The International Center for Agricultural Research in the Dry Areas (ICARDA) is participating in a large-scale programme, Subprogram 1 of the Consultative Group on International Agricultural Research (CGIAR) Generation Challenge Program, that aims to explore the genetic diversity of the global germplasm collections held by the CGIAR research centres. This project will identify a ‘composite collection’ of germplasm for individual crops, representing the range of diversity of each crop species and its wild relatives, and characterize each composite set using anonymous molecular markers, mainly simple sequence repeats (SSRs). The overall goal of this project is to study diversity across given genera and identify genes for resistance to biotic and abiotic stresses that can be used in crop improvement programmes. ICARDA was responsible for creating the composite collection for lentil. ICARDA has the global mandate for lentil and houses the largest global collection of this crop with 10,509 accessions. From this collection, a global composite collection of 1000 lentil accessions was established with the aim to represent genetic diversity and the agro-climatological range of lentil. Accessions for the composite collection were compiled from landraces, wild relatives, and elite germplasm and cultivars. The methodology presented here combined classical hierarchical cluster analyses using agronomic traits and two-step cluster analyses using agro-climatological data linked to the geographical coordinates of the accessions' collection sites. Genotyping for 30 SSR loci will be carried out for all 1000 accessions. Plants grown for DNA analysis will be harvested and progeny will be evaluated under field conditions at ICARDA.

Chickpea is one of the most important grain legume crops in the world. Large collections of genetic resources are maintained in the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) and International Center for Agricultural Research in the Dry Areas (ICARDA) genebanks. Association mapping using neutral markers has been suggested as a means to identify useful alleles in the vast reservoirs of genetic diversity existing in the germplasm collections that could be associated with the phenotypes among the population individuals. ICRISAT in collaboration with ICARDA developed a global composite collection of 3000 accessions that will be profiled using 50 polymorphic simple sequence repeat (SSR) markers. The data generated through this collaborative effort will be used to define the genetic structure of the global composite collection and to select a reference sample of 300 accessions representing the maximum diversity for the isolation of allelic variants of candidate gene associated with beneficial traits. It is then expected that molecular biologists and plant breeders will have opportunities to use diverse lines in functional and comparative genomics, in mapping and cloning gene(s), and in applied plant breeding to diversify the genetic base of the breeding populations which should lead to the development of broad-based elite breeding lines/cultivars with superior yield and enhanced adaptation to diverse environments.

The improvement of quality and quantitative traits in industrial crops is among the most important goals in plant breeding. Many traits of interest are controlled by multiple genes and improvements have so far only been obtained through conventional breeding. The use of biotechnological tools to modify quantitative traits is highly challenging. CropDesign has developed TraitMill™, an automated plant evaluation platform allowing high-throughput testing of the effect of plant-based transgenes on agronomically valuable traits in crop plants. The focus of the platform is currently on rice, a good model for other important cereals such as maize and wheat. TraitMill™ offers a high-throughput prediction of gene function. Genes of validated function that confer trait improvement can then be transferred to other cereal crop species such as maize, but also to dicots, trees and ornamentals. TraitMill™ involves the following key components: (i) selection of candidate trait improvement genes among genes involved in signal transduction, cell cycle control, transcription, nutrient metabolism, etc.; (ii) a suite of validated constitutive or tissue-specific promoters from rice allowing for the selection of the most appropriate promoter–gene combination in view of the desired trait improvement; (iii) an industrialized plant transformation system generating tens of thousands of transgenic plants annually; and (iv) a robotized trait evaluation set-up for plant evaluation, proprietary image analysis software for measuring plant performance parameters and statistical analysis of results.

