Despite the recognized threats that invasive species pose to human health, economic activities, and biodiversity, our understanding of the factors driving most biological invasions remains limited (Li et al., 2016 X. Li; X. Liu; F. Kraus; R. Tingley; Y. Li Risk of biological invasions is concentrated in biodiversity hotspots, Frontiers in Ecology and the Environment, Volume 14 (2016) no. 8, pp. 411-417 | DOI; Chinchio et al., 2020 E. Chinchio; M. Crotta; C. Romeo; J. A. Drewe; J. Guitian; N. Ferrari Invasive alien species and disease risk: An open challenge in public and animal health, PLOS Pathogens, Volume 16 (2020) no. 10, e1008922 | DOI; Diagne et al., 2021 C. Diagne; B. Leroy; A.-C. Vaissière; R. E. Gozlan; D. Roiz; I. Jarić; J.-M. Salles; C. J. A. Bradshaw; F. Courchamp High and rising economic costs of biological invasions worldwide, Nature, Volume 592 (2021) no. 7855, pp. 571-576 | DOI). This knowledge gap is particularly pronounced for insects, despite them being the most prevalent and destructive group of animal invaders in terrestrial ecosystems (Bradshaw et al., 2016 C. J. A. Bradshaw; B. Leroy; C. Bellard; D. Roiz; C. Albert; A. Fournier; M. Barbet-Massin; J.-M. Salles; F. Simard; F. Courchamp Massive yet grossly underestimated global costs of invasive insects, Nature Communications, Volume 7 (2016) no. 1, 12986 | DOI). Many phytophagous insects have a significant impact, attacking crops, ornamental plants, or forest trees of ecological and commercial importance. Yet very few insect genomes are available to address the many fundamental and applied questions related to biological invasions, including insights into genetic adaptations, invasion pathways, and mechanisms underlying their impacts on native ecosystems.

Among the numerous invasive insect species, Cydalima perspectalis (the box tree moth), Leptoglossus occidentalis (the western conifer seed bug), and Tecia solanivora (the Guatemalan tuber moth) are notable for their rapid spread and severe impacts on their respective host plants. C. perspectalis is a major defoliator of Buxus species, causing widespread damage to natural and ornamental box plants. Native to East Asia, it was first detected in Europe in 2007 and has since expanded its range to North America (Krüger, 2008 E. Krüger Glyphodes perspectalis (Walker, 1859)-new for the European fauna (Lepidoptera: Crambidae, Entomologische Zeitschrift mit Insekten-Börse, Volume 118 (2008), pp. 81-83 | DOI; Bras et al., 2022 A. Bras; E. Lombaert; M. Kenis; H. Li; A. Bernard; J. Rousselet; A. Roques; M.-A. Auger-Rozenberg The fast invasion of Europe by the box tree moth: an additional example coupling multiple introduction events, bridgehead effects and admixture events, Biological Invasions, Volume 24 (2022) no. 12, pp. 3865-3883 | DOI; Coyle et al., 2022 D. R. Coyle; J. Adams; E. Bullas-Appleton; J. Llewellyn; A. Rimmer; M. J. Skvarla; S. M. Smith; J.-H. Chong Identification and Management of Cydalima perspectalis (Lepidoptera: Crambidae) in North America, Journal of Integrated Pest Management, Volume 13 (2022) no. 1 | DOI). L. occidentalis, originally from western North America, is a major seed pest of conifers, affecting the regeneration of valuable tree species. It has progressively invaded eastern North America since the 1950s and Europe since 1999, where it has established successful populations (Schaffner, 1967 J. Schaffner The occurrence of Theognis occidentalis in the midwestern united states (Heteroptera: Coreidae, Journal of the Kansas Entomological Society, Volume 40 (1967), pp. 141-142 | DOI; Taylor et al., 2001 S. Taylor; G. Tescari; M. Villa A Nearctic pest of Pinaceae accidentally introduced into Europe: Leptoglossus occidentalis (Heteroptera: Coreidae) in northern Italy, Entomological News, Volume 112 (2001), pp. 101-103 | DOI; Lesieur et al., 2019 V. Lesieur; E. Lombaert; T. Guillemaud; B. Courtial; W. Strong; A. Roques; M.-A. Auger-Rozenberg The rapid spread of Leptoglossus occidentalis in Europe: a bridgehead invasion, Journal of Pest Science, Volume 92 (2019), pp. 189-200 | DOI). T. solanivora, a specialist pest of potatoes, has had a devastating impact on potato production in the area it has invaded. Native to Central America, it has spread to South America since the 1970s and was introduced to the Canary Islands in 1999 (Povolny, 1973 D. Povolny Scrobipalpopsis solvanivora sp. n.-a new pest of potato (Solanum tuberosum) from Central America, Acta Universitatis Agriculturae, Facultas Agronomica, Volume 21 (1973), pp. 133-146 | DOI; Puillandre et al., 2008 N. Puillandre; S. Dupas; O. Dangles; J.-L. Zeddam; C. Capdevielle-Dulac; K. Barbin; M. Torres-Leguizamon; J.-F. Silvain Genetic bottleneck in invasive species: the potato tuber moth adds to the list, Biological Invasions, Volume 10 (2008) no. 3, pp. 319-333 | DOI).

