- In 2014, faced with a fragmented Illumina assembly for the chloroplast genome of the green algal holobiont Caulerpa ashmeadii, we applied to the MAP to test the prospects of nanopore reads to investigate such intricacies, as well as further explore the hologenome of this species with native and hybrid approaches.. - In hybrid assembly, our modest nanopore data sets showed encouraging results to improve assembly length, contiguity, repeat content, and binning of the larger nuclear and bacterial genomes. - Profiling of the holobiont with nanopore or Illumina data unveiled a dominant Rhodospirillaceae (Alphaproteobacteria) species among six putative endosymbionts. - Conclusion: Our findings relying on a very modest number of nanopore R9 reads as compared to current output with newer chemistries demonstrate the promising prospects of the technology for the assembly and profiling of an algal hologenome and resolution of structural variation. - 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0. - Full list of author information is available at the end of the article. - All Oxford nanopore platforms function on the basis of the same principle (and nanopore chemistries) for DNA sequencing, in which a single-stranded DNA (ssDNA) unfolding from a double stranded DNA (dsDNA) molecule threads through a nanopore embed- ded in a membrane to which voltage is applied. - The translocation of the ssDNA through the nanopore re- sults in a drop in the ion current (raw electrical current aggregated into ‘events. - threading), while the latter relates to DNA quality and fragment size dis- tribution of the sequencing library. - In 2014, faced with a fragmented Illumina assembly for the chloroplast genome of the green algal holobiont Cau- lerpa ashmeadii Harvey 1858 (Bryopsidales, Ulvophyceae, Chlorophyta) (Fig. - are popular in the aquarium trade [14] and infamous. - Raw reads were basecalled in the cloud with the default version of Metrichor available at time of sequencing. - In the present study, we relate our experience with the MAP for the sequencing of DNA extracted from the fronds of C. - Based on modest nanopore 1D and 2D data sets collected during our last testing of the technology with R9 chemistry (August 2016), we report on simple mitigation strategies to improve read size dis- tribution and results of assemblies and genome polishing performed in native and hybrid frameworks. - Among the two R9 flow cell tested, the first one was run with raw genomic DNA (Library Lib RAW ) and led to poor results because of the presence of. - 3), few active pore numbers, and probably inadequate DNA concentration of the sequencing li- brary (inaccuracy of NanoDrop measurements, see methods and Additional file 1). - However, the CP genome could only be circularized in the nanopore as- semblies (both 1Df and 2D). - fragmented), while in the hybrid assemblies, the ‘stitching’ of SVs on the scaffold’s extremities, prevented their circularization. - 2 Caulerpa ashmeadii in the field. - Our modest nanopore data sets were insufficient to as- semble the bacterial and nuclear genomes as seen from the short cumulative length of the 1Df and 2D assemblies (Table 3). - Note the low/uneven coverage of the larger nuclear and bacterial genomes. - Prodigal in the MyCC workflow [18. - Overall, in- vestigation of the reported COGs with BLASTp for the different bins consistently identified six bacterial taxa. - coverage-wise), the CP and MT genome were the most abundant mol- ecules in the cell (Table 2).. - Note structural variation of the psbA intron and interspersed repeats ASH1.1/ASH1.2 recombining in repeat ASH1.3. - Note also the large number of introns in the mitochondrial genome and lower overall gene content and compactness. - Polishing significantly raised the overall quality of the CP and MT genome contigs (Table 4, Additional file 11: Figure S9). - Through the different steps of the polishing pipeline and in the case of the 1Df CP genome, the Canu assembled contigs were >. - Interestingly, the quality of the polished 2D CP genome was overall similar (Table 4) despite its lower read coverage (~215X vs. - In the MT, these tools called numerous SNPs in the atp1 region, which upon close inspection were revealed in error from the mapping of the CP reads containing the conserved atpA gene. - Abundance of genomic compartments in the Nanopore (1Df and 2D) and Illumina PE dataset based on read counts determined via mapping on the best hybrid assembly (PE+1Df) (Nuclear and bacterial) or the curated organellar genomes (chloroplast and mitochondrion). - Note the important abundance of the bacterial compartment and of the Rhodospirillaceae sp1 among bacterial species. - The gene psbA was found in