Conservation of crop germplasm diversity involves the establishment of in situ and ex situ genebanks. The major activities for ex situ genebanks include assembling, conserving, characterizing and providing easy access to germplasm for scientists. More than six million accessions are currently assembled in over 1300 genebanks worldwide. ICRISAT is one of the 15 CGIAR centres, with headquarters at Patancheru, India, and conserves genetic resources of sorghum, pearl millet, chickpea, pigeonpea, groundnut, and six small millets. The ICRISAT genebank holds 114,870 accessions from 130 countries, including both archival materials from various organizations throughout the world, and from fresh collections resulting from 213 missions in 62 countries. The ICRISAT genebank supplies annually over 40,000 germplasm samples to scientists worldwide. Sixty-six varieties selected from the basic germplasm have been released for cultivation in 44 countries, and ICRISAT has restored/repatriated crop germplasm to eight countries. The research focus is on germplasm diversity assessment, developing core and mini-core collections, and using a molecular characterization approach to both enhance the utilization of germplasm in research and improve the efficiency of germplasm management. Following these approaches, we have been able to identify a significant number of accessions with traits potentially relevant for crop improvement.

Genomic and proteomic diversity provide the basis of evolutionary change by natural selection under abiotic and biotic stresses, and the human-driven evolutionary process of domestication by artificial selection. Described here are some of the regional and local genomic and proteomic long-term multidisciplinary studies conducted at the Institute of Evolution, University of Haifa, Israel, during 1975–2005 (see publications at http://evolution.haifa.ac.il), involving both wild barley, Hordeum spontaneum, the progenitor of cultivated barley and wild emmer sheat, Triticum dicoccoides, the progenitor of modern tetraploid and hexaploid cultivated wheat. Wild cereals harbour large amounts of as yet untapped adaptive genetic resources for crop improvement (resistances against abiotic and biotic stresses, micronutrient metal deficiencies, storage proteins, amylases and photosynthetic yield, among others). The adaptive genomic diversity of wild cereals, including cryptic beneficial alleles at specific quantitative trait loci of T. dicoccoides and H. spontaneum is the best genomic resource to be conserved in situ and ex situ for utilization by classical and modern biotechnologies, to enrich the genetically impoverished and stress-vulnerable food cultivars, advance crop improvement, and thereby increase and optimize world food production in a second genetic green revolution.

Drought tolerance (DT) is a very complex trait. For the purpose of developing DT rice cultivars, a research programme was initiated in 1997 in China, consisting of four coherent parts: the development of a DT field screen facility and evaluation standard; the collection, evaluation and enhancement of DT rice resources; DT gene/quantitative trait locus (QTL) discovery; and DT rice breeding. More than 2000 rice accessions, mostly from China, were collected and evaluated in a new DT screening facility with a powerful water management system. Eighty-six entries were selected to serve as a core DT collection. A set of 187 recombinant inbred lines was developed for the genetic mapping of DT and high yield-related QTL under drought conditions. Several DT rice cultivars adapted to southern and central China were released. DT rice CMS lines were developed and distributed to most parts of China to encourage the development of DT or water-saving hybrid rice.