Despite their ecological and economic importance, genomic resources for these species remain scarce. Although a genome assembly has been previously published for C. perspectalis (Broad et al., 2024 G. R. Broad; D. Boyes; R. Poloni The genome sequence of the Box-tree Moth, Cydalima perspectalis (Walker, 1859), Wellcome Open Research, Volume 9 (2024), 272 | DOI), it was generated from an individual of the typica morph (light-colored). Here, we provide a new draft genome for this species based on an individual of the fusca morph (dark-colored), allowing for comparative genomic studies between these morphs. Furthermore, no reference genome was available for L. occidentalis or T. solanivora, limiting our ability to study the genetic drivers of their invasions. To fill this gap, we generated draft genome assemblies and transcriptomic data for these three species. These datasets, which cover species with diverse genomic characteristics (Table 1), provide valuable resources for investigating invasion dynamics and developing more effective management strategies.

Long-reads, transcriptome RNA-seq and Hi-C sequences were obtained via PacBio and Illumina platforms. Raw reads were deposited onto the NCBI Sequence Read Archive under BioProject PRJNA1140410. All BioSample and SRA accession numbers are listed in Table 2. Large quantities of reads were produced, and the sequence qualities were high for all libraries (Table 2). Mapping the reads from each library onto the genome assemblies built from these same data resulted in good alignment rates for long reads (nearly 100%) and Hi-C sequences (close to 97%). For RNA-seq, mapping rates were more variable, with an average of 72.5% and some individuals exhibiting low values (see Table 2 for details).

For each of the three species, C. perspectalis, L. occidentalis and T. solanivora, draft de novo genomes were assembled. Hi-C sequencing enhanced scaffolding for C. perspectalis (public read set: ERR11217097) and L. occidentalis (read set from this study). N50 values indicate a high level of contiguity for all three assemblies, exceeding 15 Mb in each case. C. perspectalis had the least fragmented assembly, with a total length of 469.1 Mb, only 52 scaffolds and a high sequencing depth of 75X (Table 3). Despite its larger genome size (1.77 Gb) and lower sequencing depth (22.5X), L. occidentalis exhibits strong contiguity, as reflected by its high N50 value of 147.7 Mb (Table 3). Additional quality indicators, including BUSCO scores and Mercury QV metrics, further validated the overall high quality of all three assemblies (Table 3).

Consistent with the relatively high heterozygosity observed in all three genomes (Table 1), K-mer spectra-cn plots reveal two distinct peaks based on HiFi reads, each represented once in the assembly (Figure 1). In the genomes of L. occidentalis and T. solanivora, slight additional k-mer fractions appear more than once in the assembly (Figure 1b and Figure 1c), suggesting incomplete removal of the second haplotype. However, due to the low sequencing depth for both species, the possibility of ancient duplications cannot be ruled out.

Alignments against closely related reference genomes yielded mixed results (Figure 2). For C. perspectalis, the alignments showed high concordance between scaffolds and the reference chromosome of the same species across the entire genome (Figure 2a). By contrast, L. occidentalis and T. solanivora were aligned with more distantly related species, Leptoglossus phyllopus and Teleiodes luculella, respectively, resulting in lower overall concordance as expected (Figure 2b and Figure 2c). Nonetheless, large blocks of collinearity were observed for both species, confirming the strong accuracy and contiguity of our assemblies.

