Background & Summary

Advances in sequencing technologies have enabled researchers to gain deeper insights into genome structure, function, and evolution. Despite the availability of some genomic resources for the oriental armyworm (Mythimna separata)1,2, this pest remains a persistent and evolving challenge in agricultural pest management. Renowned for its destructive potential, M. separata causes severe damage to a wide range of crops, including rice, maize, and wheat, across Asia and Australia3,4. It is recognized as one of the most significant polyphagous and invasive insect species.

Global shifts in climate, crop planting patterns, varietal distribution, and cultivation systems have driven M. separata to develop new patterns of adaptability5. These changes have enhanced its ability to thrive in diverse environmental habitats, further intensifying its threat to agricultural systems. The migration behavior of M. separata plays a key role in its outbreaks, enabling its polyphagous impact across vast geographical regions. Its range extends from northern areas, including Korea and Japan, to southern regions such as the Indochina Peninsula, the Philippines, and Malaysia6,7. While limited cold tolerance prevents M. separata from overwintering in most regions, seasonal migrations allow it to reach suitable areas for survival and reproduction8,9. These biological traits highlight the importance of M. separata as a critical subject for in-depth research.

In this study, we present an enhanced chromosome-level 714.5 Mb genome assembly for M. separata, including the complete assembly of both sex chromosomes, featuring the longest and most complete W chromosome reported for this species to date. This genome version offers high sequence accuracy and improved contiguity. Additionally, we developed markers for sex detection based on W chromosome-specific sequences, enabling the identification of M. separata’s sex even at the larval stage. By addressing gaps in existing genomic data, this improved assembly provides a robust foundation for understanding the genetic architecture underlying M. separata’s environmental adaptability and offers valuable resources for advancing pest management strategies.

Methods

Sample information

The Mythimna separata samples used in this study were derived from an inbred strain reared on a noctuid artificial diet under controlled laboratory conditions (25 ± 2 °C, 14L:10D photoperiod, and 75 ± 5% relative humidity) at the Agricultural Genomics Institute of Chinese academy of Agricultural Sciences at Shenzhen. A single-pair mating approach was employed to produce subsequent generations, followed by consistent sibling mating to ensure genetic homozygosity. After four successive generations, one female pupa was selected for genomic sequencing using PacBio HiFi and Illumina technologies, while another female pupa from the same generation was chosen for Hi-C sequencing. Additionally, individuals from the same generation, representing various developmental stages, were collected for transcriptome sequencing (Table 1). These included larvae at the 2nd, 3rd, 4th, and 5th instars, as well as one male and one female pupa, and one male and one female adult.

Table 1 Overview of sequencing reads utilized in this study.
Full size table

Genome sequencing and assembly

Genomic DNA (gDNA) was extracted using the DNeasy Blood & Tissue Kit (Cat. no. 69506, Qiagen). The quality and quantity of the gDNA were assessed using a NanoDrop One UV-Vis spectrophotometer (Thermo Fisher Scientific) and a Qubit 3.0 Fluorometer (Invitrogen), following the manufacturers’ protocols. Approximately 0.5 μg of gDNA was used to generate a PCR-free Illumina genomic library with a 350 bp insert size, prepared using the TruSeq Nano DNA HT Sample Preparation Kit (Illumina, USA). This library was sequenced in 2 × 150 bp format on the Illumina NovaSeq 6000 platform, generating 46.59 Gb of raw Illumina data, achieving a 68.41× sequence coverage of the estimated M. separata genome (Table 1). After quality control with fastp (v0.20.1)10 using default parameters, the clean reads were utilized to construct a 17-mer frequency distribution map using Jellyfish (v2.3.1)11. The genome size of M. separata was subsequently estimated to be 681 Mb using GenomeScope (v1.0)12.

Additionally, 5 μg of gDNA from the same individual was used to create ~15 kb SMRTbell insert libraries, which were sequenced on the PacBio Sequel II system (Pacific Biosciences, USA). A total of 38.71 Gb of clean PacBio HiFi reads was generated, representing 56.89× coverage of the estimated genome (Table 1). The PacBio HiFi reads were then assembled using Hifiasm (v0.3.0)13 with default parameters. Haplotigs were removed using Purge_Dups (v1.2.3)14. The initial genome was assembled into 121 contigs with a total size of 714.46 Mb. The contig N50 was measured 20.45 Mb and the maximum contig length was 25.71 Mb.

Hi-C sequencing and chromosome scaffolding

Hi-C library construction was carried out using the same female pupa used for DNA extraction. The DNA was cross-linked in situ, extracted, and digested with the restriction enzyme DpnII. The resulting fragments were ligated to form chimeric junctions, followed by purification and amplification of the Hi-C libraries using 12–14 cycles of PCR. The libraries were sequenced on the Illumina NovaSeq 6000 platform using a 150 bp paired-end configuration, producing 64.01 Gb of Hi-C read data.

