Background & Summary

Euglena, a genus of single-celled flagellate eukaryotes, is ubiquitously distributed in both freshwater and saltwater environments. Possessing photosynthetic chloroplasts, Euglena exhibits autotrophic characteristics akin to plants, while also displaying heterotrophic attributes similar to animals1,2,3. E. gracilis, a prominent species within the genus, serves as a widely utilized model organism in both academic and industrial research due to its rich array of valuable compounds, including pigments, unsaturated fatty acids, vitamins, amino acids, and the distinctive β-1,3-glucan, paramylon—an advantageous functional food ingredient4,5,6. Notably, recent studies, such as Wu et al.’s pilot-scale fermentation achieving maximal biomass and paramylon content7, underscore the industrial potential of E. gracilis.

Despite substantial advancements in genetic modification8,9,10,11,12,13, hindered by the absence of a high-quality genome, E. gracilis remains a subject of limited genetic engineering tools and applications. In 2019, Ebenezer et al. presented an initial genome assembly of E. gracilis (1.43 Gb), which, though informative, proved significantly fragmented14. Consequently, researchers have resorted to omics approaches, including de novo transcriptome assembly14,15 and proteomic analysis1,14, to explore physiological and genomic aspects. Nevertheless, a definitive high-quality genome assembly remains a critical prerequisite for advancing genetic engineering and synthetic biology applications in E. gracilis6.

This study addresses the existing gap by introducing a chromosome-level genome assembly of E. gracilis through the integration of Illumina, PacBio, Bionano, and Hi-C technologies (Table 1). The resulting assembly, spanning 2.37 Gb, with contig N50 of 619 Kb and scaffold N50 of 1.12 Mb, exhibits superior continuity (Table 2). Anchoring to 46 chromosomes (Fig. 1a) achieved a remarkable 99.83% rate, unveiling structural insights. Repetitive elements, constituting 58.84% of the genome, contribute to its complexity. The annotation of 39,362 protein-coding gene models and the assessment of 80.39% gene completeness attest to the high quality of this genome. This achievement marks a pivotal step in enhancing our comprehension of E. gracilis, offering a genetic foundation for both experimental and computational inquiries in this species.

Table 1 Statistical analysis of sequencing reads from Illumina, Pacbio, Bionano and Hi-C.
Full size table
Table 2 Assembly statistics and comparison to previous published data.
Full size table
Fig. 1
figure 1

Chromosome-level assembly of the E. gracilis genome. (a) Genome landscape of the E. gracilis. From the outer ring to the inner ring are the distributions of chromosome length, gene density, transposable element (TE) density, tandem repeat (TR) density, and GC content, with densities calculated within a 1 Mb window. (b) Distribution estimation of 19-kmer. (c) Estimation based on flow cytometry. (d) Hi-C interaction heatmap illustrating the genomic interactions within the E. gracilis genome. The colour bar indicates contact density, ranging from red (high) to white (low).

Full size image

Methods

Sample collection and sequencing

Sample preparation

The E. gracilis Z strain (CCAP 1224/5Z) was purchased from CCAP (Culture Collection of Algae and Protozoa, United Kingdom) and cultivated in our laboratory under autotrophic conditions using CM medium at 26 °C, with a continuous white light intensity of 80 μmol photons·m−2·s−1. Cellular samples were harvested during the mid-log phase, rapidly frozen with liquid nitrogen, and subsequently preserved at −80 °C for subsequent sequencing library preparation.

