Transcript expression analysis of metabolic detoxification genes in the Colorado potato beetle
Data files
Jun 04, 2026 version files 109.11 MB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_AARS.csv
3.40 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_HARS.csv
5.06 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_MD.csv
4.87 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_ME.csv
4.87 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_NC.csv
4.25 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_NY.csv
4.87 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_OR.csv
4.87 KB
-
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_VA.csv
4.87 KB
-
2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_AARS_2nd_gen.csv
5.83 KB
-
2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_HARS_2nd_gen_E20.csv
6.40 KB
-
2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_HARS_2nd_K1.csv
6.21 KB
-
2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_ND.csv
5.73 KB
-
Admire_Assays_-_probit_points.csv
12.06 KB
-
ISM-555_Adult_Assays_v20200819.xlsx_-_All_Data.csv
77.90 KB
-
ISM-555_Probit_Points.csv
3.89 KB
-
Quantification_hisat_salmon_2.zip
108.90 MB
-
README.md
10.50 KB
-
sample_ids.csv
20.99 KB
-
tx_doc.csv
19.42 KB
Abstract
This dataset contains raw and derived data from adult Colorado potato beetle (Leptinotarsa decemlineata) insecticide bioassays and targeted transcript expression experiments conducted across multiple U.S. populations. Phenotypic data include replicate-level counts of alive, incapacitated, and dead beetles following exposure to a range of concentrations of imidacloprid and isocycloseram, recorded at 48 and 96 hours post-exposure. These data are used to estimate dose-response relationships, lethal dose metrics, and resistance ratios for each population and exposure duration. The dataset also includes targeted transcript expression measurements for a set of 299 genes previously associated in the literature with metabolic detoxification and insecticide resistance. Transcript expression data were generated for selected populations and treatments to support analyses linking phenotypic resistance patterns with molecular responses to insecticide exposure. The dataset may be reused for comparative analyses of insecticide susceptibility, resistance evolution, cross-insecticide resistance, and detoxification-related gene expression in agricultural pest populations.
Dataset DOI: 10.5061/dryad.d7wm37qf3
Description of the data and file structure
Data were collected as part of controlled laboratory experiments designed to quantify insecticide susceptibility and resistance in adult Colorado potato beetle (Leptinotarsa decemlineata) populations from multiple U.S. locations. Adult beetles were exposed to a range of concentrations of two insecticides, imidacloprid and isocycloseram, and mortality and incapacitation were recorded at 48 and 96 hours post-exposure. These experiments were conducted to estimate dose-response relationships, lethal dose metrics, and resistance ratios within and across populations.
In addition, targeted transcript expression data were generated for a select set of 299 genes previously associated with metabolic detoxification and insecticide resistance. Transcript expression was quantified for selected populations and treatments to evaluate molecular responses to insecticide exposure and to support analyses linking phenotypic resistance patterns with detoxification-related gene expression.
Files and variables
Phenotypic resistance bioassay data (Colorado potato beetle)
Imidacloprid (Admire®) adult bioassays
The following CSV files contain raw dose-response bioassay data for adult Colorado potato beetle populations exposed to imidacloprid. Each file corresponds to a single population and assay version. Rows represent individual experimental replicates conducted at a specific dose and exposure duration.
2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_AARS.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_HARS.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_MD.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_ME.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_NC.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_NY.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_OR.csv2021_CPB_Adult_Assay_Data_v20210702.xlsx_-_VA.csv2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_ND.csv2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_AARS_2nd_gen.csv2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_HARS_2nd_K1.csv2021_CPB_Adult_Assay_Data_v20210809.xlsx_-_HARS_2nd_gen_E20.csv
Isocycloseram (Plinazolin®) adult bioassays
ISM-555_Adult_Assays_v20200819.xlsx_-_All_Data.csv
This file contains raw adult dose-response bioassay data for isocycloseram, with all populations combined in a single table. Rows represent individual replicates at a given dose and exposure duration.
Derived dose-response and resistance metrics
The following files were generated during statistical analysis and are included to support reproducibility and reuse.
Admire_Assays_-_probit_points.csv
Probit regression outputs for imidacloprid assays, including LD20, LD50, and LD90 estimates with 95 % confidence intervals for each population and exposure duration.ISM-555_Probit_Points.csv
Probit regression outputs for isocycloseram assays, including LD20, LD50, and LD90 estimates with 95 % confidence intervals.
