Chickpea Innovation Lab Overview, Activities and Highlights
presented at the 4th Annual Meeting, December 2017


Download the entire program -- includes agenda, overview, abstract and participant list.

Chickpea Innovation Lab Dec 2017 Program Final.pdf


Chickpea Innovation Lab Overview

Chickpea is a pulse legume of critical importance in low-income food insecure countries, in advanced developing economies, and in developed countries. Paradoxically, countries with the highest nutritional demand for chickpea are also those with the lowest yields, often ½ to ¼ of yields found in the developed world. Whole genome sequencing reveals that ~95% of genomic variation was lost from modern elite cultivars during domestication. This has profound implications, because corresponding reductions to trait variation limit the ability to adapt the crop to changing environments and to meet emerging needs, raising an urgent need for new sources of diversity.

We address these needs by harnessing the expanded genetic potential of chickpea’s wild relatives, focusing on traits related to tolerance to biotic and abiotic stress, improved seed nutrient density, agronomic characters and symbiotic nitrogen fixation. As described in this report, we have collected and are characterizing a large and systematic set of wild Cicer species from a representative range of natural environments. Genomic technologies have been used to characterize genetic diversity among ~1,162 wild accessions and to nominate particular wild plant accessions as targets of detailed phenotyping and breeding. Phenotyping of wild genotypes reveals variation for traits that are high priorities for crop improvement, including variation in flowering time, resistance to fungal diseases and insect pests, nitrogen fixation and drought tolerance, while genome wide association studies (GWAS) identify loci underlying wild traits.

In parallel we have surveyed diversity among ~1,000 landraces of cultivated chickpea, collected from the major historical centers of crop diversity in Ethiopia, India, Pakistan and Turkey, including material collected one century ago by the renowned crop geneticist Nikolay Vavilov. We have determined that landraces represent an additional reservoir of diversity, largely absent from modern elite genotypes. In combination with recurrent phenotyping conducted at the Vavilov Research Institute, and with global information system (GIS) technologies to extract climate data, we are using GWAS studies to identify trait-associated genomic variation and to prioritize that variation for breeding.

Our wild germplasm collection forms the basis of a large program of genetic crossing that is introducing wild genetic diversity into phenology-normalized backgrounds of elite cultivated genotypes. The base populations involve twenty-three diverse wild donor accessions crossed into six cultivated elite varieties. To date, ~10,000 unique wild x cultivated lineages have been produced directly by the Innovation Lab, with several thousand additional lineages produces by partners. Thousands of lineages and tens of thousands of seed have been distributed, grown and phenotyped for a range of agronomic traits under the prevailing field conditions in India, Ethiopia, the US and Russia. All progeny derived from crosses to one early flowering recurrent cultivated parent (ICCV96029) have been genotyped by sequencing, and phenotyped at F2, F3 and F4 stages for a variety of traits. GWAS and QTL analyses have identified haplotypes that underlie traits, with the goal of rapid varietal improvement based on knowledge of valuable alleles. Current efforts focus on domestication and yield related traits (e.g., pod shattering, biomass conversion efficiency, seed size, total seed production), while a second generation of traits is under analysis and yielding a range of extremely interesting trait values (e.g., for drought and heat tolerance, pod borer pest resistance, Ascochyta blight resistance, nitrogen fixation, flowering time and plant architecture).

With the objective of developing inoculum for nitrogen-fixation that can be delivered to farmers, we have used hierarchical sampling and genomics to understand the genomic diversity of endemic strains of chickpea’s symbiont, Mesorhizboium spp. >1,500 pure strain and metagenome sequencing experiments have identified >20 species of Mesorhizobium that nodulate chickpea globally, compared to three species identified from surveys at the crop’s center of origin in Mesopotamia. Our results indicate that subsequent to movement of the crop from Mesopotamia over the past six thousand years, horizontal gene transfer of a symbiotic gene cassette, representing 10% of the Mesorhizboium genome, recruited new species of symbiotic microbes to chickpea. Thus, 90% of the collective (>20 species) symbiont genome is highly diverse and we aim to understand and harness this diversity for improved crop performance. Towards that end, we have conducted and continue extensive greenhouse and field trials. We have developed methods to quantify symbiotic performance and to identify bacterial strains and host genotypes that yield optimal nitrogen fixation. A representative subset of Mesorhizobium species are under test in field trials using commercial grade inocula produced by our corporate partner Rizobacter, with the goal of providing farmers with superior microbial solutions for nitrogen fixation.

