Communications Biology (Sep 2023)

Improved pearl millet genomes representing the global heterotic pool offer a framework for molecular breeding applications

  • Punna Ramu,
  • Rakesh K. Srivastava,
  • Abhijit Sanyal,
  • Kevin Fengler,
  • Jun Cao,
  • Yun Zhang,
  • Mitali Nimkar,
  • Justin Gerke,
  • Sriram Shreedharan,
  • Victor Llaca,
  • Gregory May,
  • Brooke Peterson-Burch,
  • Haining Lin,
  • Matthew King,
  • Sayan Das,
  • Vaid Bhupesh,
  • Ajin Mandaokar,
  • Karunakaran Maruthachalam,
  • Pobbathi Krishnamurthy,
  • Harish Gandhi,
  • Abhishek Rathore,
  • Rajeev Gupta,
  • Annapurna Chitikineni,
  • Prasad Bajaj,
  • S. K. Gupta,
  • C. Tara Satyavathi,
  • Anand Pandravada,
  • Rajeev K. Varshney,
  • Raman Babu

DOI
https://doi.org/10.1038/s42003-023-05258-3
Journal volume & issue
Vol. 6, no. 1
pp. 1 – 11

Abstract

Read online

Abstract High-quality reference genome assemblies, representative of global heterotic patterns, offer an ideal platform to accurately characterize and utilize genetic variation in the primary gene pool of hybrid crops. Here we report three platinum grade de-novo, near gap-free, chromosome-level reference genome assemblies from the active breeding germplasm in pearl millet with a high degree of contiguity, completeness, and accuracy. An improved Tift genome (Tift23D2B1-P1-P5) assembly has a contig N50 ~ 7,000-fold (126 Mb) compared to the previous version and better alignment in centromeric regions. Comparative genome analyses of these three lines clearly demonstrate a high level of collinearity and multiple structural variations, including inversions greater than 1 Mb. Differential genes in improved Tift genome are enriched for serine O-acetyltransferase and glycerol-3-phosphate metabolic process which play an important role in improving the nutritional quality of seed protein and disease resistance in plants, respectively. Multiple marker-trait associations are identified for a range of agronomic traits, including grain yield through genome-wide association study. Improved genome assemblies and marker resources developed in this study provide a comprehensive framework/platform for future applications such as marker-assisted selection of mono/oligogenic traits as well as whole-genome prediction and haplotype-based breeding of complex traits.