Variable definitions (raw bioassay files)
The following variables are common to all raw bioassay CSV files unless otherwise noted:
- population: Population identifier (site name and generation, where applicable)
- date: Date the bioassay was conducted
- dose_ppm: Insecticide concentration (parts per million, ppm)
- time_hrs: Exposure duration in hours
- rep: Replicate identifier
- n: Number of insects tested in the replicate
- alive: Number of insects alive at scoring
- incap: Number of insects incapacitated at scoring
- dead: Number of insects dead at scoring
- harmed: Total number of insects dead or incapacitated
- total: Total number of insects scored
Variable definitions (derived probit output files)
- prob: Target response probability (e.g., 0.5 for LD50)
- prob_label: Label corresponding to the target probability (e.g., LD20, LD50, LD90)
- dose_ppm: Estimated lethal dose (ppm)
- dose_ppm_lower: Lower 95 % confidence limit of the dose estimate
- dose_ppm_upper: Upper 95 % confidence limit of the dose estimate
- time_hrs: Exposure duration associated with the estimate
- population: Population identifier
Missing values
Missing values are indicated as NA. Blank cells are not used.
Genetic / transcriptomic data files
RNA-seq sample metadata
sample_ids.csv
Metadata table linking each RNA-seq sample to population, treatment, and dose information. This file is used to define sample identities and experimental factors for transcriptomic analyses.
Variables (sample_ids.csv):
- id: Unique sample identifier used to match RNA-seq quantification directories
- location: Population/collection location
- chemical: Treatment group (Control, ISM, Imidacloprid)
- sample: Sample number within a population × treatment group
- rep: Biological replicate identifier
- dose_ppm: Insecticide dose (ppm); controls recorded as 0
Derived variables used in analysis:
- dose_log: log10-transformed dose (ppm), with control values set to 0
- dose_log_sc: Scaled (z-scored) version of
dose_logused in statistical models - sample_id: Composite identifier combining location, chemical, sample number, and dose
Transcript-to-gene mapping
tx_doc.csv
Reference table mapping transcript identifiers to gene identifiers. This file is used to aggregate transcript-level RNA-seq quantifications to gene-level counts.
Variables (tx_doc.csv):
- tx: Transcript identifier
- gene: Corresponding gene identifier
RNA-seq quantification files
Quantification_hisat_salmon_2.zip/(directory)
Directory containing per-sample Salmon quantification outputs generated from HISAT2-aligned RNA-seq reads. Each subdirectory corresponds to a single biological sample.
Each sample directory (e.g., AARS1_R1_t.bam.quant/) contains:
AARS_R1_quant.sf: Transcript-level abundance estimates produced by Salmonaux_info/: Auxiliary Salmon metadatacmd_info.json: Command and parameter informationlogs/: Salmon run logslibParams/: Library parameter information
Variables (AARS_R1_quant.sf):
- Name: Transcript identifier
- Length: Transcript length (base pairs)
- EffectiveLength: Effective transcript length used for quantification
- TPM: Transcripts per million
- NumReads: Estimated number of reads assigned to the transcript
Processing note: Transcript-level quantifications were imported using tximport with length-scaled TPMs to generate gene-level count matrices for downstream differential expression analyses.
Missing values
Missing values in tabular files are indicated as NA. In RNA-seq quantification files, transcripts with zero estimated abundance may be absent from individual AARS_R1_quant.sf files.
Code/software
Metabolic Detox Phenotypic.Rmd
R Markdown file used to process and analyze adult phenotypic bioassay data for imidacloprid and isocycloseram. This document imports raw assay CSV files, performs data cleaning and correction, fits dose–response models using probit regression, calculates lethal dose estimates (LD20, LD50, LD90) with confidence intervals, computes resistance ratios (RR50) relative to a susceptible reference population, and generates summary tables and figures used in downstream analyses.
Inputs:
- Raw adult bioassay CSV files (imidacloprid and isocycloseram assays)
Outputs:
- Probit model summaries
- LD estimate tables
- Resistance ratio (RR50) tables
- Figures visualizing dose–response and cross-insecticide resistance patterns
Metabolic Detox Transcript Exp.Rmd
R Markdown file used to analyze RNA-seq gene expression data associated with metabolic detoxification responses. This document imports transcript-level quantifications generated by Salmon, aggregates transcripts to genes using a transcript-to-gene mapping file, constructs DESeq2 objects, performs normalization and variance stabilization, and conducts differential expression analyses across treatments and doses. Results are summarized in tables and figures for interpretation of detoxification-related gene expression patterns.