The impact and capacity of the Chickpea Innovation Lab has been increased by leveraging funding from a range of sponsors, including new partners and sponsorships from government (BBSRC in the United Kingdom to the University of Edinburgh) and corporate (Nestle S.A.) entities. These new sponsorships provide the means to expand efforts to include analysis of root architecture, which is a component of stress adaptation in many crops, and seed protein nutrition. In total, the project has increased the original USAID investment 3-fold, with ~$17.5 million funding from 21 sponsors*.

The Chickpea Innovation Lab also has an active training and capacity building component. The efforts of thirty-nine students, postdoctoral scientists and visiting professors (20 males and 19 females) are supported by core USAID resources or by leveraged funding. Among 21 PhD and 4 Masters students, 20 are funded for the duration of their degrees. The largest contingent is 10 Ethiopian scholars, including 8 PhD and 1 Masters student and a postdoctoral scholar, primarily funded by USAID resources. Five Indian scholars include three postdocs, one PhD student and one visiting professor that are funded primarily by leveraged resources. Short term exchange visits of up to 6 months form the basis of an active student exchange program among partners in the US, India, Ethiopia and Turkey. During visits to advanced research institutes (ARIs), our students gain experience with specialized infrastructure and advanced analytical capacities, including breeding and phenotyping, genomics and bioinformatics, and statistical methods; conversely, short term visits and field work in Ethiopia and India exposes students and scholars from ARIs to the real world challenges that their efforts are intended to address.

Outcomes of this project will be high-yielding, climate-resilient chickpea varieties within the context of user-preferred traits: climate resilience, seed quality and nutrient density, reduced inputs through nitrogen fixation, and biotic stress resistance among them. We have a clear focus on research-for-development, with all upstream activities (i.e., germplasm collection, genomics and population development) predicated on the need to facilitate downstream phenotyping and breeding activities. In the course of this work we aim to identify and introduce newly collected wild alleles into diverse high performing elite cultivars that increase crop productivity, food and nutritional security for smallholder farmers. The Chickpea Innovation Lab’s long-term impact will derive from the value of the biological tools and resources that we develop, the genomic and phenotyping data that we archive and make available, and the increased capacity for research and agricultural applications that our students acquire and express.

*Thanks to our sponsors: 1USAID (Washington, India and Ethiopia), 2US National Science Foundation, 2USAID-sponsored Partnership for Enhanced Engagement in Research, 2USAID-sponsored Legume Innovation Lab, 2USAID-sponsored US Borlaug Fellows in Global Food Security program, 2USAID-sponsored US-Pakistan Center for Advanced Studies, 2Pakistan Higher Education Commission, 2Global Crop Diversity Trust, 2TATA Trusts, 2Two Blades Foundation, 2Mars Incorporated, 2CGIAR Consortium Research Program 3.5 on Grain Legumes, 2Indo-US Science and Technology Forum, 3Australian Grains Research and Development Corporation, 3Saskatchewan Ministry of Agriculture, 3Saskatchewan Pulse Growers, 3Biotechnology and Biological Research Council (BBSRC) of the UK, 2Nestle SA, 3,4Rizobacter, 3Fulbright Visiting Scholars Program, 3Chinese Scholarship Council.

1Chickpea Innovation Lab activities funded by USAID resources; 2Chickpea Innovation Lab activities funded and managed by the Lab using leveraged resources; 3Partner activities funded by non-Chickpea Innovation Lab resources, may or may not involving UC Davis oversight; 4in-kind support, not counted in total funding.


