A chromosomal-level genome assembly of Batocera lineolata Chevrolat, 1852 (Coleoptera: Cerambycidae)

Abstract

The white-striped longhorned beetle, Batocera lineolata, is an important forestry pest, with larvae that damage vascular and woody tissues, frequently leading to tree mortality. The lack of a high-quality genome has constrained research into the mechanisms underlying its damage. Here, we report a chromosome-scale genome of B. lineolata, constructed through an integrated approach combining PacBio HiFi, Hi-C, and Illumina RNA sequencing data. The assembled genome of B. lineolata spans 327.96 Mb across 18 scaffolds, with a scaffold N50 of 34.45 Mb. Notably, 94.5% of the assembly (309.92 Mb) was anchored to 10 chromosomes. Synteny analysis identified chromosome 10 as the X chromosome and confirmed its conservation across Coleoptera. BUSCO analysis indicated a high genome completeness of 98.8%, comprising 98.7% single-copy and 0.1% duplicated genes. Repetitive elements accounted for 28.04% (91.97 Mb) of the genome, and 12,030 genes were annotated. This genome provides a foundation for studying the ecology and genetics of B. lineolata and for developing more effective wood-boring pest control strategies.

Background & Summary

The superfamily Chrysomeloidea, one of the largest groups within Coleoptera, primarily includes the families Chrysomelidae (leaf beetles), Cerambycidae (longhorned beetles), and Megalopodidae (leaf beetles)¹. Among these, Cerambycidae represents one of the largest and most diverse beetle families and is of considerable ecological and economic importance, with over 36,000 species described worldwide^2,3. Members of this family are nearly cosmopolitan, inhabiting environments from lowlands to mountainous regions above 4,000 m, with the exception of Antarctica^4,5. Several cerambycid species are serious pests in agriculture, horticulture, and forestry, where larval feeding within woody and vascular tissues can severely impair plant growth or cause host mortality⁶. In addition to direct damage, some species act as vectors of harmful plant pathogens, including nematodes and fungi, thereby exacerbating disease transmission and increasing economic losses⁷. Among these species, the white-striped longhorned beetle, Batocera lineolata, is primarily distributed across China, Vietnam, Myanmar, India, Japan, and Korea⁸. B. lineolata is a significant forestry pest whose adults and larvae bore into the bark of Juglans regia, Populus spp., Fraxinus spp., and Ligustrum lucidum, disrupting vascular tissues and accelerating plant wilting and mortality⁹. This species completes one generation every two years and overwinters as larvae or adults within tree trunks¹⁰. High-quality genomes offer invaluable resources for elucidating the biology, ecology, and evolutionary history of Cerambycidae. Genomic analyses of Anoplophora glabripennis have identified glycoside hydrolase enzymes acquired via horizontal gene transfer, which are capable of breaking down plant cell walls and likely enhance the beetle’s ability to extract nutrients from otherwise nutrient-poor and highly recalcitrant woody tissues^11,12. In recent years, chromosome-level genome assemblies have been generated for four additional xylophagous longhorned beetles of considerable ecological and economic importance^13–16, including Monochamus alternatus, Monochamus saltuarius, Arhopalus rusticus, and M. scutellatus. These genomes provide valuable resources for investigating the genomic basis of conifer feeding and the adaptive traits associated with forest environments. The absence of a complete genome for B. lineolata has hindered studies of its biology and the design of effective control measures.

In this study, we present a chromosome-level genome of B. lineolata, generated by integrating PacBio HiFi, Illumina, and Hi-C sequencing data. Synteny analysis was employed to accurately identify the X chromosome. We further performed comprehensive annotation of repetitive elements, non-coding RNAs, and protein-coding genes. This high-quality genome provides a valuable resource for investigating the ecology, genetics, and evolution of B. lineolata and lays the foundation for developing more effective management strategies against wood-boring pests.

Methods

Sample collection and sequencing

A single adult male specimen of B. lineolata was collected in Xinyang City, Henan Province, China, on 26 June 2025 for dual DNA and RNA sequencing. To reduce contamination from residual gut contents, the specimen was fasted for several days. Muscle tissues from the pronotum and posterior abdominal segments were then dissected, flash-frozen in liquid nitrogen, and stored at −80 °C until sequencing.