Library preparation and sequencing

Genomic DNA of high quality was extracted using the CTAB method. Paired-end libraries were constructed using NEBNext Ultra II DNA Library Prep Kit for Illumina (NEB, USA) and sequenced on an Illumina HiSeq2500 platform (Illumina, USA), which generated a total of 264.2 Gb Illumina data, providing approximately 111-fold coverage of the genome (Table 1). In total of 50 mg DNA were used to construct the PacBio Sequel sequencing libraries, then sequencing was performed to produce raw reads. For Bionano sequencing, high molecular weight DNA with a fragment distribution greater than 150 kb were isolated and used for DNA nicking using Nb.BssSI (NEB). The nicks were labelled and then loaded onto the Saphyr Chip nanochannel array (Bionano Genomics) and imaged using the Saphyr system and associated software (Bionano Genomics) according to the Saphyr System User Guide. The PacBio Sequel and Bionano platforms contributed 377.5 Gb and 306.6 Gb data, achieving coverages of approximately 159X and 129X, respectively (Table 1). Hi-C libraries was prepared with the standard procedure described. After digesting the genomic DNA with a restriction enzyme MboI, the sticky ends of the digested fragments were biotinylated, diluted, and then ligated to each other randomly. The prepared sequencing library was sequenced on a NovaSeq platform (Illumina, USA), which yielded a total of 402.3 Gb data with the Illumina sequencing platform (Table 1). Library preparation and sequencing of Illumina survey libraries, PacBio Sequel libraries, Bionano libraries, and all transcriptome libraries were executed by Nowbio Biotechnology Company (Yunnan, China). Frasergen Bioinformatics Co., Ltd (Wuhan, China) undertook the preparation and sequencing of Hi-C libraries on their sequencing platform.

Genome survey and assembly

K-mer frequency analysis

K-mer frequencies (K = 19) were computed from filtered Illumina reads using Jellyfish16 (v2.2.10), serving as the basis for a genome survey conducted with GenomeScope17 (v2.0). The estimated genome size for E. gracilis was determined to be 2.25 Gb (Fig. 1b), aligning closely with the genome size estimations derived from flow cytometry analysis (2.14–2.34 Gb) (Fig. 1c).

Genome assembly

To assemble the genome, NextDenovo18 (v2.2-beta.0) was employed to generate contigs utilizing PacBio reads, followed by three rounds of Illumina read correction using NextPolish19 (v1.0.1). The corrected contigs underwent assembly with Bionano data using Sovle (v3.3). Subsequently, the assembled scaffolds were organized into chromosomes utilizing the 3D-DNA pipeline20 (v201008), followed by manual curation with JuiceBox21 (v2.20.00). The final assembly comprised 46 chromosomes (Fig. 1d), collectively spanning 2.37 Gb, accounting for approximately 99.83% of the entire genome assembly (Table 2), while the individual chromosome lengths ranged from 121.4 Mb (Chr4) to 22.7 Mb (Chr35) (Table 3). Comparing with the previous genome assembly14 of E. gracilis presented by Ebenezer et al., our assembly has much longer N50 and higher BUSCO completeness score (Table 2), which fully suggested that our result is a high-quality assembly, with superior continuity.

Table 3 Length of the assembled chromosome of the E. gracilis genome.
Full size table

Genome repeat and ncRNA analysis

Repeat sequence prediction

A hybrid approach, incorporating both ab initio and homology-based methodologies, was employed to predict repeat sequences within the genome. For ab initio prediction, LTR_FINDER22 (v1.07) and ltrharvest23 (v1.5.10) were used to predict LTR retrotransposons, and the results were integrated using LTR_retriever24 (v2.8). Meanwhile, RepeatModeler25 (v2.0) was also used to identify repeats. Then the results of LTR_retriever and RepeatModeler were merged as a custom library and fed to Repeatmasker26 (v.4.0.9) to predict TEs. Simultaneously, homology-based annotation employed RepeatMasker26 (v.4.0.9) and RepeatProteinMask26 (v.4.0.9) against Repbase27 (Release 20181026). TRF28 (v4.0.9) was used for searching tandem repeats. Following redundancy elimination, a total of 1.4 Gb of repeat sequences were identified, constituting 58.84% of the E. gracilis genome. The repeat sequences predicted by TRF, Repeatmasker, Proteinmask and ab initio pipeline covered 9.85%, 1.89%, 2.07% and 52.75% of the genome sequence, respectively. Within the repeat elements, 32.73% remained unclassified, while long terminal repeats (LTRs) represented 32.81% of the genome. DNA elements, long interspersed nuclear elements (LINEs), and short interspersed nuclear elements (SINEs) accounted for 4.60%, 1.49%, and 0.11% of the genome, respectively (Table 4).