2017 Program Activities and Highlights

Introgression population development ­– Among ~10,000 project-generated independent lineages (twice that proposed in the original project), roughly 5,700 lineages were produced at UC Davis and 4,500 were produced at Harran University in Turkey. International partners in Canada and Australia produced several thousand additional lineages. Populations range in development from F3 to F5. To simplify analysis, we have focused genotyping and phenotyping on progeny of a single cultivated parent (an early flowering Indian genotype “ICCV96029”) crossed with the panel of wild donor genotypes. ~2,500 independent F3 lineages from this population were imported to India and Ethiopia and advanced to F4 in the fall of 2016, with a subset is under advancement to F5 in the fall of 2017.

Drought phenotyping ­– We have quantified significant variation in wild donor genotypes for physiological responses to limiting soil moisture (classical drought) and to vapor pressure deficit (atmospheric drought). Wild genotypes are often more tolerant of soil and atmospheric drought than the best tolerant cultivated checks. Interestingly, as we hypothesized in our original proposal, the tolerance of wild accessions to moisture deficit is correlated with moisture availability at field locations where they originated. We are currently exploring drought responses and related whole plant phenotypes in segregating pre-breeding populations: 473 phenology-normalized F3 families have been tested on the field-based lysimetry platform and data are under analysis. Transcriptional profiling reveals differences in drought responses between wild and cultivated accessions, and between tolerant and susceptible genotypes, providing a complement to the ongoing physiological and breeding work.

Pod borer phenotyping ­– A lack of resources prevented further progress on pod borer breeding. However our earlier observations of high levels of pod borer resistance in wild genotypes, combined with the magnitude of pod borer losses in India and Ethiopia, make developing genetic resistance against this pest an extremely high priority where we believe we can make a significant positive impact.

Heat tolerance ­– Field testing at ICRISAT has identified high levels of heat tolerance (seed set at >40°C) in our pre-breeding populations (249 accessions among 480 F3:F4 families). Trait values in segregating populations, calculated as percentage seed set under typically non-permissive temperatures, often significantly exceeded those of the most tolerant cultivated checks. Selected accessions are being prepared for additional field testing in Ethiopia and India.

Ascochyta blight ­– Two field trials were planted at EIAR field sites in Alementa and Dera, Ethiopia. These field locations are characterized by high Ascochyta blight pressure, which was unusually extreme in the late summer of 2017. Representative families from our pre-breeding populations were planted in replicate between the two field sites. We observed a range of disease tolerance phenotypes (tolerant to immune) in a subset of F4 families, with the same families performing well at both locations. These results are consistent with simple genetic control of Ascochyta resistance, which if so should be amenable to breeding. Importantly, tolerance is present in accessions derived from both Cicer reticulatum and C. echinospermum, suggesting the likelihood of distinct genes. The availability of distinct genes would enable staking of genes to increase the durability of resistance phenotypes, which is a long sought after objective, especially in the highly susceptible and farmer-preferred Desi varieties. We have obtained pathogen isolates from these field locations and we are preparing for more detailed phenotyping of the host and genomic analyses of the pathogen.

Nitrogen fixation ­–  ~85 additional symbiont strains were isolated in pure culture and their genomes sequenced. New strains from Ethiopia are primarily from low pH soils with the objective of pH resistant inocula, while strains from India represent the beginning of a living strain collection that will enable inoculum trials with endemic diversity. Our corporate partner, Rizobacter, is developing commercial grade liquid inoculum, providing capacity to uniformly inoculate 100’s of thousands of seed. Sachets of inoculum were delivered to UC Davis and to our partners in Ethiopia, where a field trial is currently ongoing. We have developed greenhouse methods to quantify nitrogen fixation effectiveness and we are exploring host factors that determine effective symbiosis, providing a rational basis to evaluate and compare additional strains as a prelude to field testing. One challenge is the need to scale these assays to many additional strains and host genotypes, with the goal of achieving predictive power from genome affinities (doing so would have large implications for effective use of inocula).

