensis. In other similar experiments with the black Sigatoka (BS)-resistant cultivar 'Calcutta IV' an elicited cDNA encoding a high mobility group (HMG) protein, possibly involved in changing chromatin architecture, has been identified (unpublished). A series of resistance gene analogues (RGAs) have been isolated, using degenerate PCR primers targeting highly conserved regions in proven plant resistance genes (e.g. kinase or transmembrane-encoding domains, or leucine-rich repeat sequences, to name only few). On one hand, the laboratory of J.L. Dale (QUT, Brisbane, Australia) identified five low-copy RGAs from both resistant and susceptible banana cultivars that were transcribed. If the RGA sequences from resistant and susceptible plants were compared, differences emerged between the five classes, and also within each group. L. Sagi in Rony Swennen's laboratory at KUL (Leuven, Belgium) used primers complementary to conserved nucleotide binding site (NBS)/leucine-rich repeat (LRR) domains of plant resistance genes (e.g. N from tobacco, RPS2 from Arabidopsis thaliana) to isolate Cf orthologues from the banana landrace Zebrina GF DNA . The author's laboratory has isolated a whole series of RGAs from the somaclonal mutant CIEN-BTA-03 (resistant to both M. fijiensis and M. musicola) and the parent Williams that fall into two classes: nucleotide-binding site-leucine-rich repeat-containing kinases, and serine-threonine protein kinases of the pto type. All the RGAs were fully sequenced, and eight of them are also transcribed in the mutant, its parental genotype, Pisang Mas and a tetraploid M. acuminata (Gimenez et al., in preparation).
This list is by no means complete, but it portrays the situation in banana research. Few genes are targeted, some sequences are known, fewer publications have appeared, and no banana gene whatsoever has been applied in any way (e.g. for transformation). Also, no attempt has yet been made to design expression chips with families of genes whose sequences are derived from either cDNAs (cDNA microarray), or oligonucleotides, or from clones obtained from related or unrelated plants. Expression chips of various densities are now routinely used in many laboratories around the world, although their construction still demands a lot of expertise in bioinformatics, coupling chemistry and fluorescence detection techniques. Nevertheless, expression chips of various kinds are currently the state-of-the-art for expression analysis in plants and animals. It should be possible to start such solid-phase low density expression analysis by simple means (e.g. spotted cDNAs on membrane filters, and hybridisation with radioactively labelled probes and subsequent autoradiography). But liquid-phase, massively parallel and potentially automated expression techniques are also available, but not used in banana transcriptome analysis. Whereas massively parallel signature sequencing (MPSS) is much too complex and expensive, serial analysis of gene expression (SAGE) has already been tested successfully with plants, though originally developed for human expression studies . MPSS stands for high-throughput sequencing of millions of cDNAs conjugated to 32-mer capture oligonucleotide tags on the surface of 5 µm diameter microbeads, where each microbead harbours some 100,000 identical copies of a particular cDNA ('microbead library'), that circumvents separate cDNA isolation, template processing and robotic procedures. Without referring to details, the abundance of each and every mRNA of a cell can be estimated by counting the number of clones with identical signatures. However, the multitude of steps (e.g. combinatorial synthesis of capture oligonucleotides, cDNA synthesis and restriction, addition of capture oligos to the 3'-end of each cDNA, cloning with PCR handles, PCR, denaturing of the product, ligation of annealed sequences, fluorescence-activated cell sorting, sequencing of about 16-20 bases at the free ends of the cloned templates, flow-cell management, adaptor ligation, etc.) will certainly prevent the common use of this otherwise extremely interesting, ultrahigh throughput technique for expression profiling. However, the less demanding, but nevertheless highly informative SAGE method has already been introduced into plant transcriptomics , enabling the simultaneous detection, identification and quantification of virtually all transcripts (and therefore genes) in a given cell at a given time, and allowing monitoring of gene expression patterns at various developmental stages or after specific treatments of the plants. SAGE is based on the isolation of short 9-14 bp SAGE tags from the 3'-end of the transcripts, ligation of tags from different cDNAs, and their amplification. Then 30-50 such tags, serially ligated in a single DNA molecule, are cloned and sequenced. The frequency with which each tag is represented correlates with the number of mRNAs originally present in the target cell.
