Haplotype-resolved and gap-free genome of a floating aquatic plant from the Oryzeae tribe, Hygroryza aristata

Abstract

Objectives

Hygroryza aristata, an aquatic plant native to Southeast Asia, shows a high degree of adaptability to aquatic environments. H. aristata, which belongs to the Oryzeae tribe and is closely related to rice (Oryza sativa), holds potential for crop improvement, particularly in flood tolerance. This study aimed to sequence and assemble the genome of H. aristata.

Data description

We assembled the genome of H. aristata using 31.91 Gb of Pacific Biosciences (PacBio) High-fidelity (HiFi) data and 22.36 Gb of ultra long Oxford Nanopore Technology (ONT) data, resulting in two gap-free haplotype genomes, hap1 (349.74 Mb) and hap2 (347.98 Mb), each with 12 chromosomes and 23 telomeres. The continuity of chromosomes was supported by High-throughput chromosome conformation capture (Hi-C) data. The assemblies demonstrated high completeness, with > 99.8% of coverage rates, 98.4% of Benchmarking Universal Single-Copy Orthologs (BUSCO) scores, and > 11.0 of Long Terminal Repeat Assembly Index (LAI) scores per haplotype. RNA sequencing (RNA-seq) data (176.06 Gb) of six tissues was generated for genome annotation, identifying 39,139 and 38,746 protein-coding genes in hap1 and hap2, respectively.

Objective

Hygroryza aristata (Retz.) Nees ex Wight & Arn. is a perennial aquatic plant native to the wet tropical regions, distributed across Southern China and Southeast Asia [1, 2]. This species has been designated as a second-level nationally protected wild plant in China since 2021. H. aristata floats on the water surface, absorbing nutrients directly from the water and rapidly growing and spreading through stoloniferous growth. Morphologically, H. aristata produces adventitious roots and branches at the stem nodes, and its leaf sheaths turn purple and inflate, which contributes its floating (Data file 1) [3]. This species shows a high degree of adaptability to aquatic environments.

Taxonomically and evolutionarily, H. aristata (Hygroryza, Zizaniinae, Oryzeae, Poaceae) is closely related to rice (Oryza sativa), and they both belong to the Oryzeae tribe [4]. Unlike most grasses, which favor dry ecosystems, species of the Oryzeae tribe, including H. aristata, tend to adapt to aquatic environments [5, 6]. Especially, the floating trait is particularly rare within the Poaceae family.

Although the chloroplast and mitochondrial genomes of H. aristata have been previously reported [7, 8], its nuclear genome has not yet been sequenced. This study presents the complete genomes of the two haplotypes of H. aristata. These genomic resources are pivotal for elucidating the evolutionary adaptation to aquatic environments within the Oryzeae tribe, and will contribute to the gene resource development for rice, such as those related to flood tolerance.

Data description

Genomic DNA and RNA samples of H. aristata were extracted from plants clonally propagated from a single individual. Long-read sequencing of Pacific Biosciences (PacBio) High-fidelity (HiFi) and ultra-long (UL) Oxford Nanopore Technology (ONT) (read lengths > 100 kb), and short-read sequencing of High-throughput chromosome conformation capture (Hi-C), whole-genome sequencing (WGS), and RNA sequencing (RNA-seq), were performed (Data file 2) [3].

For genome assembly, 31.91 Gb of PacBio HiFi and 22.36 Gb of UL-ONT sequencing data sets were utilized (Data file 2; Data set 1) [3, 9]. Assembly was conducted using HiFiAsm (v0.20.0-r639) [10] under HiFi + UL-ONT mode with the following parameters: -l 3 -r 5 -a 6 -n 10 --ctg-n 10 -w 63 -k 63. Chromosome IDs and strand directions were determined by aligning the assemblies to the rice genome [11]. The resulting assemblies of unphased two haplotypes, designated as hap1 and hap2, were obtained (Data set 2, 3) [12, 13].

Both hap1 and hap2 are complete genomes, each comprising 12 chromosomes with genome sizes of 349.74 Mb and 347.98 Mb, respectively (Data file 3) [3]. Notably, both haplotypes are gap-free (Data file 3) [3]. Telomere detection using Seqtk telo (v1.4-r122) [14] displayed that each haplotype contains 23 telomeres (Data file 3) [3]. In conclusion, this study presents a haplotype-resolved and gap-free genome assembly.

To assess genome, genomic reads mapping, Benchmarking Universal Single-Copy Orthologs (BUSCO) [15] and Long Terminal Repeat Assembly Index (LAI) [16] methods were employed. Hi-C reads (83.06 Gb; Data file 2; Data set 1) [3, 9] were aligned to hap1 and hap2 using Juicer (v2.0) [17], respectively. Short-read WGS data (45.49 Gb; Data file 2; Data set 1) [3, 9] and long reads were aligned to the diploid genome assembly (hap1 + hap2, ~ 700 Mb) using bwa mem (v0.7.18-r1243) [18] and minimap2 (v2.28-r1209) [19], respectively.

The continuity of hap1 and hap2 chromosomes was supported by the Hi-C contact results (Data file 4) [3]. The alignments of WGS data achieved a 99.864% coverage rate with a mean depth of 63.1×, while PacBio HiFi and UL-ONT alignments reached 99.996% and 99.995% coverage rates with mean depths of 44× and 31.9×, respectively. The complete BUSCO scores were 98.4% for both hap1 and hap2 assemblies, and the LAI scores were 11.98 for hap1 and 11.19 for hap2.

For genome annotation, 176.06 Gb of RNA-seq data (Data file 2; Data set 1) [3, 9] and proteins from 16 genomes of 15 species in Poaceae (Data file 5) [3], were employed. Gene structure prediction was conducted using Braker3 (v3.0.8) [20], combining homology prediction, transcriptional evidences and ab initio prediction. Repeat annotation was conducted using EDTA (v2.0.1) [21]. The number of protein-coding genes of hap1 and hap2 were 39,139 and 38,746, respectively (Data set 2, 3) [12, 13]. The content of repeat sequences of hap1 and hap2 were 49.08% and 49.57%, respectively (Data set 4, 5) [3]. Finally, we annotated genes, including non-redundant protein database (NR), Universal Protein Knowledgebase (UniProt), Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) and Protein families database (PFAM), as well as other related public databases (Data set 6, 7) [3] (Table 1).

Table 1 Overview of data files/data sets

Limitations

Despite the high quality of the current genome assembly of H. aristata, there is still room for enhancement to achieve an error-free level. This can be accomplished by integrating various assembly outcomes from different assemblers, conducting fine revisions, and leveraging cytological experimental data, among other approaches.

The genome annotation of H. aristata can be further improved. For examples, non-coding RNAs have not been identified in the current version of the annotation. Future updates to the annotation are planned, which will incorporate data from third-generation full-length transcriptome sequencing of additional tissue samples to enhance the annotation.