Overall, our assemblies show high contiguity and completeness, despite the presence of heterozygosity and repeat content (Table 1). The use of HiFi long reads, combined with Hi-C scaffolding where available, allowed us to mitigate these challenges and produce high-quality genomic resources.

Annotation across all three genomes predicted a similar number of genes, ranging from 19,326 to 20,895 (Table 4). BUSCO scores were consistently high, with the lowest completeness score reaching 96.0% (Table 4).

All individuals were collected and extracted between March and September 2022 (Table 5). C. perspectalis samples were obtained from a rearing facility (in CEFE, Montpellier, France), L. occidentalis samples were collected from pine trees in southern France, and T. solanivora samples were collected from a potato field in central Colombia. Before further processing, the C. perspectalis sample used for long-read sequencing was stored dry at -80°C, while those used for RNA-seq were stored in RNALater at -20°C. L. occidentalis and T. solanivora samples were stored dry at -80°C and in RNALater at -20°C, respectively.

High-molecular-weight DNA was extracted from one individual of each species. The QIAGEN Genomic Tip 100/G kit was used for C. perspectalis, and the PROMEGA Wizard Genomic DNA purification kit for L. occidentalis and T. solanivora.

Library preparation and sequencing were performed at GeT-PlaGe core facility, INRAe Toulouse according to the manufacturer’s instructions “Procedure & Checklist – Preparing whole genome and metagenome libraries using SMRTbell® prep kit 3.0”. At each step, DNA was quantified using the Qubit dsDNA HS Assay Kit (Life Technologies). DNA purity was tested using the nanodrop (Thermofisher) and size distribution and degradation assessed using the Femto pulse Genomic DNA 165 kb Kit (Agilent). Purification steps were performed using AMPure PB beads (PacBio) and SMRTbell cleanup beads (Pacbio). We made a DNA repair step with the “SMRTbell Damage Repair Kit SPV3” (PacBio). DNA was purified then sheared using the Megaruptor1 system (Diagenode) (Only the Te91 sample was not sheared). After a nuclease step using “SMRTbell® prep kit 3.0”, the libraries were size-selected, using a cutoff on the Pippin HT Size Selection system (Sage Science) with “0.75% Agarose, 6-10 kb High Pass, 75E” protocol.

Using Binding kit 3.2 (primer 3.2, polymerase 2.2) and sequencing kit 2.0, the libraries were sequenced by Adaptive Loading onto three SMRTcells (one per species) on Sequel2 instrument at 90 pM with a 2-hour pre-extension and a 30-hour movie.

For L. occidentalis, Hi-C was performed using the Arima High Coverage Hi-C kit (reference number A101030) according to the manufacturer’s instructions. Briefly, tissues from two frozen adult individuals were crosslinked and homogenized. Chromatin was then digested using four restriction enzymes. Chromatin ends were biotinylated and ligated by proximity ligation. After reverse crosslink and purification, the DNA was used for library preparation using Arima Library Prep Module. Sequencing library was performed using the Arima Library Prep (reference number A303010). The Hi-C library was sequenced on an Illumina NovaSeq6000 platform to generate 2 x150 bp read pairs.

We used the QIAGEN RNeAsy mini plus kit to extract RNA from various dissected tissues of three larvae and three adults of C. perspectalis, full bodies of one juvenile and one adult of L. occidentalis, and full bodies of two larvae of T. solanivora (see Table 5 for details).

RNAseq was performed at the GeT-PlaGe core facility, INRAe Toulouse. RNA-seq libraries were prepared according to Illumina’s protocols using the Illumina TruSeq Stranded mRNA sample prep kit to analyze mRNA. Briefly, mRNA was selected using poly-T beads. RNA was then fragmented to generate double stranded cDNA and adaptors were ligated to be sequenced. 11 cycles of PCR were applied to amplify libraries. Library quality was assessed using a Fragment Analyser and libraries were quantified by QPCR using the Kapa Library Quantification Kit. RNA-seq experiments have been performed on an Illumina NovaSeq 6000 using a paired-end read length of 2x150 pb with the Illumina NovaSeq 6000 sequencing kits.