three confor- mations with variation of the number of exons and exon- intron junctions resulting in ORF presence/absence (ORF6, ORF20 and ORF21/22). - The interspersed repeats ASH1.1 and ASH1.2 shared 99% identity and harbored a palin- dromic site GTTTAAAC possibly acting as a restriction site for an endonuclease to excise the genomic segment contain- ing ORF8, subsequently mediating recombination of the re- peat into ASH1.3 (Fig. - In this plot, runs exhibiting some of the highest realized yield by users (up to >. - Genome quality quickly increased through steps of the selected polishing pipeline. - Overall, we noted that while Racon offered some quality im- provement, Nanopolish provided much of the ‘legwork’. - (including greatly extending homopolymers, Additional file 2: Table S6) and Pilon fixed most (but not all) of the remaining errors. - Both Nanopolish and Pilon seemed sensitive to read coverage since the quality of the polished 1Df MT genomes, which exhibited much lower read coverage than the CP genome (42X vs. - However, when comparing the quality of the 1Df and 2D CP gen- ome polished with high read coverage (815X vs. - Unexpectedly, testing for SNPs with Nanopolish highlighted potential is- sues when polishing individual genomes from metagen- omes because of the cross-mapping of conserved genes from different genomic compartments (e.g. - Unfortunately, while we could phase the entire psbA to ORF9 genomic region of the main chlorotype (Additional file 15: Figure S13), we could not deter- mine the pattern of association of the SVs (Fig. - Heteroplasmy was recently documented in the. - Beyond the Ulvophyceae, heteroplasmy was also documented in the model unicellular green microalga Chlamydomonas reinhardtii P.A.Dangeard (Chlorophyceae) and the model brown macroalga Ectocarpus siliculosus (Dillwyn) Lyngbye (Phaeophy- ceae) [25, 26].. - [28, 29], and as suggested by BLASTp reports (Additional file 16: Figure S14 and Additional file 17: Figure S15) of the corresponding ORFs in C. - Consider- ing that psbA represents a critical polypeptide of the photosystem II that is translated at high rates in light conditions [30], we hypothesize that the above mecha- nisms could help maintain efficient repair/transcription/. - We also hypothesize that recombination of in- terspersed repeats ASH1.1 and ASH1.2 into ASH1.3, which results in the excision of ORF8 (Fig. - Future resequencing may help reconstruct longer SVs and further decipher the mech- anistic of DNA recombination in the CP genome. - We did not detect SVs nor SNPs in the MT genome. - Over- all, single-cell genomic experiments [32] of isolated gam- etes may represent an interesting avenue of research to shed further light on all of the above.. - best N50, cu- mulative length and binning/COGs, Table 3, Additional file 8: Figure S6) as a reference hologenome to profile the abundance of the different genomic compartment/. - 24.4 Mbp (Table 3) is in very close range with the above estimates, indicating that the majority of the nuclear genome is present in our current assembly despite being fragmen- ted. - polyploidy of an organ within a diploid organism) may occur in the frond (the. - Resequencing of the nuclear genome at higher coverage than presently available (Table 2) will allow the characterization of SNPs and eventual (chromosome-scale) structural variants to gain further insights into C. - Since the early days of the MinION Access Program in 2014, nanopore sequencing has seen rapid improve- ments, especially in the first two years of the program up to chemistry R9. - low ac- tive pore numbers), this problem has most likely been streamlined by ONT, especially since the commercial launch of the MinION platform. - Total genomic DNA was extracted from live fronds (15 g blotted-dry) originating from a single individual of the C. - The herb- arium voucher is maintained in the personal collections of the primary author.. - Following sequencing, we compared the read size distribution and flow cell out- put of the three libraries to assess effectiveness of the above LMW decontamination strategies.. - NanoDrop), libraries Lib GEL and Lib MAG were each loaded on the second flow cell (ID# FAB38981) with their entire eluted volume of 25 μL, while adjusting the amount of H 2 O in the final loading mix. - All libraries were basecalled with Metrichor (agent 2.3.8.3) in the cloud (metrichor.com) with the corresponding - script (see above). - Prior to assembly, we concatenated each of the 1Df and 2D fastQ data sets (Lib RAW + Lib GEL + Lib MAG ) and filtered them for reads >. - calculated as to exceed the total number of bases/number of reads of the corresponding dataset). - The subsequent assembly step of the