Inputs:
- Salmon quantification files (
AARS_R1_quant.sf) - Sample metadata file (
sample_ids.csv) - Transcript-to-gene mapping file (
tx_doc.csv)
Outputs:
- Normalized gene-level count matrices
- Differential expression results tables
- Figures summarizing transcriptomic responses
All data files are provided in non-proprietary formats (CSV) and can be viewed using standard spreadsheet software or any text editor. Reproduction of analyses and generation of derived data products require the R statistical computing environment.
Software environment
- R (version ≥ 4.3)
- RStudio (Posit Software, PBC) used as the integrated development environment
R packages used
Analyses were conducted using the following open-source R packages:
Data manipulation and visualization
- tidyverse (includes dplyr, stringr, readr, ggplot2)
- patchwork (figure assembly)
Dose–response, statistics, and modeling
- ecotox (dose-response and toxicity-related utilities)
- car (regression diagnostics and hypothesis testing)
- ARTool (aligned rank transform models for nonparametric factorial analyses)
- emmeans (estimated marginal means and post hoc contrasts)
Transcriptomic analysis
- tximport (import and aggregation of transcript-level RNA-seq quantifications)
- DESeq2 (differential gene expression analysis)
- vsn (variance stabilizing normalization)
File system utilities
- fs (file system operations used to manage quantification directories)
Access information
Other publicly accessible locations of the data:
- None. All data associated with this study are archived in this Dryad repository.
Data was derived from the following sources:
- None. All phenotypic bioassay and targeted transcript expression data were generated by the authors as part of the described experiments. Reference annotations used for transcript-to-gene mapping are described in the associated manuscript and accompanying documentation.
Methods for Data Collection
Colorado potato beetle Collection
Newly emerged, overwintered adult L.decemlineata were requested from several field locations around the continental U.S. in the late spring and early summer of 2020 and 2021 for leaf-dip bioassays against two test substances/active ingredients (imidacloprid (Admire® Pro Systemic Protectant) and isocycloseram (Plinazolin®)).
Approximately 300 adult beetles were collected from field locations by cooperators and mailed overnight in refrigerated containers to the University of Wisconsin, Vegetable Entomology Laboratory. Upon arrival, all adult insects were released into cages containing untreated, 3-5-week-old greenhouse-grown potato plants for a 72-96-hour acclimation period in advance of the assay. Adults (n = 216-270) from each field-collected population were assayed to determine estimates of the median lethal dose (LC50, the dose necessary to kill 50 % of test insects) associated with serial exposures to both imidacloprid and isocycloseram for that population.
Lethal Concentration Assessments within Populations
Following the acclimation period, healthy adults were randomly selected from among the test population and divided into nine groups representing the serial dilutions of the test substance (n = 24 individuals per group), with six biological replicates. Serial dilutions in water for imidacloprid measured in ppm were: 0.001, 0.01, 0.1, 0, 1, 10, 100, and 1000. Serial dilutions for isocycloseram measured in parts per million (ppm) included: 0.001, 0.005, 0.01, 0.05, 0.1, 0.5, 0, 1, and 10. Dilution ranges were chosen based on preliminary dose response assays performed in the laboratory.
To begin the bioassay, a terminal trifoliate potato leaf was removed from a fully expanded, greenhouse-grown potato plant (Solanum tuberosum cv. Silverton Russet) for each dilution of the test substance. To ensure consistency over the assay period, the terminal trifoliate leaf was selected to be sufficiently large so as not to limit feeding by the adults prior to the conclusion of the four-day bioassay period. Briefly, leaflets were individually dipped in the nine representative dilutions of the respective test substance by grasping the exposed end of a petiole and completely submerging the remainder of the leaflets. Leaflets were allowed to completely dry within a chemical fume hood. In advance of the initial exposure of adults, filter paper was placed in the bottom of 150 x 15 mm Petri dishes. Air-dried leaflets were then placed into labeled dishes and four L. decemlineata adults were transferred using forceps. Adults were allowed to feed on treated foliage for 96 hours in an incubator held at 26 °C, 70 % humidity, and a light and dark (L:D) cycle of 16:8 hours, and a second series of treated foliage containing the same serial dose was provided after 48 hours. After 48 and 96 hours, the number of live, incapacitated (defined as beetles unable to right themselves when flipped over), and dead adults were recorded. Incapacitated and dead beetles were pooled and LC50 values for each population were calculated and fitted with 95 % confidence limits. Abbott’s correction was used to normalize the data, and a log10 Probit regression analysis was performed in R version 1.2.1 using the ecotox package. Surviving beetles from every dose in the 48-hour assays were flash frozen in liquid nitrogen and stored at -80°C for downstream genetic analyses.