Read quality was checked with two methods. First by computing the average base pair quality value. Second by aligning the reads to the assembly and checking the rate of primary mappings. RNA-seq reads were preprocessed using fastp version 0.23.4 (Chen et al., 2018 S. Chen; Y. Zhou; Y. Chen; J. Gu fastp: an ultra-fast all-in-one FASTQ preprocessor, Bioinformatics, Volume 34 (2018) no. 17, pp. i884-i890 | DOI) with default parameters before alignment. Alignments were performed with minimap2 version 2.24 (Li, 2018 H. Li Minimap2: pairwise alignment for nucleotide sequences, Bioinformatics, Volume 34 (2018) no. 18, pp. 3094-3100 | DOI) for long reads (with the -x map-hifi parameter) and Hi-C reads (-x sr parameter), and with STAR v2.7.11b (Dobin et al., 2013 A. Dobin; C. A. Davis; F. Schlesinger; J. Drenkow; C. Zaleski; S. Jha; P. Batut; M. Chaisson; T. R. Gingeras STAR: ultrafast universal RNA-seq aligner, Bioinformatics, Volume 29 (2013) no. 1, pp. 15-21 | DOI) for RNA-seq reads. The alignment files were transformed into BAM format with samtools sort version 1.14 (Li et al., 2009 H. Li; B. Handsaker; A. Wysoker; T. Fennell; J. Ruan; N. Homer; G. Marth; G. Abecasis; R. Durbin The Sequence Alignment/Map format and SAMtools, Bioinformatics, Volume 25 (2009) no. 16, pp. 2078-2079 | DOI), and the alignment metrics were extracted using samtools flagstat with default parameters.

All three assemblies were generated using hifiasm (Cheng et al., 2021 H. Cheng; G. T. Concepcion; X. Feng; H. Zhang; H. Li Haplotype-resolved de novo assembly using phased assembly graphs with hifiasm, Nature Methods, Volume 18 (2021) no. 2, pp. 170-175 | DOI, 2022 H. Cheng; E. D. Jarvis; O. Fedrigo; K.-P. Koepfli; L. Urban; N. J. Gemmell; H. Li Haplotype-resolved assembly of diploid genomes without parental data, Nature Biotechnology, Volume 40 (2022) no. 9, pp. 1332-1335 | DOI). Due to the differing data availability timelines, Cydalima perspectalis reads were assembled with version 0.16.1, while Leptoglossus occidentalis and Tecia solanivora were assembled with version 0.18.8, applying default parameters in all cases. Before scaffolding, the assembly of Leptoglossus occidentalis was processed with purge_dups version 1964aaa using default parameters to remove large k-mer duplications. This reduced the total assembly size from 2.098 Gb to 1.769 Gb and the number of contigs from 7,377 to 4,976.

C. perspectalis and L. occidentalis were scaffolded into chromosomes using public and novel Hi-C reads respectively. The public C. perspectalis hi-C read set ERR11217097 was downloaded from the NCBI. Both scaffoldings were performed using juicer version 1.6, 3D-DNA release 529ccf4 (Dudchenko et al., 2017 O. Dudchenko; S. Batra; A. Omer; S. Nyquist; M. Hoeger; N. Durand; M. Shamim; I. Machol; E. Lander; A. Aiden; E. Aiden De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffolds, Science, Volume 356 (2017), pp. 92-95 | DOI), followed by a manual curation with Juicebox version 1.11.08 (Durand et al., 2016 N. Durand; J. Robinson; M. Shamim; I. Machol; J. Mesirov; E. Lander; E. Aiden Juicebox Provides a Visualization System for Hi-C Contact Maps with Unlimited Zoom, Cell Systems, Volume 3 (2016), pp. 99-101 | DOI).