Fusarium genomics and phenotyping ­– With the effort of two of our Ethiopian students and funding from the Two Blades foundation, we completed the DNA sequencing of 284 Fusarium oxysporum genomes that we collected in a systematic survey of chickpea growing regions in Ethiopia. We tested and optimized a bioinformatics pipeline for genome assembly and analysis. Genomic and phylogenetic diversity analyses, combined with GIS coordinate of strain origins, allowed us to reject the hypothesis that genomic diversity is related to geographic distribution. Multi-locus phylogenies with ~2,000 conserved genes provide an unprecedented view of species diversity, which we anticipate can guide breeding strategies in the crop.

Whole genome sequencing ­– We have developed error corrected Pacific Biosciences assemblies for each of our target species (C. arietinum, C. reticulatum and C. echinospermum). These genomes are significantly improved over previously published genomes and should greatly facilitate the use of genomics in crop improvement. They contain 34%-60% more DNA sequence within the non-gapped contig fraction, and an order of magnitude increase in sequence contiguity. Efforts are currently underway to scaffold these assemblies using optical mapping for which the data is complete; high-density linkage mapping is ongoing to anchor, order and orient these molecules. Genome annotation is being conducted in partnership with the National Center for Biotechnology Information.

Trait-marker associations ­– We have made significant progress in the discovery of yield-related traits through genome wide association studies (GWAS) and QTL analysis. DNA was extracted from 5,700 F2 individuals, and ~2,500 progeny were genotyped using genotyping-by-sequencing approaches. An initial round of SNP discovery was conducted using an older genome draft of the Cicer arietinum genome (Nature Biotechnology 2013). With that data we have filtered for fixed differences between wild and cultivated accessions and performed GWAS, discovering major loci for seed shattering, 100 seed weight, biomass conversion efficiency, plant architecture, total biomass and total seed yield. These data should increase in precision and accuracy as we begin to incorporate the improved genome drafts that we have developed.

Data management ­– All sequencing data can be found in the NCBI Umbrella BioProjectPRJNA353637, which includes Illumina data from genotyping-by-sequencing (GBS) data for of over 1,000 wild Cicer accessions (PRJNA416006), landraces (PRJNA396092), whole genome sequencing (WGS) for 280 plant accessions (PRJNA416007), the trace reads for Fusarium genomes (PRJNA412392), and reference genome assemblies for C. arietinum (PRJNA418058), C. reticulatum (PRJNA418059) and C. echinospermum (PRJNA418060). The BioProject will also host Illumina data from GBS sequencing for our F2-derived pre-breeding lines, WGS assemblies and sequencing data for strains of Mesorhizobium spp. and Fusarium oxysporum f.sp. ciceris, and RNAseq data for gene expression. Most data currently available are the trace archives, however as assemblies are available they are uploaded, and updated as appropriate; for example, then initial assemblies for the three reference Cicer genomes are available. Our policy is to make all data available in the public domain without waiting for publication. Moreover, rather than releasing trace archive data only, we provide assembled genomes. Genome annotation is being performed by NCBI, facilitated in part by transcript data that we are providing. Any request for data is met by making that data available immediately, even if the data are not yet organized for submission to NCBI.

Germplasm management ­– Germplasm management is a major challenge given the size of our collection. Mars Inc has provided gifts to purchase instrumentation that will aid long-term seed preservation at UC Davis, though personnel to accomplish this task limit us. All seed is minimally at F3, and portions of the collections have been increased to F4 and F5 stages at ICRISAT in India and at EIAR in Ethiopia. Discussions are underway about a possible partnership in the area of germplasm development and curation.

Communications ­– Chickpea IL’s website ( continues to be updated.

Outreach ­– In partnership with the Indian Agricultural Research Institute, we hosted a meeting of US and Indian scientists in Delhi: “Indo-US Bilateral Workshop Genomic Approaches for Yield Enhancement and Biological Nitrogen Fixation in Chickpea”, January 29-31, 2017. Participants included 47 Indian scientists and 9 US scientists. In conjunction with the Ethiopian Value Chain Activity (a USAID funded project), the Chickpea Innovation Lab hosted a workshop in Ethiopia, brining together representatives from private, public, non-governmental and university organizations, to discuss needs and opportunities related to development of an inoculum industry in Ethiopia.