Low-quality raw reads were removed using fastp v0.20.110 with default parameters. The purified reads were then mapped to the assembled contigs using Juicer (v1.5)15. The 3D-DNA pipeline16 was employed for contig clustering, ordering, and phasing, resulting in preliminary chromosomal scaffolds. The chromosome interaction matrix was further manually adjusted using JuiceBox v1.11.0817 at a resolution of 500 kb (Fig. 1). Through Hi-C scaffolding, the final genome assembly measured 714.5 Mb, with a scaffold N50 of 22.7 Mb and a GC content of 38.8%. A total of 32 pseudochromosomes were obtained, representing 97.2% of the total genome size.

Fig. 1
figure 1

Chromosomal architecture of M. separata. (a) Hi-C contact heatmap displaying genomewide chromatin interactions across 32 pseudochromosomes. The intensity along the diagonal patches reflects the frequency of physical interactions between chromatin regions. (b) Karyotypic analysis of the diploid chromosome complement, illustrating a total of 62 chromosomes.

Full size image

To identify the sex chromosomes from the assembled pseudochromosomes, a synteny analysis was performed by comparing the Mythimna separata genome with those of Spodoptera litura (GCA_002706865.1) and Noctua pronuba (GCA_905220335.1). BLAST v2.226+ was used with a stringent E-value cutoff of <1E−10 to align the amino acid sequences of protein-coding genes from these species. Syntenic blocks were then constructed using MCScanX18 with default parameters, and the relationships were visualized using the TBtools (v2.008)19 package. The analysis revealed that Chr1 corresponds to the Z chromosome, with a length of 34.92 Mb, while Chr2 corresponds to the W chromosome, with a length of 30.55 Mb (Fig. 2). This represents the longest and most accurate W chromosome sequence reported to date (Table 2).

Fig. 2
figure 2

Syntenic relationships between chromosomes of M. separata and those of S. litura (a) and N. pronuba (b). The sex chromosomes, W and Z, are highlighted in red text for each species. Notably, the W chromosome of M. separata lacks a corresponding counterpart in S. litura but aligns with a chromosome in N. pronuba.

Full size image
Table 2 Comparative characteristics of M. separata genome assemblies.
Full size table

RNA sequencing and genome annotation

Total RNA was extracted using the TransZol Up Plus RNA Kit (Cat. no. ER501-01-V2, TransGen), following the manufacturer’s protocol. The integrity and purity of the RNA were evaluated using the same methods employed for gDNA quality assessment, including a NanoDrop One UV-Vis spectrophotometer and a Qubit 3.0 Fluorometer. Indexed cDNA libraries were constructed from the extracted RNA using the NEBNext Ultra RNA Library Prep Kit for Illumina. Libraries with insert sizes of 250–300 bp were sequenced on the Illumina NovaSeq 6000 platform with a paired-end 150 bp strategy, yielding a total of 64.01 Gb of sequencing data.

Several comprehensive approaches were utilized to predict and annotate the genomic features of the M. separata genome. The repeat element identification followed a two-step process. First, a de novo repeat library was constructed using RepeatModeler (v1.0.11)20 based on the assembled genome, with default parameters. Repeat elements were then identified through homology searches against this library using RepeatMasker (v4.07) (https://www.repeatmasker.org/). Transposable element (TE) sequences, including SINEs, LINEs, LTR elements, DNA transposons, and other types, were categorized across the chromosomes. The results of repeat sequence annotation indicate that the W chromosome contains the highest proportion of repetitive sequences, with a particularly significant enrichment of LTR (long terminal repeat) elements compared to the other chromosomes (Table 3).

Table 3 Repetitive sequence and GC content distribution across the 32 chromosomes in the M. separata genome.
Full size table

Gene model prediction for the M. separata genome employed a multi-approach strategy, integrating ab initio, transcriptome-based, and homology-based methods. Initially, the genome was processed with RepeatMasker for masking to minimize interference from repetitive elements. Ab initio gene prediction was carried out using AUGUSTUS (v3.2.2)21 with default parameters, where the model was trained using RNA-Seq-derived transcripts to enhance prediction accuracy. Quality-controlled RNA-Seq reads were aligned to the assembled M. separata genome using TopHat2 (v2.0.12)22 and processed with Cufflinks (v2.2.1)23 to generate transcript predictions. For homology-based prediction, the genome sequences of M. separata were aligned with the protein sequences of closely related species (Agrotis ipsilon, Spodoptera litura, Helicoverpa armigera, Bombyx mori, and Ostrinia furnacalis) using BLAST. Gene structures were subsequently refined and predicted using GeneWise v2.4.1 (https://www.ebi.ac.uk/~birney/wise2/). Finally, Evidence Modeler (v1.1.1)24 was utilized to integrate the predictions from the three approaches, resulting in a unified dataset of 19,879 predicted protein-coding genes. Notably, 824 protein-coding genes were predicted on the W chromosome.