Table 4 Classification of the TE sequences in the E. gracilis genome.
Full size table

Noncoding RNA annotation

To annotate noncoding RNA (ncRNA), tRNAScan-SE29 (v1.3.1) and blast30 (v2.2.26) were applied for tRNA and rRNA prediction, respectively. Additionally, Rfam31 (v9.1) and INFERNAL32 (v0.81) were utilized for miRNA and snRNA prediction on the genome. This comprehensive approach identified four types of ncRNAs within the E. gracilis genome, encompassing 188 miRNAs, 4882 tRNAs, 223 rRNAs, and 165 snRNAs.

Gene prediction and annotation

Pre-processing and de novo assembly

The Illumina RNA-seq data underwent initial filtration utilizing Trimmomatic33 (v0.32) to obtain clean reads, subsequently employed in Trinity34 (v2.1.1) for de novo assembly. The Pacbio full-length RNA-seq dataset was refined to derive consensus sequences using smrtlink (v6.0.0).

Transcript integration and ab initio prediction

The two distinct sets of transcripts were amalgamated via PASA35 (v2.4.1) for ab initio gene prediction, utilizing Augustus36 (v2.5.5) and SNAP37 (2006-07-28). Homology annotation was conducted with ten representative species, including Bodo saltans, Naegleria gruberi, Phytomonas sp., Chlamydomonas reinhardtii, Leishmania major Friedlin, Nannochloropsis gaditana, Trypanosoma brucei, Cyanidioschyzon merolae, Leptomonas pyrrhocoris, and Perkinsela sp., downloaded from NCBI. The comprehensive integration of all data and generation of the predicted gene set were accomplished using MAKER38 (v3.01.02). The ensuing analysis revealed a total of 32,806 genes and 39,362 coding DNA sequences (CDSs) within the E. gracilis genome, with an average CDS length of 1,149 bp and an average of 8 exons per gene.

Functional annotation

For functional annotation, blastp30 (v2.2.26) was applied to align protein-coding genes with KEGG39 database. The GO Ontology40 (GO) and InterPro41 function were obtained using InterProScan. The subsequent functional annotation of CDSs demonstrated coverage of 28.2%, 40.6%, and 50.2% across the GO, InterPro, and KEGG databases, respectively, with a cumulative 57.3% of CDSs annotated in at least one database.

Data Records

Sequencing data deposit

The comprehensive E. gracilis genome project has been archived in the Genome Sequence Archive42,43 (GSA) under the accession44 CRA013190, except that the Illumina RNA-seq data have been archived in the SRA at NCBI SRP35377445.

Assembly deposit

The assembly of the E. gracilis genome, along with its corresponding annotation file, is available at figshare46 and NCBI GenBank with accession number GCA_039621445.147.

Technical Validation

Genome assembly quality assessment

The quality assessment of the E. gracilis genome assembly was executed through two distinct methodologies. Firstly, the completeness of the assembly was rigorously validated utilizing compleasm48 (v0.2.2), an improved BUSCO49 workflow based on miniprot, with specific parameters (-m lite–min_identity 0.8–min_length_percent 0.9–min_rise 0.9), and employing the eukaryota_odb10 (v5, 2020-09-10) reference gene set (n = 255). The final BUSCO analysis yielded a completeness score of 80.39%, comprised of 162 (63.53%) single-copy BUSCOs, 43 (16.86%) duplicated BUSCOs, 11 (4.31%) fragmented BUSCOs, and 39 (15.29%) missing BUSCOs. Secondly, to affirm the accuracy and integrity of the genome survey, the filtered Illumina short reads utilized were aligned back to the E. gracilis genome utilizing the Burrows-Wheeler aligner50 (BWA, v0.7.17-r1188). This meticulous alignment process revealed an impressive mapping rate of 99.42% for the short reads against the genome. The combination of these validated results attests to the high-quality nature of the E. gracilis genome assembly.