2017_pearlMillet_genome_NatureBiotechnology
salinity and limited rainfall. Pearl millet reliably produces grain in regions that have a mean annual precipitation as low as 250 mm. In the same drought conditions maize (Zea mays), rice (Oryza sativa), sorghum (Sorghum bicolor), bread wheat (Triticum aestivum) and durum wheat (Triticum durum) are likely to fail5.
Pearl millet is cultivated on ~27 million hectares worldwide and is the staple food for more than 90 million farmers living in poverty. Millet grain is highly nutritious, with 8–19% protein, low starch, high fiber (1.2 g/100 g)6, and higher micronutrient concentrations (iron and zinc) than rice, wheat, maize and sorghum7. Importantly, the potential of this crop to tolerate air temperatures >42 °C during the reproductive phase means that it can be cultivated using irrigation in the very hot summers of northwestern India8.
Despite the clear importance of pearl millet in agriculture, the production and productivity of this staple crop are very low, with an average grain yield of just 900 kg/ha. This is because pearl millet is mainly grown in dryland conditions, which are marginal production environments, and with minimal use of commercial inputs, such as, adequate irrigation, fertilizers and pesticides. Genetic gains, the rate of increase in yield over a given time period, during 1996–2013 in pearl millet have averaged around 24 kg of grain/ha/year in India, which has the highest millet productivity and production of the main pearl millet growing countries9. Pearl millet is vulnerable to several foliar diseases including downy mildew (caused by Sclerospora graminicola), Pyricularia leaf spot or blast (Pyricularia grisea), and rust (Puccinia substriata var. indica). Indeed, these pathogen infections can result in massive yield losses and reduced fodder quality. A limited range of genomics tools for pearl millet have impeded the ability of researchers and breeders to exploit methods for improvement, until now.
To accelerate pearl millet crop improvement, we sequenced the whole genome of reference genotype Tift 23D2B1-P1-P5. We also resequenced 994 pearl millet genotypes, including 963 inbred lines and single plants from each of 31 wild accessions, in order to understand the population structure, genetic diversity and domestication of this staple crop. We carried out a genome-wide association study (GWAS) to predict yield-associated traits in both irrigated and drought conditions. We also used genomic prediction to predict hybrid performance. These applications highlight the utility of our resequencing data set for accelerating breeding and enhancement of genetic gains in pearl millet.
