A new super-pangenome pipeline reveals domestication signatures of conserved noncoding sequences in the orange subfamily

Advances in genome sequencing have accelerated the assembly of hundreds of plant genomes. Because of limited tools, current pangenome approaches overlook conserved noncoding sequences, though these sequences potentially harbor diverse regulatory elements. Here, we present panCG, an integrative pipeline for family level super-pangenome analysis across coding and noncoding sequences. PanCG enables the reference-free construction of pangene and pan-conserved noncoding sequences (panCNS) indices and integrative analyses between them. Applying this pipeline to the orange subfamily (Aurantioideae) with 22 representative genomes, we identified 167,989 panCNSs, 44.63% of which exhibited potential functionality, as evidenced by multiomic data (eg ATAC-seq, ChIP-seq). Functionally, approximately half of duplicated genes showed expression divergence, likely mediated by sequence variants in these panCNSs. Although transposable element (TE) mobilization resulted in substantial CNS loss, it also contributed to the origin of a small portion of functional CNSs. By integrating panCNS and pangene data, we identified 1,391 conserved colocalized blocks of CNSs and genes across all selected genomes, providing clear evidence for coevolution between CNSs and their target genes. Notably, leveraging these pan-genomic resources, we pinpointed numerous CNSs and gene-CNS regulatory modules as candidates underlying the domestication of citrus fruit traits from wild relatives. Our study presents a powerful tool for interspecific super-pangenome analysis, whereas shedding new light on the functional roles and evolutionary dynamics of CNSs in citrus, and provides a valuable resource of candidate regulatory elements for citrus fruit biology and breeding research.