Genomic DNA was isolated via the CTAB method and subsequently purified with a Grandomics Genomic DNA kit for downstream sequencing. Total RNA was isolated from the same individual using TRIzol Reagent (Thermo Fisher Scientific). Genomic libraries were constructed with the Illumina TruSeq DNA PCR-Free Kit and sequenced to generate 150-bp paired-end reads. Hi-C libraries were prepared by cross-linking chromatin, followed by MboI digestion, end repair, proximity ligation, and subsequent DNA purification, in accordance with standard protocols¹⁷. All short-read sequencing was performed using the Illumina NovaSeq 6000 platform. A HiFi library with an insert size of 20 kb was constructed using the PacBio SMRTbell Express Template Prep Kit 2.0 (Pacific Biosciences, USA) and sequenced on the PacBio Sequel II platform. All libraries were sequenced by Berry Genomics (Beijing, China). A total of 101.04 Gb of sequencing data were obtained, including 24.86 Gb of PacBio HiFi reads (75.82 × coverage), 41.31 Gb of Illumina reads (125.96×), 24.37 Gb of Hi-C data (74.30×) and 10.50 Gb of RNA data (Table 1).

Table 1.

Statistics of the sequencing data used for genome assembly.

Libraries	Insert sizes (bp)	Clean data (Gb)	Sequencing coverage (x)
Illumina	350	41.31	125.96
PacBio HiFi	20000	24.86	75.82
Hi-C	350	24.37	74.30
RNA	350	10.50	—

Genome assembly

Raw Illumina reads generated for the genome survey were processed with Fastp v0.23.2¹⁸ to trim adapters, remove duplicates, and filter low-quality reads. Genome survey analysis of the filtered reads was performed using GenomeScope v2.0¹⁹. The B. lineolata genome was estimated to be 339.53 Mb in size, with repetitive elements accounting for approximately 20.1% of the genome and a heterozygosity rate of 0.69% (Fig. 1).

Fig. 1.

The genome size, heterozygosity, and repeat content of Batocera lineolata were estimated using GenomeScope, predicting a genome size of 339.53 Mb with 0.69% heterozygosity and a duplication rate of 20.1%.

The primary assembly of B. lineolata was generated using Hifiasm v0.19.8²⁰, and redundant heterozygous sequences were removed with Purge_Dups v1.2.5²¹ using a haploid cutoff of 80 to eliminate haplotigs. Hi-C reads were mapped to the draft assembly with Juicer v1.6.2²² after quality filtering. Chromosome-scale scaffolding was subsequently performed by integrating the primary contigs with Hi-C data using 3D-DNA v180922²³. The genome assembly was visually inspected and manually curated using Juicebox v1.11.08²², after which scaffolds were finalized with 3D-DNA. Potential contaminant sequences were screened using MMseqs 2 v11.1²⁴ against the NCBI nucleotide and UniVec databases. In addition, vector contamination was detected by BLASTN (BLAST + v2.11.0)²⁵ searches against the UniVec database. Sequences showing > 90% identity to entries in the database were flagged as potential contaminants. No significant contaminants were detected in the assembly, demonstrating the high standard of sample preparation and reliable sequencing.

To distinguish autosomes from the X chromosome, two complementary validation approaches were employed. First, the final genome assembly was remapped to the PacBio HiFi and Illumina reads using Minimap2 v2.17²⁶. Sequencing depth for each chromosome was calculated with SAMtools v1.22²⁷ as the total number of mapped bases divided by chromosome length. The X chromosome was identified based on its approximately half coverage relative to the autosomes (Table 3). Notably, chromosome 10 exhibited PacBio HiFi and Illumina coverages of 21.05 × and 45.59×, respectively, roughly half that of the autosomes, consistent with the expected hemizygous state of sex chromosomes in males. Second, synteny analysis with Tribolium castaneum was performed, leveraging the conserved features of insect X chromosomes. Syntenic blocks were identified and analyzed using MCScanX²⁸ and visualized with TBtools²⁹. Chromosome 10 was further supported as the X chromosome based on its high synteny with the X chromosome of T. castaneum (Fig. 3). Therefore, chromosome 10 was assigned as the X chromosome (Table 3). Scaffolds longer than 1 Mb were retained in the B. lineolata genome for downstream analysis. The final chromosome-level assembly of B. lineolata spanned 327.96 Mb, with a GC content of 32.85%. It consisted of 18 scaffolds and 58 contigs, with scaffold and contig N50 values of 34.45 Mb and 16.24 Mb, respectively. Notably, 94.50% of the assembly (309.92 Mb) was anchored to 10 chromosomes, which ranged in length from 11.80 Mb to 49.65 Mb (Fig. 2, Fig. 4; Table 2). The final assembly size is highly consistent with the 339.53 Mb estimate obtained from GenomeScope analysis.

