The apricot (Prunus armeniaca L.) genome elucidates Rosaceae evolution and beta-carotenoid synthesis

Abstract

Apricots, scientifically known as Prunus armeniaca L, are drupes that resemble and are closely related to peaches or plums. As one of the top consumed fruits, apricots are widely grown worldwide except in Antarctica. A high-quality reference genome for apricot is still unavailable, which has become a handicap that has dramatically limited the elucidation of the associations of phenotypes with the genetic background, evolutionary diversity, and population diversity in apricot. DNA from P. armeniaca was used to generate a standard, size-selected library with an average DNA fragment size of ~20 kb. The library was run on Sequel SMRT Cells, generating a total of 16.54 Gb of PacBio subreads (N50 = 13.55 kb). The high-quality P. armeniaca reference genome presented here was assembled using long-read single-molecule sequencing at approximately 70× coverage and 171× Illumina reads (40.46 Gb), combined with a genetic map for chromosome scaffolding. The assembled genome size was 221.9 Mb, with a contig NG50 size of 1.02 Mb. Scaffolds covering 92.88% of the assembled genome were anchored on eight chromosomes. Benchmarking Universal Single-Copy Orthologs analysis showed 98.0% complete genes. We predicted 30,436 protein-coding genes, and 38.28% of the genome was predicted to be repetitive. We found 981 contracted gene families, 1324 expanded gene families and 2300 apricot-specific genes. The differentially expressed gene (DEG) analysis indicated that a change in the expression of the 9-cis-epoxycarotenoid dioxygenase (NCED) gene but not lycopene beta-cyclase (LcyB) gene results in a low β-carotenoid content in the white cultivar “Dabaixing”. This complete and highly contiguous P. armeniaca reference genome will be of help for future studies of resistance to plum pox virus (PPV) and the identification and characterization of important agronomic genes and breeding strategies in apricot.

Chromosome-level and haplotype-resolved genome assembly enabled by high-throughput single-cell sequencing of gamete genomes

“…We aligned protein sequences from the database UniProt/Swiss-Prot, Arabidopsis thaliana and Prunus persica to each haplotype assembly using the tool Exonerate 43 with the options “--percent 60 --minintron 10 --maxintron 60000”. We mapped RNA-seq reads from reproductive buds, vegetative buds, flowers, leaves, fruits (except seeds) and bark tissues, as well as a published Apricot RNA-seq dataset 23 , using HISAT 44 , and we assembled the transcripts using StringTie 45 . Finally, we used the tool EvidenceModeler 46 to integrate the above evidence in order to generate consensus gene models for each haplotype assembly.…”

“…All genome assemblies, including ‘Rojo Pasion’ haplotypes, ‘Chuanzhihong’ apricot ( Prunus armeniaca ) 23 , Japanese apricot ( Prunus mume ) 25 and ‘Lovell’ peach ( Prunus persica ) 26 , were aligned to each other using nucmer from the MUMmer4 51 toolbox with parameters ‘-max -l 40 -g 90 -b 100 -c 200’. The alignments were further filtered for alignment length (>100 bp) and identity (>90%), with which structural rearrangements and local variations were identified using SyRI 12 .…”

“…Despite high levels of synteny, the two assemblies revealed large-scale rearrangements (23 inversions, 1,132 translocation/transpositions and 2,477 distal duplications) between the haplotypes making up more than 15% of the assembled sequence (38.3 and 46.2 Mb in each of assemblies; Supplementary Table 1). Using a comprehensive RNA-seq dataset sequenced from multiple tissues of ‘Rojo Pasion’ including reproductive buds, vegetative buds, flowers, leaves, fruits (seeds removed) and barks as well as a published apricot RNA-seq dataset 23 , we predicted 30,378 and 30,661 protein-coding genes within each of the haplotypes (with an annotation completeness of 96.4% according to a BUSCO 27 analysis). Mirroring the huge differences in the sequences, we found the vast amount of 942 and 865 expressed, haplotype-specific genes in each of the haplotypes (Methods; Supplementary Tables 2-3).…”

“…Having proved the haplotype accuracy, the assemblies were polished resulting in final assemblies (N50: 25.5 Mb and 25.8 Mb; Table 1; Methods). To further assess the quality of the scaffolded chromosome structure, we compared our assemblies with recently assembled genomes, including those of very closely-related species such as the heterozygous ‘Chuanzhihong’ apricot ( Prunus armeniaca ) 23 and the Japanese apricot ( Prunus mume ) 25 , and a more distantly-related species, peach ( Prunus persica: doubled-haploid genome) 26 , using SyRI 12 (a tool designed for the comparison of chromosome-level assemblies). Our assemblies showed high consistency in the synteny to these assemblies, reflecting the reliability of the genetic map and the assembled genome structure (Fig.…”

The genome of low-chill Chinese plum ‘Sanyueli’ (Prunus salicinaLindl.) provides insights into the regulation of the chilling requirement of flower buds

“…In the present study, a 308.06 Mb genome of low-chill Chinese plum 'Sanyueli' with N50 of 815.7 kb was generated. The presence of the majority of BUSCO genes (97.01%) in our assambly indicated that the quality of 'Sanyueli' plum genome is comparable to that of sweet cherry (96.0%) 64 , almond(96.0%) 66 and apricot(98.0%) 68 . The genome size of 'Sanyueli' plum is larger than that of mume 62 , peach 63 , almond 66 and apricot (38.28%) 68 .…”