Apart from these sophisticated technologies, a number of less complex techniques are available that produce enough information for an initial transcriptome analysis (e.g. random activation of gene expression, RAGE ). No matter what technique is finally used, the correlation of transcribed sequences with physiological or developmental processes (e.g. ripening of fruits) would add tremendously to our understanding and, as a consequence, would define target genes for genetic improvement (e.g. critical alleles of genes for increased agronomic value).
The production of transgenic bananas harbouring genes encoding an improved genotype requires strong promoters for both constitutive and regulated high-level expression of the transgenes. Generally speaking, not enough promoters are available, though several groups are actively working to isolate useful promoter sequences, or at least fragments with promoter activity. At KUL, promoters from banana streak virus (BSV) isolates from Australia proved to drive high expression of GUS reporter genes in transgenic monocot (banana, barley, maize, millet, sorghum) and dicot plants (canola, sunflower, tobacco), and also in Pinus radiata and Nephrolepis cordifolia . At least in greenhouse experiments, the BSV promoter ('My fragment') was superior to the widely used maize ubiquitin and cauliflower mosaic virus 35S promoters. Another promoter, originating from sugarcane bacilliform badnavirus (ScBV), which could infect banana as well, showed high activity in transient expression assays both in mono- and dicotyledonous hosts, but could also be used to drive transgenes nearly constitutively. Moreover, the group of J.L. Dale (QUT, Brisbane, Australia) derived promoter regions from banana bunchy top nanovirus (BBTV) satellites S1 and S2 and from the banana actin gene that are all active in transgenic banana plants. For example, BBTV S1 and S2 promoters supported vascular tissue-associated reporter gene expression, and can certainly be exploited as tissue-specific promoters. The ACT1 gene promoter directed strong transcription of reporter genes in leaves and roots. A series of promoters has also been isolated from genes induced after challenge with a fungus (Sagi, L. pers. comm.). It seems as if the availability of suitable promoters is no longer a bottleneck for banana transgenics, especially since their isolation by e.g. TAIL-PCR  or other walking techniques are routine these days.
It comes as no surprise that the number of transgenic bananas is continuously increasing . Transformation protocols, including up- and downstream tissue culture techniques, suitable transformation constructs with modified promoters  driving one or more transgenes, appropriate transformation techniques such as particle bombardment [41, 42] and Agrobacterium-driven gene transfer (which is superior in view of the small number of insertion sites and the comparably low number of transgene copies per genome), or other suitable techniques such as co-transformation , the detection of the transgenes (preferably by PCR) and characterization of their insertion sites (copy number), are well developed ('tool box'). Transgenes have been recruited exclusively from heterologous sources (e.g. antifungal peptide genes from Raphanus sativus), and mostly encoded proteins conferring resistance to Mycosphaerella fijiensis or Fusarium oxysporum generally, rather than specifically. Other transgenes exploited sense suppression by antisense sequences. For example, fruit-specific promoters that drive antisense genes to inhibit ethylene synthesis in fruits so that the ripening process was delayed; transgenics retained this new trait over multiple generations, even under field conditions.
Since all the cultivated varieties of banana are sterile and therefore do not set seed, traditional breeding by hybridisation is more difficult than genetic transformation using molecular techniques. Although attempts to produce transgenic bananas and plantains are still proceeding too slowly, public acceptance of these novel plants and their products should already be prepared for through sound information and risk assessment, although the chances of transfer of transgenes from transgenic field material to wild species (the major public concern) are to be expected to be negligible in view of the sterility of many cultivars.
6. WHOLE GENOME SEQUENCING
A logical consequence of the various successful genome sequencing projects, and the ever-increasing rate of development of novel and more effective sequencing techniques and ultrahigh-throughput platforms, is the plan to sequence the banana (M. acuminata) genome. In 2001, the Global Musa (Banana) Genomics Consortium, comprising scientists from 11 countries and various academic, governmental, non-profit and profit organizations started to sequence the approximately 550-600 Mb of DNA per haploid genome, divided into a total of 11 chromosomes, within the next four years. Given the continuous participation of the renowned Institute of Genome Research (TIGR) located in Rockville (MD, USA), this project may even be finished earlier. In an accompanying paper, we have made a strong case for sequencing the much smaller genome of Mycosphaerella fijiensis as well (see Chapter 11 in this volume).