Additionally, all three mitochondrial genomes were assembled using MitoHiFi v2.2 (Uliano-Silva et al., 2023 M. Uliano-Silva; J. G. R. N. Ferreira; K. Krasheninnikova; M. Blaxter; N. Mieszkowska; N. Hall; P. Holland; R. Durbin; T. Richards; P. Kersey; P. Hollingsworth; W. Wilson; A. Twyford; E. Gaya; M. Lawniczak; O. Lewis; G. Broad; F. Martin; M. Hart; I. Barnes; G. Formenti; L. Abueg; J. Torrance; E. W. Myers; R. Durbin; M. Blaxter; S. A. McCarthy MitoHiFi: a python pipeline for mitochondrial genome assembly from PacBio high fidelity reads, BMC Bioinformatics, Volume 24 (2023) no. 1, 288 | DOI).

The assemblies in their final states were validated with four methods. First assembly metrics were calculated using assemblathon_stats.pl script (Bradnam et al., 2013 K. R. Bradnam; J. N. Fass; A. Alexandrov; P. Baranay; M. Bechner; I. Birol; S. Boisvert; J. A. Chapman; G. Chapuis; R. Chikhi; H. Chitsaz; W.-C. Chou; J. Corbeil; C. Del Fabbro; T. R. Docking; R. Durbin; D. Earl; S. Emrich; P. Fedotov; N. A. Fonseca; G. Ganapathy; R. A. Gibbs; S. Gnerre; É. Godzaridis; S. Goldstein; M. Haimel; G. Hall; D. Haussler; J. B. Hiatt; I. Y. Ho; J. Howard; M. Hunt; S. D. Jackman; D. B. Jaffe; E. D. Jarvis; H. Jiang; S. Kazakov; P. J. Kersey; J. O. Kitzman; J. R. Knight; S. Koren; T.-W. Lam; D. Lavenier; F. Laviolette; Y. Li; Z. Li; B. Liu; Y. Liu; R. Luo; I. MacCallum; M. D. MacManes; N. Maillet; S. Melnikov; D. Naquin; Z. Ning; T. D. Otto; B. Paten; O. S. Paulo; A. M. Phillippy; F. Pina-Martins; M. Place; D. Przybylski; X. Qin; C. Qu; F. J. Ribeiro; S. Richards; D. S. Rokhsar; J. G. Ruby; S. Scalabrin; M. C. Schatz; D. C. Schwartz; A. Sergushichev; T. Sharpe; T. I. Shaw; J. Shendure; Y. Shi; J. T. Simpson; H. Song; F. Tsarev; F. Vezzi; R. Vicedomini; B. M. Vieira; J. Wang; K. C. Worley; S. Yin; S.-M. Yiu; J. Yuan; G. Zhang; H. Zhang; S. Zhou; I. F. Korf Assemblathon 2: evaluating de novo methods of genome assembly in three vertebrate species, Gigascience, Volume 2 (2013) no. 1 | DOI). Second BUSCO scores were calculated using the insecta_odb10 database (75 genomes and 1367 BUSCOs) with version 5.7.1 of the BUSCO software package (Manni et al., 2021 M. Manni; M. R. Berkeley; M. Seppey; F. A. Simão; E. M. Zdobnov BUSCO Update: Novel and Streamlined Workflows along with Broader and Deeper Phylogenetic Coverage for Scoring of Eukaryotic, Prokaryotic, and Viral Genomes, Molecular Biology and Evolution, Volume 38 (2021) no. 10, pp. 4647-4654 | DOI), with the -m geno option. Third merqury QV and completeness statistics were generated (Rhie et al., 2020 A. Rhie; B. P. Walenz; S. Koren; A. M. Phillippy Merqury: reference-free quality, completeness, and phasing assessment for genome assemblies, Genome Biology, Volume 21 (2020) no. 1, 245 | DOI). Fourth, d-genies dot-plots (Cabanettes & Klopp, 2018 F. Cabanettes; C. Klopp D-GENIES: dot plot large genomes in an interactive, efficient and simple way, PeerJ, Volume 6 (2018), e4958 | DOI) were generated versus a phylogenetically close related species (itself in the case of Cydalima perspectalis for which a reference genome is available).

Genome metrics corresponding to estimated genome size, repeat fraction and heterozygosity found in Table 1 were all extracted from the HiFi reads using genomescope2 (Ranallo-Benavidez et al., 2020 T. R. Ranallo-Benavidez; K. S. Jaron; M. C. Schatz GenomeScope 2.0 and Smudgeplot for reference-free profiling of polyploid genomes, Nature Communications, Volume 11 (2020) no. 1, 1432 | DOI). The Cydalima perspectalis estimated genome size is very close to the size of the good quality public assembly found at the NCBI GCA_951394215.1 (483.7 Mb).

