Global food systems are overly reliant on a few staple crops, making them vulnerable to environmental and social instability. Diversifying food systems by improving underutilized orphan crops is crucial. *Lablab purpureus* (lablab) is a climate-resilient, multipurpose legume native to Africa, cultivated for food and forage. It's drought-tolerant and rich in bioactive compounds. Genome-assisted breeding can accelerate lablab improvement. While genomes of many crops have been sequenced, African researchers are underrepresented in sequencing their indigenous crops due to limited resources and infrastructure. This study addresses this underrepresentation by presenting an Africa-led collaborative effort to generate a high-quality lablab genome.

Previous work on lablab has included a draft genome assembly using short-read sequencing, but this lacked the resolution for a comprehensive understanding of its genomic architecture. Studies have suggested a dual domestication origin for lablab, with two-seeded and four-seeded gene pools. Research has also highlighted the antinutritional effects of trypsin inhibitors in lablab, impacting its use as feed and fodder. This research builds on this existing knowledge base, aiming to provide a more complete genomic resource for lablab improvement.

The researchers used Oxford Nanopore Technology (ONT) MinION for cost-effective genome sequencing of lablab cv. Highworth. This generated 4.7 million reads with a mean length of 6.1 kb, achieving 67x coverage. The reads were assembled into contigs, polished using Illumina short reads, and scaffolded into 11 pseudomolecules using Hi-C data. Gene annotation employed a pipeline combining protein homology, transcript evidence, and ab initio predictions, resulting in 30,922 gene models. Repeat annotation identified 168,174 transposable element sequences. Gene family analysis compared lablab with other legumes using orthogroup analysis and CAFE for gene family expansion and contraction. Trypsin inhibitor genes were cataloged and analyzed for genomic organization and expression. Resequencing of wild and domesticated lablab accessions investigated the dual domestication hypothesis. Finally, a collection of 203 lablab accessions was analyzed using DArTseq for population structure, diversity, and genome-wide association studies (GWAS).

The study produced a chromosome-scale assembly of the lablab genome (417.9 Mbp, N50 of 38.1 Mbp), significantly improving upon previous drafts. The assembly's high quality was confirmed by BUSCO scores (98.5% and 98.2% against embryophyta and fabales, respectively) and a high LTR Assembly Index (LAI) of 19.8. Analysis of gene families revealed 119 lablab-specific orthogroups enriched for fatty acid biosynthesis and arabinose metabolism. Expanded gene families were enriched for lignin and pectin metabolism and photosynthesis, potentially related to drought tolerance. The trypsin inhibitor gene family, an antinutritional factor, was found to be organized in five clusters, primarily on chromosomes Lp01, Lp04, Lp06, and Lp11, with a major cluster on Lp04. Resequencing confirmed the dual domestication origin of lablab, with clear separation between two-seeded and four-seeded gene pools. Genetic diversity was higher in the four-seeded group. Analysis of 203 lablab accessions revealed two main population clusters, largely corresponding to the two-seeded and four-seeded gene pools, with further sub-structuring within the four-seeded group. GWAS identified 18 markers associated with several agronomic traits, explaining 7–24% of phenotypic variation.

The high-quality chromosome-scale assembly of the lablab genome addresses the need for a comprehensive genomic resource for this important orphan crop. The findings on trypsin inhibitor gene organization provide targets for breeding programs aimed at reducing antinutritional factors while maintaining plant defense. Confirmation of the dual domestication origin highlights the importance of considering the unique genetic diversity within each gene pool for breeding strategies. The population structure analysis and GWAS results identify valuable markers for marker-assisted selection, accelerating the improvement of key agronomic traits. This research contributes significantly to lablab breeding efforts and enhances our understanding of legume domestication.

This study delivers a high-quality lablab genome assembly, significantly advancing our understanding of this important orphan crop. The findings on gene family evolution, trypsin inhibitor gene organization, and population structure provide valuable resources for breeding programs. The Africa-led collaborative model sets a precedent for future genomics research in low- and middle-income countries, emphasizing inclusive research practices. Future research could focus on functional characterization of identified genes, exploring the genetic basis of drought tolerance, and refining marker-assisted selection strategies for lablab improvement.

The study primarily focused on lablab cv. Highworth for genome assembly, potentially limiting the generalizability of some findings to other lablab varieties. While the GWAS identified markers associated with agronomic traits, further validation and fine mapping are needed to identify the causal genes. The phenotypic data used for GWAS was collected across several years, introducing potential environmental variation as a confounding factor.