← Back to the plain-language case
Abstract gives the genome length as 259.28 Mbp; the main text and deposited assembly give 1.27 Gbp
Genome Biol (2025) · doi:10.1186/s13059-025-03542-5 · PMC11959729
Summary
The abstract describes the chromosome-level genome of Ophiura sarsii vadicola as "259.28 Mbp in length with a scaffold N50 length of 66.91 Mbp". The main text describes the same assembly, with the same 66.91 Mbp N50, as 1.27 Gbp, which is 4.9 times larger. Every other size-related number in the paper agrees with 1.27 Gbp and not with 259.28 Mbp: the k-mer estimate (1.21 Gbp), the genome size implied by the stated sequencing coverage (1.255 Gbp), the chromosome lengths in Supplementary Table S3 (summing to 1,268,118,785 bp with the unplaced sequence), and the assembly deposited at NCBI (1,267,624,991 bp). The abstract figure is the headline description of the paper's main genomic resource, so readers who see only the abstract or an indexed summary get a genome size about one fifth of the real one; none of the regeneration findings depend on it.
What the paper reports
Abstract, Results:
Here, we report a chromosome-level genome of Ophiura sarsii vadicola, 259.28 Mbp in length with a scaffold N50 length of 66.91 Mbp.
Main text, subsection "Genome assembly and characterization", first paragraph (the first subsection of the journal's Results; in the PMC copy it follows directly under the Background heading, see Status):
A total of 143.7 Gbp (114.47×) of PacBio SMRT long reads and 218.32 Gbp (173.92×) of Illumina short reads were generated for genome assembly (Additional file 2: Table S1). K-mer analysis estimated the size of O. sarsii vadicola genome to be 1.21 Gbp with a heterozygosity rate of 2.21% (K-mer = 17, Additional file 1: Fig. S1). Using a hybrid assembly approach with 362.02 Gbp of sequence data, we assembled a 1.27 Gbp genome, consisting of 1988 contigs with an N50 of 2.44 Mbp and GC content of 37.2% (Additional file 2: Table S2). After HI-C scaffolding, 99.62% of the assembled contigs were anchored into 19 chromosomes, yielding a final genome size of 1.27 Gbp across 216 scaffolds with an N50 of 66.91 Mbp (Fig. 1A, Additional file 1: Fig. S2 and Additional file 2: Tables S3–S4).
Methods, "Genome sequencing and assembly":
PacBio sequencing generated 143.7 Gbp of data with 114.47× coverage and an N50 length of 20,943 bp.
Contigs were polished with Quiver and aligned to 218.32 Gbp of Illumina data (173.92×) using the BWA v0.6 [81].
Discussion, "Chromosome-level genome features in the phylum Echinodermata":
Our genome showed higher completeness (94.0%) and comparable assembly quality, with a scaffold N50 of 66.91 Mbp, relative to O. sarsii (93.5%, 78.3 Mbp) and A. filiformis (92.7%, 68.8 Mbp) (Additional file 2: Table S4).
Back matter, "Data availability":
The data of genome sequencing and the genome assembly data of Ophiura sarsii vadicola in this study have been deposited at NCBI under the BioProject PRJNA732836 [119].
Additional file 2 (supplementary tables), as downloaded from the publisher: Table S2 gives a total scaffold length of 1,268,118,785 bp (216 scaffolds, scaffold N50 66,915,169 bp, L50 8); Table S3 lists 19 chromosome lengths totalling 1,263,260,924 bp (99.62%) plus 4,857,861 bp unplaced; Table S4 gives the size of this genome as 1,268.11 Mb.
What the paper's own numbers show
Method. I put every genome-size figure the paper gives, and every figure that can be derived from its inputs, on the same scale. Data volume divided by stated fold-coverage gives the genome size the authors used for coverage. The 19 chromosome lengths in Table S3 give the assembly size directly and let N50 and L50 be recomputed. The N50 also constrains the total: half the assembly must lie in scaffolds at least 66.91 Mbp long. Finally I compared these with the assembly deposited under the cited BioProject (NCBI GCA_043643665.1). Every number below is printed by recompute.py.txt.