As has been found from other sequencing projects (e.g. Arabidopsis thaliana or Oryza sativa), the enormous amount of data, if properly managed, will reveal hitherto unknown features of the genome. These include sequence composition of various genomic regions, an inventory of the various genic and non-genic sequences (genes and repetitive DNA such as satellites, mini- and microsatellites, pseudogenes, retropseudogenes, retrotransposons, LINES, SINES, DNA transposons and many others), the distribution of the various elements along the chromosomes, potential duplications, translocations, inversions, macro- and microsynteny with related or unrelated plant genomes, structure of centromeres and telomeres, the exact genome size, number of open reading frames, which approximates the number of genes, to name only few. Since the genus Musa comprises autopolyploids (e.g. AAA, AAAA, AAAAAA), various allopolyploids (AAB, ABB, AABB, AAAB), diploids (AA, BB) and hybrids (AB), the interactions of parental genomes at the sequence level, the identity of each genome and the transfer of sequences from one genome to the other can easily be deciphered. With the immense knowledge accumulated by genome sequencing, many still intractable problems of banana research could be tackled much better (e.g. genome evolution), but the problems of data handling remain, since bioinformatics for banana genomics has not yet been developed (data repository, software design). So, this project should be welcomed by all those who want bananology to be at the forefront of modern science.
The author's research has been supported by IAEA (CRP 302-02-GFR-8148), EU (ERBIC-18/CT-970192) and Deutsche Forschungsgemeinschaft (Ka 332/15-1 and 15-2).
 INTERNATIONAL HUMAN GENOME SEQUENCE CONSORTIUM, Initial sequencing and analysis of the human genome, Nature 409 (2001) 860-921.
 ADAMS, M.D., et al., The genome sequence of Drosophila melanogaster, Science 287 (2000) 2185-2195.
 THE ARABIDOPSIS GENOME INITIATIVE, Analysis of the genome sequence of the flowering plant Arabidopsis thaliana, Nature 408 (2000) 796-815.
 THE C. ELEGANS SEQUENCING CONSORTIUM, Genome sequence of the nematode C. elegans: a platform for investigating biology, Science 282 (1998) 2012-2018.
 GOFFEAU, A., et al., The yeast genome directory, Nature 387 (1997) 1-105.
 WOOD, V., et al., The genome sequence of Schizosaccharomyces pombe, Nature 415 (2002) 871-880.
 VELCULESCU, V.E., et al., Serial analysis of gene expression, Science 270 (1995) 484-487.
 BRENNER, S., et al., Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays, Nature Biotechnol. 18 (2000) 630-634.
 DOLEZEL, J., Flow cytometric analysis of nuclear DNA contents in higher plants, Phytochem. Anal. 2 (1991) 143-154.
 DOLEZEL, J., et al., Flow cytometric estimation of nuclear DNA amount in diploid bananas (Musa acuminata and M. balbisiana), Biol. Plant. 36 (1994) 351-357.
 DOLEZEL, J., et al., Use of flow cytometry for rapid ploidy determination in Musa species, InfoMusa 6 (1997) 6-9.
 LYSAK, M.A., et al., Flow cytometric analysis of nuclear DNA content in Musa, Theor. Appl. Genet. 98 (1999) 1344-1350.
 D'HONT, A., et al., The interspecific genome structure of cultivated banana, Musa spp., revealed by genomic DNA in situ hybridisation, Theor. Appl. Genet. 100 (2000) 177-183.
 DOLEZELOVA, M., et al., Physical mapping of the 18S-25S and 5S ribosomal RNA genes in diploid bananas, Biol. Plant. 41 (1998) 497-505.
 BALINT-KURTI, P.J., et al., Identification and chromosomal localization of the monkey retrotransposon in Musa sp., Mol. Gen. Genet. 263 (2000) 908-915.
 HARPER, G., et al., Integration of banana streak badnavirus into the Musa genome: molecular and cytogenetic evidence, Virology 255 (1999) 207-213.
 SHEPHERD, K., Cytogenetics of the genus Musa, INIBAP, Montpellier, France (1999).