Table 3.

Statistics for chromosomes sequence length.

Chromosome ID	Sequences Length (Mb)	Coverage Short Reads	Coverage Long Reads
Chr01	49.65	99.92	41.37
Chr02	40.24	147.16	49.25
Chr03	36.27	107.47	48.66
Chr04	35.27	93.68	38.22
Chr05	34.45	94.06	39.59
Chr06	30.83	96.13	36.52
Chr07	24.41	121.05	51.95
Chr08	23.95	131.56	56.30
Chr09	23.04	144.15	60.29
ChrX	11.80	45.59	21.05

Fig. 3.

Synteny analysis between the chromosomes of Batocera lineolata (Blin) and Tribolium castaneum (Tcas), with the X chromosome indicated in red.

Fig. 2.

Chromosome-scale Hi-C contact heatmap of the Batocera lineolata genome assembly. Blue lines denote superscaffolds, and both the X and Y axes correspond to cumulative scaffold length in megabases (Mb).

Fig. 4.

A circular plot summarizes the genomic landscape of Batocera lineolata. The concentric rings, from outermost to innermost, represent chromosome length (physical scale), GC content, gene density, and the distribution of major repetitive elements, including transposable elements (DNA transposons, SINEs, LINEs, LTRs) and simple repeats.

Table 2.

Genome assembly statistics for Batocera lineolata.

Assembly	Total length (Mb)	Number scaffolds/contigs (chromosomes)	Scaffold/contig N50 length (Mb)	GC (%)		BUSCO (n = 1,367) (%)
					C		S	D	F	M
Hifiasm	512.74	544/544	13.95/13.95	32.10	99.1		98.4	0.7	0.2	0.7
Purge_Dups	362.29	89/89	16.61/16.61	32.85	98.7		98.6	0.1	0.1	1.2
3D-DNA	362.29	257/308 (10)	34.45/16.13	32.85	98.7		98.6	0.1	0.1	1.2
Final	327.96	18/58 (10)	34.45/16.24	32.85	98.8		98.7	0.1	0.1	1.1

C: complete BUSCOs; S: Complete and single-copy BUSCOs; D: complete and duplicated BUSCOs; F: fragmented BUSCOs; M: missing BUSCOs.

Genome annotation

Repetitive elements in the B. lineolata genome were identified using an integrated strategy combining homology-based and de novo predictions. De novo repeat annotation was performed using RepeatModeler v2.0.4³⁰, ncorporating the “-LTRStruct” option to improve the detection of LTR retrotransposons. The repeat library was integrated with RepBase-20230909³¹ and Dfam v3.5³² to build a custom repeat database. RepeatMasker v4.1.2³³ was used to identify repetitive sequences using the combined repeat library. Repetitive elements accounted for approximately 91.97 Mb, corresponding to 28.04% of the genome. The repetitive elements comprised 59.02 Mb (18.00%) of unclassified repeats, 20.47 Mb (6.19%) of DNA transposons, 2.17 Mb (0.66%) of LINEs, 4.95 Mb (1.49%) of LTRs, 0.043 Mb (0.01%) of SINEs, 4.44 Mb (1.35%) of simple repeats, and 1.12 Mb (0.34%) of other repeat categories (Table 4).

Table 4.

Genome assembly and annotation statistics of Batocera lineolata.