| Quantity | Location | Reported | Recomputed |
|---|---|---|---|
| Genome length | Abstract | 259.28 Mbp | matches no other figure; 0.2045 of the Table S2 total and of the NCBI total |
| Contig assembly size | Main text, Genome assembly subsection | 1.27 Gbp, 1988 contigs | Table S2 contig total 1,267,941,585 bp |
| Final assembly size | Main text, Genome assembly subsection | 1.27 Gbp, 216 scaffolds, N50 66.91 Mbp | Table S3: 1,263,260,924 + 4,857,861 = 1,268,118,785 bp (equals Table S2) |
| Ratio, body to abstract | Abstract vs main text | 1.27 Gbp vs 259.28 Mbp | 4.898 (4.891 using Table S2) |
| Genome size behind PacBio coverage | Genome assembly subsection; Methods; Table S1 | 143.7 Gbp at 114.47× | 1.2554 Gbp; at 259.28 Mbp the coverage would be 554.2× |
| Genome size behind Illumina coverage | Genome assembly subsection; Methods; Table S1 | 218.32 Gbp at 173.92× | 1.2553 Gbp; at 259.28 Mbp the coverage would be 842.0× |
| Total sequence data | Genome assembly subsection | 362.02 Gbp | 143.7 + 218.32 = 362.02 Gbp |
| K-mer genome size | Genome assembly subsection | 1.21 Gbp | 4.67 times the abstract figure |
| Mean chromosome length | Genome assembly subsection (19 chromosomes, 99.62% anchored) | — | 66.84 Mbp at 1.27 Gbp (19 × N50 = 1271.29 Mbp); 13.65 Mbp at 259.28 Mbp |
| Scaffolds of at least N50 length | Abstract; Table S3 | N50 66.91 Mbp | Table S3: 8 chromosomes ≥ 66,915,169 bp and L50 = 8, as in Table S2; these 8 sum to 662.37 Mbp, 2.55 times 259.28 Mbp; a 259.28 Mbp assembly could hold at most 3 such scaffolds |
| Largest chromosome | Table S3 | 147,934,811 bp | 0.571 of the abstract's total genome length |
| Deposited assembly | Data availability (PRJNA732836) | — | GCA_043643665.1: 1,267,624,991 bp, scaffold N50 66,915,169 bp, L50 8, 19 chromosomes |
| Scaffold N50 | Abstract | 66.91 Mbp | consistent with the main text, Table S2 and NCBI (66.915 Mbp) |
The figures agree among themselves to within 5% (1.21 to 1.27 Gbp), and the abstract's 259.28 Mbp is the only exception. Taken with Table S3, the abstract figure cannot be right. The largest chromosome alone (147.9 Mbp) is more than half of 259.28 Mbp (129.6 Mbp), so an assembly of that size containing it would have a scaffold N50 of at least 147.9 Mbp, not the 66.91 Mbp printed next to it; and the 8 chromosomes of at least 66.91 Mbp already sum to 662.37 Mbp, 2.55 times 259.28 Mbp. The 19 chromosomes listed in Table S3 sum to 1.263 Gbp. The deposited assembly differs from Table S2 by 0.039% in total length (and has 191 rather than 216 scaffolds), which does not bear on the discrepancy.
Figure
Figure 1. The two conflicting sentences, quoted verbatim (the main-text excerpt is abridged at the start and end). The abstract (left) gives the length of the chromosome-level genome as 259.28 Mbp; the main text's Genome assembly subsection (right) gives the final assembly as 1.27 Gbp. Both state a scaffold N50 of 66.91 Mbp, so they describe the same assembly. The supplementary scaffold total and the deposited NCBI assembly, shown underneath for reference, agree with the main text.
Figure 2. Every genome-size figure in the article and Additional file 2 (orange circles), and the sizes recomputed from the paper's own inputs or read from the deposited assembly (blue diamonds), on one linear axis starting at zero. The coverage-based values are data volume divided by stated fold-coverage; the Table S3 value is the sum of the 19 chromosome lengths plus the unplaced sequence. All values except the abstract's lie between 1,210 and 1,270 Mbp.
Why it matters
The genome assembly is one of the paper's two main contributions (the title leads with it, and the abstract's Conclusions call it "Our genomic resource"), and its length is the first number in the abstract's Results. The abstract understates it by a factor of 4.9. Abstracts are what indexing services, search results and many citing papers reproduce, so the 259.28 Mbp figure is the one most likely to be repeated, for instance in comparisons of echinoderm genome sizes, where it would make O. sarsii vadicola smaller than every echinoderm genome in Table S4 (the smallest listed is 383.69 Mb) rather than placing it with the other four ophiuroids listed there (1,184.53 to 2,764.32 Mb). It could also mislead anyone planning sequencing depth or resource use from the abstract.
