Associated publication

Title: FishChip: Development of a High-Throughput qPCR Assay for Freshwater Fish eDNA Quantification and Its Comparison with Droplet Digital PCR

This deposit contains tabular data and metadata used to generate Figures 2–6 of the associated manuscript. All files are comma-separated values (CSV, UTF-8 encoding) unless otherwise noted.

---

Dataset overview

This deposit comprises nine comma-separated (CSV) files and one README metadata file that documents variable definitions, units, and relationships among files. All quantitative tables relate to a panel of 19 species-specific HT-qPCR primer assays targeting nine freshwater fish species (Cyprinus carpio, Danio rerio, Hypophthalmichthys nobilis, Micropterus salmoides, Oreochromis mossambicus, Oryzias latipes, Pseudorasbora parva, Schizothorax sinensis, and Silurus meridionalis). Primer names, target loci (e.g., 16S, 12S, ND1, Cytb, COI), and species assignments are documented in the README files.

Laboratory validation data (Figures 2–3):

• Figure2_ct_standard_curve.csv: HT-qPCR cycle threshold (Ct) values for seven 10-fold dilution levels (LgN = log₁₀ copies/μL, 9–3) across all 19 primers; used for standard-curve fitting.
• Figure3AB_efficiency.csv: amplification efficiency (%) for each of the 19 assays.
• Figure3C_plasmid_DNA_concentration.csv: gradient-dilution Ct matrix for sensitivity assessment (same dilution design as Figure 2).

Field eDNA survey data (Figures 4–5):
Environmental water samples from 39 sites across six freshwater basins/lakes in China (GH, JS, LS, GL, SJ, YR).

• Figure4A_heatmap.csv: site × primer matrix of log₁₀(copies L^-1 + 1) eDNA abundance from HT-qPCR.
• Figure4B_detection_rates.csv: species–basin detection rates (detected sites / total sites per basin).
• Figure5A-F_primer_statistics.csv: per-primer summary statistics (mean, median, SD, min, max, detection count; units: copies L^-1).
• Figure5G_concentration.csv: 285 individual positive detections with Sample, Primer, Species, Ct, copies/μL, copies L^-1, and log₁₀(copies L^-1).

HT-qPCR versus ddPCR comparison (Figure 6):

• Figure6_ht_dd_compare.csv: 66 paired sample–primer observations with HT-qPCR and ddPCR concentrations (copies L^-1) and log₁₀-transformed values used for method comparison.
• Figure6_bland_altman.csv: per-pair and summary Bland–Altman agreement statistics on the log₁₀ scale.

The README file describes column names, units, missing-value conventions, derivation from raw HT-qPCR outputs, and links between files. Concentrations are expressed as copies per litre of environmental water unless otherwise noted; Ct values are instrument readouts without separate units.

The dataset documents laboratory validation and field application of a 19-primer HT-qPCR panel targeting nine freshwater fish species across six basins/lakes in China. Quantitative outputs include cycle threshold (Ct) values, amplification efficiency, absolute eDNA concentrations (copies L^-1), site-level detection rates, and paired HT-qPCR versus droplet digital PCR (ddPCR) measurements.

Target species and primers

Nine species were assayed with two or three independent mitochondrial primers each:

Primer sequences, target regions, and amplicon lengths are provided in Supplementary Table S5 of the associated manuscript.

Study regions (basin codes)

A total of 44 water samples were collected. Five sites had no positive eDNA detections for any of the nine target fish species and were removed from the dataset, and the remaining 39 sites were retained for analysis.

Sampling intensity differed among basins; cross-basin comparisons should account for unequal site numbers.

File inventory

---

Figure 2 — Standard curve Ct data

File: Figure2_ct_standard_curve.csv

Raw Ct values from HT-qPCR standard-curve experiments for nine fish species. These data were used to fit standard curves (Ct versus log copy number) and to convert field-sample Ct values into absolute eDNA copy numbers.

Structure

• Rows: 7 (one per 10-fold dilution level)
• Columns: 20 (LgN + 19 primer Ct columns)
• Missing values: none

Variables

Primer column order

Columns 2–20 correspond to: 16S_Oreo, 12S_Oreo, Sm_ND1, Sm_Cytb1, Sm_Cytb2, Msa, MiSa_16S, Lc_ND1_2, Lc_ND2, HYNO_COI_2, BHTM1, CYCA_COI, CpCyB_496, OlaND5, Ola, Dre_cytb_V132, Dre_COI, Ppar_CO1, Pp_COI.

Analysis notes

For each primer, Ct (y-axis) was regressed against LgN (x-axis): Ct = slope × LgN + intercept. A good standard curve typically has R² close to 1 and a slope near −3.3 (~100% amplification efficiency).

---

Figure 3 — Primer validation

Figure 3A–B: Amplification efficiency

File: Figure3AB_efficiency.csv

Summary (n = 19 assays): mean 101.6%, SD 4.0%, range 91.7–105.8%. Most assays (63.2%) fell within 100–105%. Values close to 100% indicate ideal exponential amplification.

Figure use: Figure 3A displays individual primer efficiencies; Figure 3B shows the frequency distribution in 5% bins (90–95%, 95–100%, 100–105%, 105–110%).