corrected/trimmed reads into contigs (Canu produces contigs rather than scaffolds) was performed with the error rate parameter set to 2.5% for 1Df reads and 1.5% for 2D reads, as recommended for R9 data [20] (errorRate of 0.015 and 0.025 in Canu v1.4. - 1000 bp) on the Illumina and hybrid assembly files to sort out genomic compartments and members of the C. - ashmeadii’s genomic com- partments in the different data sets (Illumina and ONT 1Df and 2D), we performed mapping on scaffolds binned from the best assembly (PE+1Df). - For each of the single-end nanopore 1Df and 2D data sets, computations were done for mapped reads excluding secondary align- ments (samtools flag ‘-F 2308. - Quality improvement of the genomes and protein-encoding genes was computed through polish- ing steps against the final curated version via BLASTn (as in the Data quality section further below). - Annotation of the genomes and CP structural variations were carried out in Geneious v11.1.5 (www.geneious.com [62. - We used genoplotR v0.8.9 [67] to graph a summary linear view of the complete CP and MT genomes and the discovered CP SVs. - A more detailed visual of the gene content of these circular genomes was graphed in OGDRAW v1.3.1 [68] and is available in the Additional file 5: Figure S3 and Additional file 6: Figure S4.. - To avoid ambiguous read alignments that may affect these estimates, we used a portion of the CP genome devoid of structural varia- tions or complex regions (a segment of 70,840 bp in full agreement with the Illumina-only contig). - the sum of the length of all reads) as well as the average read length generated. - Additional file 1. - Additional file 2. - Note that due to layering of the plot, numerous scaffolds <. - 20,000 bp in the Illumina and hybrid assemblies become hidden, thus two plots are shown to emphasize (a) bacterial scaffolds or (b) nuclear scaffolds.. - Note as well the successful depletion of the main bacterial kmer peak. - in the bacterial depleted dataset (a and e vs. - Sankey plot depicting MyCC bin correspondence across Illumina and hybrid assembly files (Illumina PE + 1Df or 2D) and their taxonomy/origin in the Caulerpa ashmeadii holobiont based on COGs. - Flow size linking bins represent the number of common scaffolds in the assembly files compared via reciprocal BLASTn. - Note the consistent delimitation of nuclear contigs but some instability in the binning of the Phyllobacteriaceae sp.. - (a) Edited BLASTn report showing identity and common directionality of the interspersed repeats ASH1.1 and ASH1.2 and palindrome GTTTAAAC (italicized and boxed) acting as a potential endonuclease restriction site. - Color coding represents boundaries of the putative excised (blue) and recombined genomic segments (green and red). - Impact of Canu ’ s error rate parameter on assembly of the circular chloroplast genome in the presence of structural variation (SV) (black bubbles). - interpretation and revision of the manuscript. - Acquisition of computing resources (40 cores Linux server) to conduct bioinformatic analyses at SMS was funded by the Laboratories of Analytical Biology (LAB) of the Smithsonian National Museum of Natural History.. - Flow cells were delivered free of charge by ONT as part of the MinION Access Program from 2014 to 2016 (R6 to R9 chemistry) for the testing of the technology on algal samples. - Nanopore sequencing data analysis: state of the art, applications and challenges. - Molecular evidence for the aquarium origin of the green alga Caulerpa taxifolia introduced to the Mediterranean Sea. - Reassessment of the classification of Bryopsidales. - Flip-flop organization in the chloroplast genome of Capsosiphon fulvescens (Ulvophyceae, Chlorophyta). - Divergent copies of the large inverted repeat in the chloroplast genomes of Ulvophycean green algae. - Evidence for persistence of chloroplast markers in the heteroplasmic state in Chlamydomonas reinhardtii. - Inheritance of organelles in artificial hybrids of the isogamous multicellular chromist alga Ectocarpus siliculosus (Phaeophyceae). - Inheritance pattern of chloroplast DNA is correlated with gamete types based on sex-specific arrangement of the cell fusion site in Caulerpa (Ulvophyceae, Chlorophyta). - nov., an endophytic bacterium isolated from the root of the halophyte Suaeda maritima. - Cardiobacterium hominis endocarditis: description of two patients and characterization of the organism. - Mediterranean species of Caulerpa are polyploid with smaller genomes in the invasive ones. - Molecular data suggest a hybrid origin for the invasive Caulerpa racemosa (Caulerpales, Chlorophyta) in the Mediterranean Sea.
Xem thử không khả dụng, vui lòng xem tại trang nguồn hoặc xem
Tóm tắt