 GAWEL, N.J., et al., Restriction fragment length polymorphism (RFLP)-based phylogenetic analysis of Musa, Theor. Appl. Genet. 84 (1992) 286-290.
 HOWELL, E.C., et al., The use of RAPD for identifying and classifying Musa germplasm, Genome 37 (1994) 328-332.
 JARRET, R.L., et al., Detecting genetic diversity in diploid bananas using PCR and primers from a highly repetitive DNA sequence, Euphytica 68 (1993) 69-76.
 KAEMMER, D., et al., Molecular breeding in the genus Musa: A strong case for STMS marker technology, Euphytica 96 (1997) 49-63.
 KAEMMER, D., et al., Oligonucleotide and amplification fingerprinting of wild species and cultivars of banana (Musa spp.), BioTechnology 10 (1992) 1030-1035.
 JARRET, R.L., et al., Isolation of microsatellite DNA markers in Musa. InfoMusa 3 (1995) 3-4.
 LAGODA, P.J.L., et al., Sequence tagged microsatellite site (STMS) markers in the Musaceae, Mol. Ecol. 7 (1998) 657-666.
 BROOKES, A.J., The essence of SNPs, Gene 234 (1999) 177-186.
 FAURE, S., et al., A molecular marker-based linkage map of diploid bananas (Musa acuminata), Theor. Appl. Genet. 87 (1993) 517-526.
 YANG, D., et al., Construction of a BAC contig containing the Xa5 locus in rice, Theor. Appl. Genet. 97 (1998) 1120-1124.
 BUDIMAN, M.A., et al., A deep-coverage tomato BAC library and prospects toward development of an STC framework for genome sequencing, Genome Res. 10 (2000) 129-136.
 FU, H.H., DOONER, H.K., A gene-enriched BAC library for cloning large allele-specific fragments from maize: Isolation of a 240kb contig of the bronze region, Genome Res. 10 (2000) 866-873.
 DEJONG, J.H., et al., High resolution FISH in plants - techniques and applications, Trends Plant Sci. 4 (1999) 258-263.
 JACKSON, S.A., et al., Application of fish-FISH in physical mapping of Arabidopsis thaliana, Genome 41 (1998) 566-572.
 MOZO, T., et al., A complete BAC-based physical map of the Arabidopsis thaliana genome, Nature Genet. 22 (1999) 271-275.
 KURATA, N., et al., Physical mapping of the rice genome with YAC clones, Plant Mol. Biol. 35 (1997) 101-113.
 WIAME, I., et al., PCR-based cloning of candidate disease resistance genes from banana (Musa acuminata), Acta Hort. 521 (2000) 51-57.
 MATSUMARA, H., et al., Technical advance: Transcript profiling in rice (Oryza sativa L.) seedlings using serial analysis of gene expression, Plant J. 20 (1999) 719-726.
 HARRINGTON, J. J., Creation of genome-wide protein expression libraries using random activation of gene expression, Nature Biotechnol. 19 (2001) 440-445.
 SCHENK, P. M., et al., Promoters for pregenomic RNA of banana streak badnavirus are active for transgene expression in monocot and dicot plants, Plant Mol. Biol. 47 (2001) 399-412.
 TERAUCHI, R., KAHL, G., Rapid isolation of promoter sequences by TAIL-PCR: The 5´-flanking regions of Pal and Pgi genes from yams (Dioscorea), Mol. Gen. Genet. 263 (2000) 554-560.
 SAGI, L., et al., Production of transgenic banana and plantain, Acta Hort. 521 (2000) 51-57.
 KAY, R., et al., Duplication of CaMV 35S promoter sequences creates a strong enhancer for plant genes, Science 236 (1987) 1299-1302.
 BECKER, D. K., et al., Genetic transformation of Cavendish banana (Musa spp. AAA group) cv. 'Grand Nain' via microprojectile bombardment, Plant Cell Rep. 19 (2000) 229-234.
 SAGI, L., et al., Genetic transformation of banana and plantain (Musa spp.) via particle bombardment, BioTechnology 13 (1995) 481-485.
 REMY, S., et al., Co-transformation as potential tool to create multiple and durable resistance in banana (Musa spp.), Acta Hort. 461 (1998) 361-365.