CGIG_OA_RNASEQ_PRJNA1196326

Primary-analysis sample count: 6. The counted unit and independent experimental units are described under Biological replication; counts may represent animals, pools or libraries.

Treatment

Eyed larvae whose parents were held under elevated pCO2 (acidified, pH 7.4); BioSample isolate “acidification group”, sample names S74_1-3

Control

Eyed larvae whose parents were held at ambient pH 8.1; BioSample isolate “wild type group”, sample names S81_1-3

Biological replication

6 RNA-seq libraries of eyed larvae, 3 from the acidified-parent group and 3 from the ambient-parent group. Each library is presumably a pool of many larvae; the public metadata do not say whether the three replicates are separate culture tanks, separate crosses or subsamples of one culture.

Phenotype interpretation

The BioProject frames the study as transcriptional acclimation of offspring after parental acidification exposure; no larval performance or survival phenotype is deposited, so AREE records the larval transcriptional response to parental acidification, not a measured tolerance phenotype

Analysis status

raw_reanalysis_complete_not_harmonized

Quality control and results

All 121,029,855 read pairs quantified with Salmon 2.3.4 against GCF_963853765.1/RS_2024_06 (stranded ISR); mapping rates 68.92-78.45% (median 75.99%), not different between arms (control mean 73.6% versus parental_acidified 76.3%); expressed-SNP allele frequencies at 4,106 sites show the three libraries in each arm are one culture, either one RNA sample or large pools of larvae from one culture (within-arm mean difference 0.061 against 0.051 expected from resequencing the same RNA, r = 0.90), and that the two parental groups are genetically distinct (between arms 0.235 against 0.056, comparable to the 0.271 between two unrelated DECICOMP full-sib families), so parental treatment is fully confounded with parentage and there is one culture per arm; collapsing to one sample per arm leaves 1 v 1, which DESeq2 cannot test, so the 3 v 3 contrast was run for descriptive fold changes only: 10,466 of 21,179 tested genes reach a nominal FDR < 0.05 (5,005 higher, 5,461 lower in parental_acidified), inflated by pseudoreplication; PC1 (94.2%) separates the arms

Limitations

  • No publication is linked to the BioProject, and searches of PubMed, Europe PMC (including the accession) and Crossref on 2026-09-29 found none, so there are no published methods, results or effect tables to reconcile a reanalysis against. The BioProject title matches Goncalves et al. 2016 (Mol Ecol, doi:10.1111/mec.13808) except for the word “Rapid”, but that paper studied Saccostrea glomerata and is not this dataset.
  • The contrast compares offspring of two parental groups, so parental treatment is confounded with parentage: the acidified and ambient larvae necessarily have different parents. Unless both parental groups were drawn at random from one broodstock, genetic background can differ between arms, as it did in the sister IOCAS study PRJNA826964. The reanalysis should run the expressed-SNP check (scripts/check_replicate_genotypes.py) to measure between-arm genetic distance and to test whether the three replicates per arm are independent.
  • Whether the larvae were reared at ambient pH or at their parents’ pH is not recorded, so the contrast cannot be separated into transgenerational and within-generation effects. It is recorded as the effect of parental acidification exposure on eyed-larval gene expression.
  • Parental exposure duration, pCO2, carbonate chemistry, temperature, larval age in days and rearing density are not in the public metadata.
  • Condition and replicate come from the NCBI BioSample isolate attributes (“acidification group 1-3” for S74, “wild type group 1-3” for S81). They agree with the pH in the sample-name prefixes (S74 = pH 7.4, S81 = pH 8.1), and every library is paired-end with ENA FASTQ MD5s recorded in the manifest.
  • With three libraries per arm, the contrast has four residual degrees of freedom, fewer if the replicates share a culture.
  • Not harmonized into the evidence base (decision 2026-10-02). The AREE expressed-SNP check (scripts/check_replicate_genotypes.py, 4M read pairs per library) found the three libraries per arm no more different than resequencing one culture, and the two arms as genetically distant as unrelated full-sib families, so there is one culture per arm, parentage and parental pH are inseparable, and no valid test of the parental-acidification effect exists. The 3 v 3 DESeq2 results are kept outside the repository as descriptive fold changes only.

Sources