Applied genetics combined with practical plant breeding is a powerful tool in agricultural development and for food security. The Green Revolution spurred the world's potential to meet its food, feed and fibre needs at a time when vast regions were notoriously food-insecure. Subsequent adaptations of such strategies, from the late 1980s onwards, in order to develop new plant varieties in a more participatory way, have strengthened the focus on applying technology to farmers' diverse needs, feeding research results into a variety of seed systems. During these developments, there were no major legal impediments to the acquisition of either local or formal knowledge or of the building blocks of plant breeding: genetic resources. The emergence of molecular biology in plant science is creating a wealth of opportunities, both to understand better the limitations of crop production and to use a much wider array of genetic diversity in crop improvement. This ‘Gene Revolution’ needs to incorporate the lessons from the Green Revolution in order to reach its target groups. However, the policy environment has changed. Access to technologies is complicated by the spread of private rights (intellectual property rights), and access to genetic resources by new national access laws. Policies on access to genetic resources have changed from the concept of the ‘Heritage of Mankind’ for use for the benefit of all mankind to ‘National Sovereignty’, based on the Convention on Biological Diversity, for negotiated benefit-sharing between a provider and a user. The Generation Challenge Programme intends to use genomic techniques to identify and use characteristics that are of value to the resource-poor, and is looking for ways to promote freedom-to-operate for plant breeding technologies and materials. Biodiversity provides the basis for the effective use of these genomic techniques. National access regulations usually apply to all biodiversity indiscriminately and may cause obstacles or delays in the use of genetic resources in agriculture. Different policies are being developed in different regions. Some emphasize benefit-sharing, and limit access in order to implement this (the ‘African Model Law’), while others, in recognition of countries' interdependence, provide for facilitated access to all genetic resources under the jurisdiction of countries in the region (the Nordic Region). There are good reasons why the use of agricultural biodiversity needs to be regulated differently from industrial uses of biodiversity. The International Treaty on Plant Genetic Resources for Food and Agriculture, which entered into force in 2004, provides for facilitated access to agricultural genetic resources, at least for the crops that are included in the Treaty's ‘Multilateral System of Access and Benefit-sharing’. Ratification of the Treaty is proceeding apace, and negotiations have entered a critical stage in the development of practical instruments for its implementation. Although the scope of the Treaty is all plant genetic resources for food and agriculture, there are important crops that are not covered by its Multilateral System. Humanitarian licences are being used to provide access for the poor to protected technologies: countries may need to create such a general humanitarian access regime, to ensure the poor have the access they need to agricultural genetic resources.

Advances in genomics have provided technologies for high throughput analysis of plant genomes with potential for use in gene discovery in germplasm collections. The establishment of DNA banks facilitates this screening by making DNA from large numbers of plant accessions widely available. DNA banks require the development of appropriate policies for access and benefit sharing. Tools for automating sample and data handling are essential. Standard molecular methods for fingerprinting DNA accessions for international comparisons need to be determined. New screening technologies are required to take advantage of the emerging availability of large DNA collections. The Australian Plant DNA Bank aims to collect DNA from all Australian plant species and to sample the diversity within each species. DNA from all individuals of the species is being stored for rare species. Domesticated or economically important species from all countries are also being collected and stored. International networking of DNA banks will be a key step in linking genomics tools to global plant diversity.

There is a long-standing concern that modern plant breeding reduces crop genetic diversity, which may have consequences for the vulnerability of crops to changes in pests, diseases, climate and agricultural practices. Recent molecular assessments of genetic diversity changes in existing genepools of major agricultural crops may shed some light on the impact of plant breeding on crop genetic diversity. Reviewing published assessments revealed different impacts of plant breeding on improved genepools, not only narrowing or widening their genetic base, but also shifting their genetic background. In general, the genome-wide reduction of crop genetic diversity accompanying genetic improvement over time is minor, but allelic reduction at individual chromosomal segments is substantial. More efforts are needed to assess what proportion of lost alleles is associated with undesirable traits.

The effectiveness of association mapping (AM) based on linkage disequilibrium (LD) is currently being tested in a number of crops. An important prerequisite for the application of AM is the availability of collections of accessions with a suitable level of genetic variation for target traits and with limited spurious LD due to the presence of population structure. Herein, the results of a genome-wide molecular characterization of a collection of elite durum wheat accessions well-adapted to Mediterranean environments are presented. Ninety-seven highly polymorphic simple sequence repeats and 166 amplified fragment length polymorphism markers were used to characterize 189 durum accessions, mainly cultivars and advanced breeding lines. Genome-wide significant and sizeable LD indices at a centimorgan scale were observed, while LD mainly decayed within 10 cM. On the other hand, effects due to spurious LD were notably lower than those previously observed in a durum wheat collection sampling durum gene pools of more diverse origin. These results, coupled with the high level of genetic variability detected for a number of important morpho-physiological traits and their high heritability, indicate the suitability of this collection for AM studies targeting agronomically important traits.