The error does not touch the analyses. The full text, the supplementary tables and the deposited assembly all have the correct size; the assembly quality metrics (N50, BUSCO completeness) are stated correctly; and none of the regeneration results (cell clusters, connective-tissue-derived progenitors, pseudotime ordering of muscle differentiation) depend on the genome length. A reader of the full text would notice the conflict.
Explanations considered
- A different object (contig set, subset, haplotype or an earlier stage). Does not fit. The abstract pairs 259.28 Mbp with the scaffold N50 of 66.91 Mbp, which the main text gives for the final 1.27 Gbp, 19-chromosome assembly, and the contig assembly is also 1.27 Gbp.
- A unit or decimal slip (Mbp vs Gbp). Does not fit. The ratio is 4.898, not a power of ten.
- A stated fraction of the assembly. Does not fit. 259.28 Mbp is 20.45% of the assembly. The non-repeat part would be about 574 Mbp (45.2%), and the unplaced sequence is 4.86 Mbp (0.38%). No chromosome length in Table S3 equals it (they range from 44.6 to 147.9 Mbp).
- The genome-size estimate rather than the assembly length. Does not fit. The k-mer estimate is 1.21 Gbp, and the coverage figures imply 1.255 Gbp.
- A conversion or typesetting artifact in the archived copy. Does not fit. The PMC page and the publisher-supplied JATS XML in Europe PMC both have "259.28 Mbp" as plain text with an ordinary decimal point. A rendering problem would not produce a value 4.9 times smaller.
- A size that the N50 permits. Does not fit. A 259.28 Mbp assembly with an N50 of 66.91 Mbp could contain at most 3 scaffolds of N50 length, which is incompatible with Table S3, which lists 8 chromosomes of at least that length (together 662.37 Mbp).
- A later correction. None found (see Status).
None of these accounts for the number. The body, supplement and deposited data agree with each other, so the abstract value is the one in error.
Status
As of 2026-10-24:
- Version of record. The publisher's page (genomebiology.biomedcentral.com, redirecting to link.springer.com) and its PDF could not be read; both returned a bot challenge or a login redirect. The wording was checked in the PMC article page and in the publisher-supplied JATS XML from Europe PMC, and both contain "259.28 Mbp" in the abstract and "1.27 Gbp" twice in the "Genome assembly and characterization" subsection. The local PMC archive matches. In these archived copies the only "Results" heading is in the structured abstract; the body's results subsections, including "Genome assembly and characterization", follow directly under the "Background" heading. The publisher's version may carry a Results heading there, so this write-up cites the subsection by name. Every quotation above was checked against the PMC archive; the Data availability sentence was checked in the full-text back matter.
- Supplements. Additional file 2 (13059_2025_3542_MOESM2_ESM.docx, Tables S1–S22) was downloaded from static-content.springer.com and parsed; Tables S1–S4 support 1.27 Gbp (values above). Additional file 1 (figures) and the FigShare dataset (10.6084/m9.figshare.28468406) were not needed for the size and were not examined further.
- Deposited data. NCBI Datasets record for GCA_043643665.1 (BioProject PRJNA732836, released 2024-10-24): total length 1,267,624,991 bp, 19 chromosomes, 191 scaffolds, scaffold N50 66,915,169 bp, L50 8.
- Corrections. No correction, erratum, expression of concern or retraction in Crossref (no update-to, updated-by or relation entries, which also covers the Retraction Watch data Crossref carries), PMC or Europe PMC.
- PubPeer. The PubPeer search and API returned a human-verification wall and could not be queried directly; a web search for the title and PubPeer returned no PubPeer entry.
- Side note, outside this claim. The Genome assembly subsection gives the contig N50 as "2.44 Mbp", whereas Table S2 and NCBI give 2,259,232 bp (2.26 Mbp).
Sources: PMC article; DOI; NCBI GCA_043643665.1; BioProject PRJNA732836; Crossref record; Additional file 2 from the publisher's supplementary material.
Suggested correction
In the abstract, replace "259.28 Mbp in length" with "1.27 Gbp in length", so that the sentence reads: "Here, we report a chromosome-level genome of Ophiura sarsii vadicola, 1.27 Gbp in length with a scaffold N50 length of 66.91 Mbp." The authors may also wish to check the contig N50 in the Genome assembly subsection ("2.44 Mbp") against Table S2 (2,259,232 bp).