For non-coding RNA (ncRNA) annotation, both database searches and model-based predictions were employed. Transfer RNAs (tRNAs) were identified with tRNAscan-SE v2.0.925 using eukaryote-specific parameters. Ribosomal RNAs (rRNAs) and their subunits were predicted using Barrnap v0.9 (https://github.com/tseemann/barrnap) with the parameter–kingdom euk. MicroRNAs (miRNAs), small nuclear RNAs (snRNAs), and small nucleolar RNAs (snoRNAs) were detected by querying the Rfam database (v14.10)26 with Infernal cmscan v1.1.427. This comprehensive analysis identified 4,367 tRNAs, 551 rRNAs, 71 miRNAs, 139 snRNAs, and 23 snoRNAs in the M. separata genome (Table 4).

Table 4 Non-coding RNA annotation statistics in the M. separata genome.
Full size table

Development of W-chromosome specific markers

In Lepidoptera, females possess a ZW sex-determination system, whereas males have a ZZ system28,29,30. This results in a female-specific presence of W-linked sequences, which can be used for sex detection, particularly in cases where both sexes share similar morphological characteristics, such as during the larval stage. To identify W-chromosome-specific sequences, we compared them with other chromosomes and designed two pairs of PCR primers accordingly (Table 5). PCR amplification and agarose gel electrophoresis revealed that specific bands corresponding to the W chromosome were present exclusively in females, as expected in the ZW sex-determination system of Lepidoptera species (Fig. 3). This sex-specific banding pattern, combined with the positive identification of a mitochondrial COI marker31, enabled clear and reliable detection of M. separata’s sex, even at the larval stage.

Table 5 W-chromosome-specific markers developed in this study for sex detection.
Full size table
Fig. 3
figure 3

Sex detection markers based on W chromosome in M. separata. (a) Schematic representation of the sex detection method using PCR amplification and gel electrophoresis, where W-chromosome-specific bands are observed exclusively in females. (b) Electrophoresis results for two developed sex markers, with a mitochondrial COI marker included as a positive control. Lanes 1–10 correspond to male samples, and lanes 11–20 correspond to female samples. Notably, no bands were detected in lane 16, likely due to poor DNA quality.

Full size image

Data Records

The whole genome project is accessible in NCBI’s database under BioProject accession number PRJNA1194333. The PacBio, Illumina, Hi-C sequencing and RNA-Seq raw data are available on the NCBI Sequence Read Archive under accession number SRP55021832. The assembled genome is publicly available in GenBank under accession number GCA_048418785.133. To enhance accessibility, the assembled genome and gene annotation files have been deposited in the Figshare database34.

Technical Validation

To ensure the integrity and accuracy of the assembled chromosomes, several validation approaches were undertaken. First, karyotyping was performed to determine the chromosome number and configurations of M. separata. For karyotype analysis, testis was dissected from a 5th instar larva and incubated in a colchicine solution at 25 °C for 3 hours. The tissues underwent hypotonic treatment in 1% sodium citrate solution at 25 °C for 50 minutes, followed by fixation in a 3:1 methanol–acetic acid solution at 4 °C for 2 hours. Next, the tissues were softened in 60% acetic acid for 10 minutes on pre-chilled glass slides, fully shredded using a dissecting needle, and refixed in the same fixative to create an even cell suspension. After air-drying, the slides were stained with 5% Giemsa solution (pH 6.8) for 30 minutes, rinsed with running water, and air-dried at room temperature. Chromosome morphology was observed using an Olympus BX51 microscope. The analysis revealed that the diploid complement of M. separata consists of 62 chromosomes, providing critical insights into the chromosomal architecture of this species (Fig. 1).

To evaluate assembly integrity, PacBio HiFi sequencing reads were mapped to the assembled genome using the minimap2 tool35. The alignment analysis demonstrated an average sequencing depth of 51.33× and a coverage rate of 99.99%, indicating near-complete genome representation. The completeness of the M. separata genome assembly was further evaluated using BUSCO (v5.3.2)36 with the Arthropoda gene set (odb10). The results showed that 98.6% of BUSCO genes were complete, comprising 97.1% single-copy and 1.5% duplicated genes, with 0.2% fragmented and 1.2% missing BUSCOs. Similarly, a quality assessment of the predicted gene set in protein mode revealed a BUSCO completeness of 97.8%, including 96.3% single-copy, 1.5% duplicated, 0.7% fragmented, and 1.5% missing BUSCOs. These metrics collectively affirm the assembly’s high quality and completeness.

To further validate the chromosomal organization, the raw Illumina reads were mapped to the 32 chromosomes, revealing that the sequencing depth of the W and Z chromosomes is nearly half that of the autosomes (Table 3). This is consistent with the female pattern, where the sex chromosomes are haploid, while the autosomes are diploid. Additionally, the synteny analysis demonstrated strong syntenic relationships between the M. separata Z chromosome and the Z chromosomes of the other two species (Fig. 2). In contrast, the W chromosome of M. separata exhibited high synteny with the W chromosome of N. pronuba37, which is notably absent in S. litura38 (Fig. 2). These results confirm the identity and presence of the W sex chromosome in this current genome assembly.