Figure 3C: Gradient dilution DNA concentration matrix

File: Figure3C_plasmid_DNA_concentration.csv

DNA template concentrations measured for each primer across seven 10-fold dilution levels (LgN 9 → 3). Used for the DNA concentration matrix (tile plot) in Figure 3C. Related Ct values at the same dilution design are in Figure2_ct_standard_curve.csv.

Naming note: Con9 corresponds to LgN = 9, Con8 to LgN = 8, and so on. Minor primer-name variants in this file (e.g., 12S_oreo, HYNO_COI_F2/R2) refer to the same assays as 12S_Oreo and HYNO_COI_2 elsewhere.

---

Figure 4 — Field eDNA survey

Environmental water samples were analyzed by HT-qPCR. Ct values were converted to absolute copy number using primer-specific standard curves and normalized to copies per litre of sampled water (copies L^-1).

Figure 4A: Site-level abundance heatmap

File: Figure4A_heatmap.csv

Each primer column value is computed as log₁₀(copies_L + 1). A value of 0 indicates zero or unreported copy number (log₁₀(0 + 1) = 0). To recover approximate copies L^{-1 from entry x}: copies_L ≈ 10^x − 1 (for x > 0).

Abundance is reported at the primer level; species with two or three primers appear as separate columns. Row order matches Figure 4A (not hierarchical clustering).

Figure 4B: Basin–species detection rates

File: Figure4B_detection_rates.csv

One row per species–basin combination where the species was detected at least once in that basin.

A species was counted as detected at a site if any primer assigned to that species yielded a positive result. Species–basin combinations with no detections are omitted. For visualization, detection rates are often grouped as high (0.6–1.0), medium (0.3–0.6), or low (0–0.3).

Distinction from Figure 4A: Figure 4B reports site-level presence (detection rate); Figure 4A retains continuous abundance estimates.

---

Figure 5 — Multi-primer detection performance

Figure 5A–F: Primer-level summary statistics

File: Figure5A-F_primer_statistics.csv

Summary of eDNA concentration distributions across all positive detections from all sites and regions combined. Supports chord diagrams in Figure 5A–F.

The count variable is the total number of site-level positive detections per primer, not the number of sampling sites or regions.

Figure 5G: Individual positive detections

File: Figure5G_concentration.csv

285 positive detection records from 40 unique sampling sites across six regions. Each row is one positive primer–sample combination.

Negative or undetected primer–sample combinations are not included. Figure 5G displays copies_L_log10 by primer.

---

Figure 6 — HT-qPCR versus ddPCR comparison

A subset of 32 sampling sites across six regions was selected for parallel validation. Sixty-six HT-qPCR-positive sample–primer combinations were re-quantified by ddPCR on the same eDNA extracts.

Figure 6A: Paired concentration data

File: Figure6_ht_dd_compare.csv

HT-qPCR quantification used plasmid standard curves; ddPCR used Poisson-based absolute quantification without a standard curve. Log-transformed variables HT_qPCR and ddPCR were used for Deming regression (Figure 6A).

Figure 6B: Bland–Altman agreement analysis

File: Figure6_bland_altman.csv

Per-observation and summary Bland–Altman outputs for the same 66 pairs on the log₁₀ scale.

Summary statistics (n = 66):

---

Relationships between files

Figure2_ct_standard_curve.csv ──► standard curves ──► copies_L in field data
Figure3AB_efficiency.csv       ──► primer QC (Figure 3A–B)
Figure3C_plasmid_DNA_concentration.csv ──► template amounts (Figure 3C)
Figure4A_heatmap.csv           ──► site abundance matrix
Figure4B_detection_rates.csv   ──► aggregated from site-level detections
Figure5G_concentration.csv     ──► individual positive records
Figure5A-F_primer_statistics.csv ──► summary of Figure5G_concentration.csv
Figure6_ht_dd_compare.csv      ──► paired method comparison
Figure6_bland_altman.csv       ──► derived from Figure6_ht_dd_compare.csv

---

General usage notes

• Ct: instrument readout; lower Ct indicates higher template amount.
• copies L^-1: primary unit for field eDNA abundance; concentrations depend on standard-curve calibration, extraction efficiency, and primer specificity.
• Zero or low values may reflect true absence, inhibition, or limits of quantification rather than biological zero abundance.
• Multi-primer species should be interpreted per primer unless explicitly combined in separate analyses.
• Detection versus abundance: detection rates (Figure 4B) reflect site-level presence; abundance tables (Figures 4A, 5G) retain quantitative estimates.
• Files can be opened in Excel, R, Python, or GraphPad Prism.

---

Legal and ethical considerations

This dataset contains no human subjects data and no personally identifiable information. Sampling locations are identified by site codes; geographic coordinates are described in the manuscript Methods section. Data were collected under applicable institutional and environmental sampling permissions as reported in the associated publication.

---

Citation

Although this dataset is released under CC0 and attribution is not legally required, citation of the associated manuscript and this Dryad dataset is appreciated when the data are reused.

Contact

For questions regarding primer design, HT-qPCR protocols, sampling design, or data interpretation, please contact the corresponding author of the associated publication.

License

This dataset is dedicated to the public domain under the CC0 1.0 Universal Public Domain Dedication.