Characteristics	Batocera lineolata
Genome assembly
Genome Size (Mb)	327.96
Number of scaffolds	18
Number of chromosomes	10
Scaffold N50 length (Mb)	34.45
Contig N50 length (Mb)	16.24
GC (%)	32.85
BUSCO completeness (%)	98.8%
Protein-coding genes
Number	12030
Mean gene length (bp)	6890.0
BUSCO completeness (%)	97.1
Repetitive elements
Size (Mb)	91.97 (28.04%)
DNA transposons (Mb)	20.47 (6.19%)
LINEs (Mb)	2.17 (0.66%)
LTRs (Mb)	4.95 (1.49%)
SINEs (Mb)	0.043 (0.01%)
Unclassified (Mb)	59.02 (18.00%)
Simple repeats (Mb)	4.44 (1.35%)
Other repeats (Mb)	1.12 (0.34%)
ncRNA
Number of ncRNA	695
tRNA	492
rRNA	33
miRNA	65
snRNA	58
ribozyme	2
Other ncRNA	45
Function annotation
Number of genes matching Uniprot records	11711
Number of genes with InterProScan annotations	10522
Number of genes with eggNOG annotations	11293
Number of genes with COG Functional Categories combining InterProScan and eggNOG results	10627
Number of genes with GO items combining InterProScan and eggNOG results	9609
Number of genes with KEGG pathways items combining InterProScan and eggNOG results	4698

Non-coding RNAs (ncRNAs) in the B. lineolata genome were identified using Infernal v1.1.2³⁴ with the Rfam v14.10³⁵ database, and transfer RNAs (tRNAs) were predicted with tRNAscan-SE v2.0.9³⁶. A total of 695 ncRNAs were identified, including 2 ribozymes, 58 small nuclear RNAs (snRNAs), 65 microRNAs (miRNAs), 492 tRNAs, 33 ribosomal RNAs (rRNAs), and 45 other ncRNAs (Table 4).

The annotation of B. lineolata protein-coding genes was performed with MAKER v3.01.03³⁷. This pipeline produces high-confidence gene models by integrating evidence from multiple sources, including protein homology, ab initio gene predictions, and aligned transcriptomic data. Ab initio gene prediction identifies putative genes based on intrinsic genomic features without external evidence. For this study, we improved prediction accuracy by training the models with both transcriptomic data and protein homology information. RNA-seq data were aligned to the genome using HISAT2 v2.2.1³⁸ to provide transcriptomic evidence, and the resulting alignments were then assembled de novo into transcripts with StringTie v2.1.6³⁹. Ab initio gene prediction was performed with BRAKER v2.1.6⁴⁰, integrating GeneMark-ES/ET/EP v4.68_lic⁴¹ and AUGUSTUS v3.4.0⁴². Model training was carried out automatically using both aligned transcriptome data and protein datasets from the Arthropoda OrthoDB v11 database⁴³. Protein homology-based gene prediction was performed with GeMoMa v1.9⁴⁴, using protein sequences from several model insect genomes: Drosophila melanogaster⁴⁵ (GCF_000001215.4), Apis mellifera⁴⁶ (GCF_003254395.2), Bombyx mori⁴⁷ (GCF_030269925.2), A. glabripennis (GCF_000390285.2)¹¹, and Tribolium castaneum⁴⁸ (GCF_031307605.1) (Table 5). Gene structure analysis revealed that the 12,030 predicted genes in B. lineolata had an average length of 6,890 bp. The average gene structure was composed of 6.2 exons, 5.1 introns, and 6.1 coding sequences (CDSs). The mean lengths of exons, introns, and CDSs were 312.6 bp, 1,030.1 bp, and 269.0 bp, respectively (Table 4). BUSCO analysis showed that 97.1% of Insecta orthologs were recovered in the gene annotation, including 1,321 (96.6%) single-copy, 7 (0.5%) duplicated, 4 (0.3%) fragmented, and 35 (2.6%) missing genes. These findings indicate that the gene annotation of B. lineolata is accurate, comprehensive, and of high quality.

Table 5.

Species taxonomic information and accession code of all samples used in this study.

