Genomic DNA sequencing

For Illumina short-read sequencing, the genomic DNA was isolated from of a single female adult following Chen’s protocol¹⁶, briefly, using sodium dodecyl sulfate (SDS) and proteinase K digestion, followed by phenol-chloroform extraction. The library (150bp inserts) was constructed with Nextera DNA Flex Library Prep Kit (Illumina, San Diego, CA, USA), and sequenced on the Illumina NovaSeq 6000 (Illumina, San Diego, CA, USA), generating 43.66Gb of raw data with 150bp pair-end reads. Adapters and low-quality short reads were removed by Fastp (v0.21.0)¹⁷ with default parameters, resulting in a total of 42.05Gb (~123×coverage) of clean data (Table 2). The short-read data was used for genome survey and assembly polish.

Table 2

Library sequencing data and methods used in this study to assemble the Odontothrips loti genome.

Sequencing strategy	Platform	Usage	Insertion size	Clean data (Gb)	Coverage (X)
Short-reads	Illumina	Survey Assembly	150bp	42.05	123
Long-reads	Oxford Nanopore	Assembly	10–20Kb	39.63	116
Hi-C	Illumina	Hi-C assembly	150bp	31.78	93
RNA-seq	Oxford Nanopore	Annotation	1–15Kb	10.24	30

Open in a separate window

For long-read genomic DNA sequencing, we used approximately 800 mixed-sex adult thrips. Genomic DNA was extracted using the SDS method¹⁶, and the DNA fragment size and the degree of degradation were checked on a 0.7% agarose gel. The purity and concentration of extracted DNA were determined with NanoDrop One (Thermo Fisher Scientific). The library was constructed with SQK-LSK109 kit (Oxford Nanopore Technologies, Oxford, UK) according to the manufacturer’s instructions and sequenced on the Oxford Nanopore PromethION platform (Oxford Nanopore Technologies, Oxford, UK). We obtained 41.19Gb (~120×coverage) of raw long-read data with mean length of 6,182.26bp (N50=16,150bp). We then used Oxford Nanopore GUPPY (v0.3.0, https://timkahlke.github.io/LongRead_tutorials/BS_G.html) to filter reads with quality score<7 and obtained 39.63Gb (~116×coverage) of clean reads. The cleaned long-read data were used for contig-level genome assembly (Table 2).

Hi-C library preparation and sequencing

The Hi-C sequencing library was prepared with 800 mixed-sex adult thrips. Samples were cross-linked with a 2% formaldehyde isolation buffer and then treated with DpnII (New England Biolabs, Beijing, CN) to digest nuclei. Biotinylated nucleotides were used to repair tails, and the ligated DNA was split into fragments of 300–700bp in length. The resulting Hi-C library was sequenced in Illumina Novoseq 6000 for 150bp paired-end reads. After applying the same filter criteria for short reads, a total of 31.78Gb (~93×coverage) of clean data was generated to assist the chromosome-level assembly (Table 2).

ONT-Transcriptome sequencing

For ONT-transcriptome sequencing, approximately 240 thrips including nymph and adult were mixed for RNA extraction with the RNA Easy Fast Tissue/Cell Kit (Tiangen). NanoDrop (Thermo Fisher Scientific) and Qubit 3.0 Fluorometer (Life Technologies, Carlsbad, CA, USA) were used to evaluate the quality of extracted RNA. SQK-PCS109 and SQK-PBK004 kit (Oxford Nanopore Technologies) were used for reverse transcript and construction of cDNA library, and sequencing was proceeded on the PromethION sequencer (Oxford Nanopore Technologies, Oxford, UK). A total of 10.24Gb of clean reads were generated with mean length of 1,034.61bp (N50=1,238bp), used to assist genome annotation (Table 2).

Estimation of genomic characteristics

Genomic characteristics were estimated based on 42.05Gb of short-read data using a K-mer-based statistical analysis in Jellyfish (v2.3.0)¹⁸ and GenomeScope2¹⁹ (p=2, k=19). Based on 19-mer depth analysis, the genome size and heterozygosity were estimated to be 341.3Mb and 1.49%, respectively, therefore, this genome is considered highly heterozygous (Fig. 2).

Open in a separate window

Fig. 2

Characteristics of the Illumina short-read sequencing of the Odontothrips loti genome.

Genome assembly

Hi-C scaffolding

Low-quality raw reads (quality score <20,length shorter than 30bp) and adaptors were removed using Fastp (v0.21.0)¹⁷. The clean reads were then mapped to the contig assembly using HICUP (v0.8.0)²² to filter unmapped reads, invalid pairs, dangling end and repeats resulting from PCR amplification. The valid paired-end pairs were used for contig cluster, order and orient by ALLHIC (v0.9.8)²³. The interaction between contig pairs were converted into binary files by 3D-DNA²⁴ and Juicer (v1.6)²⁵. The HiCExplorer (v3.6)²⁶ was used to generate the heat maps of contig interaction intensity and location. The Juicebox (v1.11.08)²⁷ was subsequently employed to review assembly manually. In summary, the resulting chromosome-level genome length was 346.59Mb with a scaffold N50 of 18.52Mb (Table 3), around 86.93% (301.28Mb) of the genome bases were anchored onto 16 chromosomes (Fig. 3a), and most syntenic blocks of genome presents in the low GC content region (Fig. 3b).

Open in a separate window

Fig. 3

Heatmap of genome-wide Hi-C data and circular representation of the chromosomes of Odontothrips loti. (a) The heatmap of chromosome interactions in O. loti. The frequency of Hi-C interaction links is represented by colors, which ranges from yellow (low) to red (high). (b) Circos plot of distribution of the genomic elements in O. loti. The tracks indicate (i) length of the chromosome, (ii) gene density, (iii) distribution of transposable element (TE) density, and (iv) GC density. Center: intra-genomic syntenic blocks of O. loti. The densities of genes, TEs, and GC were calculated in 500kb windows.

Predicting repeats

We used ReaptModeler (v.1.0.11, https://github.com/Dfam-consortium/RepeatModeler) to predict repeat sequence. LTR_FINDER (vOfficial, -size 1000000 -time 300)²⁸ and LTR_retriever (v2.9.0)²⁹ were used to find and de-redundant the LTR sequence. These two de novo library were combined with RepBase³⁰ for further prediction by RepeatMasker (v4.0.9,-nolow -no_is -norna)³¹. RepeatProteinMask (-noLowSimple -pvalue 0.0001) was used for homo-prediction. All results were de-redundant and merged to the final repeat sequence. In summary, 115.26Mb repeat sequences were identified, accounting for 33.26% of the O. loti genome (Table 4). Among these repeat sequences, most (18.85%) are DNA transposon, followed by 10.13% of long terminal repeats (LTRs), 3.45% of long interspersed nuclear elements (LINEs) and only 0.40% of short interspersed nuclear elements (SINEs) (Table 4).

This work was supported by National Natural Science Foundation of China (no. 31971759 to B.L.), the Beijing Innovation Consortium of Modern Agricultural Industry Technology System (no. BAIC02-2024 to B.L.) and the Ningxia Province Sci-Tech Innovation Demonstration Program of High-Quality Agricultural Development and Ecological Conservation (no. NGSB-2021-15-04 to W.S.). We are grateful to Chaoyang Zhao (National Soil Dynamics Laboratory, USDA-ARS, Auburn, AL, USA) for guidance to improve the language of manuscript. The bioinformatics analysis is supported by High-performance Computing Platform of China Agricultural University.