Using the available RNA-Seq reads sets, we performed a de novo annotation for each assembly, primarily to validate nucleotide content of the assemblies. Annotations were conducted with helixer version 0.3.1_cuda_11.2.0 (Stiehler et al., 2020 F. Stiehler; M. Steinborn; S. Scholz; D. Dey; A. P. M. Weber; A. K. Denton Helixer: cross-species gene annotation of large eukaryotic genomes using deep learning, Bioinformatics, Volume 36 (2020) no. 22-23, pp. 5291-5298 | DOI) using the invertebrate trained model and default parameters, as well as with braker (Gabriel et al., 2021 L. Gabriel; K. J. Hoff; T. Brůna; M. Borodovsky; M. Stanke TSEBRA: transcript selector for BRAKER, BMC Bioinformatics, Volume 22 (2021) no. 1, 566 | DOI). Both annotations were then merged using the agat_sp_merge_annotations.pl script from the agat suite version 1.2.0 with default parameters (Dainat et al., 2021 J. Dainat; D. Hereñú; P. Pucholt NBISweden/AGAT: AGAT-v0.7.0 software, Zenodo (2021) | DOI). Annotation validation was performed using the same version of BUSCO software as for the assemblies but with the -m tran option. Due to poor results from braker for L. occidentalis, only helixer output was kept as final annotation for this species.

Additionally, all three mitochondrial genomes were annotated using GeSeq v2.03 (Tillich et al., 2017 M. Tillich; P. Lehwark; T. Pellizzer; E. S. Ulbricht-Jones; A. Fischer; R. Bock; S. Greiner GeSeq – versatile and accurate annotation of organelle genomes, Nucleic Acids Research, Volume 45 (2017) no. W1, pp. W6-W11 | DOI).

We thank our colleagues Stéphane Dupas and Barbara Porro for Tecia solanivora and Leptoglossus occidentalis samples. We thank the Technical Platform for Experimental Ecology at CEFE Montpellier for their assistance in maintaining captive C. perspectalis stocks. We also thank Emmanuelle Murciano-Germain for administrative assistance. This work was performed in collaboration with the GeT core facility, Toulouse, France (GeT, https://doi.org/10.15454/1.5572370921303193E12). GeT core facility was supported by France Génomique National infrastructure, funded as part of “Investissement d’avenir” program managed by Agence Nationale pour la Recherche (contract ANR-10-INBS-09). Preprint version 4 of this article has been peer-reviewed and recommended by Peer Community In Genomics (https://doi.org/10.24072/pci.genomics.100425; Lacroix, 2025 V. Lacroix Three more reference genomes of invasive insect species, Peer Community in Genomics, Volume 100425 (2025) | DOI).

This work was funded by grants from ANR project GENLOADICS.

The authors of this preprint declare that they have no financial conflict of interest with the content of this article.

The sequencing data and genome assemblies from this study have been deposited at DDBJ/ENA/GenBank, project PRJNA1140410: https://www.ncbi.nlm.nih.gov/bioproject/1140410. Genome and mitogenome annotations, as well as mitogenome sequences, are available at Data INRAE via the following link: https://doi.org/10.57745/WDMFPB (Lombaert & Klopp, 2025 C. Klopp FARDOMICS assembly procedure, Zenodo, Volume v1 (2025) | DOI). All the required command-line instructions can be accessed on github (https://github.com/chklopp/FARDOMICS_assemblies/blob/main/README.md) or via Zenodo: https://doi.org/10.5281/zenodo.15573596 (Klopp, 2025 C. Klopp FARDOMICS assembly procedure, Zenodo, Volume v1 (2025) | DOI).

EL, CK and ED designed the study. EL, RP and MJ managed the choice and collection of samples. AB and RP prepared the samples. GA, CI, JL, MS and SV prepared the libraries and performed the sequencing. CK and EL analysed the data. EL, CK, CI, MS and SV wrote the paper. All authors have revised and approved the final manuscript.