In R (v4.3.2) statistical analyses were conducted to assess the potential for phenotypic cross resistance. Shapiro-Wilk tests were used to evaluate the normality of log-transformed LC values for imidacloprid and isocycloseram. Due to significant data deviation from normality (p < 0.001), we used non-parametric methods for correlation and variance analysis. Spearman’s rank correlation and Kendall’s tau were calculated to assess the monotonic association between the dose-responses values of both insecticides across all populations and calculated LC levels (LC20, LC50, LC90). To test for homogeneity of variance, we performed Levene’s test using median-centered groupings. An aligned rank transform ANOVA (ART ANOVA; art function, ARTool package) was used to test the main and interaction effects of insecticide and population on dose response, allowing for factorial analysis under non-parametric assumptions. Post hoc pairwise comparisons were conducted using estimated marginal means (emmeans package), with contrasts performed between insecticides within each population. Significance thresholds were set at α = 0.05, and p-values were adjusted for multiple comparisons where applicable.
Sequencing, Quantification, and Quality Control
Total RNA was extracted for sequencing following select pesticide exposures for each population, which were chosen using estimated LC20 and LC50 values for each substance. Approximate doses that most closely matched LC20 and LC50 values for respective populations were chosen to reflect the induced genetic response to both insecticides, while avoiding extreme cellular toxicosis. For imidacloprid, extracted dilutions included 0 ppm, 0.1 ppm, and 10 ppm, with the addition of 100 ppm for two more phenotypically resistant populations (NY and ME). For isocycloseram, extracted dilutions included 0 ppm, 0.005 ppm, 0.5 ppm for all beetle populations. Total RNA was extracted from adult beetles (n = 6) at each chosen dilution using Trizol (Life Technology) and the quality and quantity of RNA was measured using a Nanodrop spectrophotometer (Thermo Fisher Scientific). Extracted RNA samples were equalized using Qubit Fluorometric Quantification system and converted to cDNA using SuperScript III kits (Thermo Fisher Scientific). A total of 20 μl of beetle cDNA for each biological replication (n = 6) was submitted to Floodlight Genomics LLC (https://floodlightgenomics.com/) for targeted amplicon sequencing. Gene targets for the amplification (n = 299) were chosen based on data from previous studies that identified candidate resistance genes associated with metabolic detoxification and target site resistance. Gene families on this list include, but are not limited to, cytochrome p450s, GSTs, ABC transporters, and esterases. Custom primer sequences were designed to specific transcripts of interest using the Primer-BLAST tool.
Primers and cDNA were provided to Floodlight Genomics for processing using their MonsterPlex workflow. MonsterPlex, an optimized amplicon sequencing approach, includes triplicate PCR amplification of each biological replicate to minimize variation and sequencing on the Illumina HiSeq X platform following the manufacturer's directions. Floodlight Genomics delivered sample-specific raw transcript sequence reads, which were aligned to annotated sequences. The data were sequenced using the NovaSEQ X Plus device with a 2x150bp configuration, merging all paired reads during processing to enhance coverage, which ranged from approximately 100,000 reads to over 1,000,000 per sample. A replicated library of samples was also provided by Floodlight Genomics for quality control.
The quality of the raw cDNA amplicon sequencing reads was assessed using MultiQCv1.23 and then trimmed using Cutadapt v1.18 to improve the quality of analysis. Salmon v1.10.2 was then employed for sequencing data analysis to quantify transcript abundances, accounting for sample-specific parameters and biases. Using a quasi-mapping approach, Salmon processes raw sequencing reads directly without generating intermediate alignment files, enhancing efficiency and reducing computational requirements. However, to further increase read coverage, pre-computed alignments were first generated using HISAT2 v2.2.1 by mapping to the L. decemlineata reference genome (NCBI Refseq accession: GCF_000500325) and then processed by Salmon to estimate read counts.