Data and code from: Contemporary concentrations of microplastics in aquatic ecosystems correlate with molecular stress responses in fish
Data files
Jun 18, 2026 version files 23.03 MB
-
coldata_pheno2.txt
6.81 KB
-
DESeq_final_filtered.robj
22.87 MB
-
README.md
4.51 KB
-
WGCNA.R
141.72 KB
Abstract
Microplastics are a pervasive pollutant with impacts on organisms across ecosystems, including effects on behavior, reproduction, feeding, and metabolism. As species increasingly live in the presence of plastic pollution, there is a need to correspondingly deepen our understanding of biological responses to microplastics. To this end, we performed an experiment exposing fathead minnows (Pimephales promelas) to different microplastic treatments. We tested two microplastic concentrations, reflecting both current and predicted future conditions, of two plastic origins: pre-consumer plastic (never exposed to natural environments) and plastic gathered from Lake Ontario. We performed weighted gene co-expression analysis on liver tissue using directional mRNA sequencing data to evaluate gene expression among different treatments across sexes and to explore correlations with ecotoxicological metrics. We addressed the following questions: 1) is there a significant effect of microplastic exposure on gene expression, 2) if there are changes in gene expression attributed to microplastic exposure, do these effects differ across plastic concentration and/or plastic origin, and 3) does the effect of microplastics exposure vary by sex? Our findings provide evidence of metabolic stress-response changes in fish exposed to microplastics. Many of the gene modules found in our analysis correspond to cellular stress responses, further increasing the evidence that microplastic exposure has molecular effects on organisms. Furthermore, we found these molecular effects differed by sex, with a greater response found in females. As fathead minnows are an important toxicological model species, these results have important implications across aquatics species and ecosystems. Understanding how persistent pollutants such as microplastics may stress organism across levels of biological control will be key to mitigating anthropogenic change across levels of biological organization, from the molecular level through populations and communities.
Dataset DOI: 10.5061/dryad.nvx0k6f5r
Description of the data and file structure
Files and variables
File: WGCNA.R
Description: R script containing all code and analyses relevant to the dataset
File: DESeq_final_filtered.robj
Description: Starting DESeq2 object containing the combined readcounts of three library runs with locus tags containing no read counts removed. This file was generated from raw read count data generated by HTSeq v 0.11.2 read into R using the DESeq2 package. This file can be loaded into any R environment with DESeq2 loaded.
File: coldata_pheno2.txt
Description: phenotypic and ecotoxicological data relevant for analyses
Variables
- sample: ID for sample
- treatment: Control, Env_High, Env_Low, Pre_High, Pre_Low
- plastic: pre-consumer (Pre), environmental (Env), or Control (C)
- concentration: high (H), low (L), or control (C)
- sex: male (M) or female (F)
- sex_female: binary for WGCNA, 1=female 0=male
- conc: numeric, 0=control, 1=low, 2=high
- env_plastic: binary for WGCNA
- pre_plastic: binary for WGCNA
- Env_High: binary for WGCNA
- Env_Low: binary for WGCNA
- Pre_High: binary for WGCNA
- Pre_Low: binary for WGCNA
- Control: binary for WGCNA
- egg_diameter_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- eggshell_thickness_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- male_gonad_maturity_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- liver_lipid_female_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- liver_lipid_male_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- female_29_dph_length_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- female_178_dph_length_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- male_178_dph_length_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- female_178dph_CF_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- male_178dph_CF_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- HSI_female_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- HSI_male_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- percent_immature_118dph_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- clutches_per_female_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- total_eggs_per_female_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- average_clutch_size_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- broken_eggs_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- proportion_mal_larvae_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
- larval_length_EF: ecotoxicological metric for effect size, binned based on the effect size from Bucci et al. (2024)
Binned based on the effect size from Bucci et al. (2024) https://doi.org/10.1093/etojnl/vgag167 where 3 = a very large effect size, a confidence interval that does not overlap zero, and statistical significance (p < 0.05), 2 = a very large effect size and a confidence interval that does not overlap zero, but no statistical significance (p > 0.05), 1 = a large effect size, but the confidence interval overlaps zero and the change is not statistically significant, and 0 = no effect.
Code/software
All submissions can be viewed using R; coldata_pheno2.txt can be viewed using any text editor
Access information
Other publicly accessible locations of the data:
- N/A
Data was derived from the following sources:
- N/A