Species	Order	Family	Source
Tribolium castaneum	Coleoptera	Tenebrionidae	NCBI (GCF_031307605.1)
Anoplophora glabripennis	Coleoptera	Cerambycidae	NCBI (GCF_000390285.2)
Apis mellifera	Hymenoptera	Apidae	NCBI (GCF_003254395.2)
Bombyx mori	Lepidoptera	Bombycidae	NCBI (GCF_030269925.2)
Drosophila melanogaster	Diptera	Drosophilidae	NCBI (GCF_000001215.4)

The predicted proteins were functionally annotated using a combination of complementary approaches. Homology searches against the UniProtKB database were carried out using DIAMOND v2.0.11.15⁴⁹. GO terms and KEGG pathway annotations were annotated via eggNOG-mapper v2.0.15⁵⁰. InterProScan v5.53-87.0⁵¹ was employed for protein domain annotation, incorporating five curated databases, including Pfam⁵², SMART⁵³, SUPERFAMILY⁵⁴, Gene3D⁵⁵, and CDD⁵⁶. The outputs from all annotation tools were subsequently combined to produce a comprehensive functional annotation dataset for the predicted genes of B. lineolata. Functional annotation identified 12,688 UniProt entries in B. lineolata, with 10,105 genes annotated with GO terms, 4,990 genes assigned to KEGG pathways, 2,929 genes linked to EC numbers, and 11,422 genes categorized into COG categories (Table 5). The circular visualization of the genome, generated from genome annotation data and visualized using TBtools, illustrates genomic features such as repeat elements, gene density, and GC content across the pseudochromosomes.

Data Records

The raw sequencing data and the assembled genome of B. lineolata have been deposited and are publicly accessible through the National Center for Biotechnology Information (NCBI). The sequencing datasets, comprising Hi-C (SRR35275989)⁵⁷, transcriptome (SRR35275990)⁵⁸, Illumina short reads (SRR35275991)⁵⁹, and PacBio HiFi long reads (SRR35275992)⁶⁰, have been deposited and are publicly accessible under their respective accession numbers. The assembled genome is publicly available at NCBI (accession GCA_052815245.1)⁶¹, and the corresponding annotation files, including repeat elements, predicted gene models, and functional annotations, can be accessed via Figshare⁶².

Technical Validation

Genome quality for B. lineolata was evaluated through a combination of two complementary methods. To assess completeness, the assembly was analyzed with BUSCO v5.0.4⁶³ using the Insecta gene set (n = 1,367), which indicated a high completeness of 98.8%, including 98.7% single-copy, 0.1% duplicated, 0.1% fragmented, and 1.1% missing BUSCOs. To evaluate the quality of the final assembly, Merqury v1.3⁶⁴ was applied to detect possible sequence errors using k-mer-based analysis and quality value (QV) scoring. The genome of B. lineolata showed a k-mer completeness of 95.6% and a QV score of 46.8286, reflecting excellent base-level accuracy. Combined with the BUSCO assessment, these results confirm that the genome assembly is highly complete and accurate. Assembly accuracy was further evaluated by mapping PacBio, Illumina, and RNA-seq reads to the genome using Minimap2 and SAMtools, resulting in mapping rates of 99.75%, 98.05%, and 96.86%, respectively. Overall, these findings indicate that the B. lineolata genome is highly accurate.

Ethics statement

This research involved an invertebrate insect species and did not include vertebrate animals. According to the relevant institutional and national regulations, ethical approval is not required for studies involving insects. All procedures were conducted in accordance with accepted ethical standards for invertebrate research.

Author contributions

Y.C. and Z.M. contributed to the research design. J.M. and Jin.J. collected the samples. Y.C. and Z.Y. analyzed the data. Y.C. and Z.Y. wrote the draft manuscript and revised the manuscript. All co-authors contributed to this manuscript and approved it.

Data availability

This dataset is original and previously unpublished. The authors declare no competing interests. Raw sequencing reads and the assembled genome for Batocera lineolata are available in the NCBI BioProject database under accession PRJNA1309302. Annotation files, including repeat elements, gene models, and functional predictions, are publicly available on Figshare (https://doi.org/10.6084/m9.figshare.31076632).

Code availability

No specific script was used in this work. All commands and pipelines used in data processing were executed according to the manual and protocols of the corresponding bioinformatic software. Software versions are reported in the Methods section, and default parameters were used unless explicitly specified.

Competing interests

The authors declare no competing interests.