Cucumber MET Optimization

This repository is related to a study titled: Optimizing the U.S. Pickling Cucumber Multi-Environment Trials Using Envirotyping, Production Importance, and Future Climate Projections.

Integrating envirotyping, environmental clustering, optimization, and future climate comparison to design more efficient cucumber multi-environment trial networks.

---

Overview

This repository contains the data, scripts, and workflow documentation used for the cucumber multi-environment trial (MET) optimization project. The project combines historical trial data, environmental covariates, clustering, scenario-based MET optimization, and future climate projections to identify more informative and cost-effective testing strategies.

The overall objective is to improve trial-network design by addressing four linked questions:

• Which environmental covariates are most relevant for cucumber yield?
• How are historical and county-level environments structured into meaningful clusters?
• How much can the testing network be reduced while preserving useful genetic information?
• How should present and future environmental structure inform trial recommendations?

---

Repository structure

Cucumber_MET_Optimization/
├── Data/
│   ├── raw/
│   └── processed/
├── Figures/
├── Outputs/
├── Scripts/
│   ├── 01_multiyear_env_data_and_predictor_selection_simple.R
│   ├── 02_multilocation_env_data_and_predictor_selection_simple.R
│   ├── 03_combine_final_predictors_simple.R
│   ├── 04_present_counties_env_processing_simple.R
│   ├── 05_met_optimization_models_FINAL_from_analysis.R
│   ├── 06_present_future_clustering_and_alluvial.R
│   ├── README_env_data_processing.md
│   └── future_climate_projections/
│       ├── cmip6_future_weather_colab.ipynb
│       ├── cmip6_future_weather.py
│       └── README_future_climate_projections.md
├── Cucumber_MET_OPT.Rproj
└── README.md

---

Main workflow

The core environmental analysis workflow is located in Scripts/:

1. 01_multiyear_env_data_and_predictor_selection_simple.R
   Multi-year trial (MYT) environmental extraction, processing, predictor selection, and clustering.

2. 02_multilocation_env_data_and_predictor_selection_simple.R
   Multi-location trial (MLT) environmental extraction, processing, predictor selection, and clustering.

3. 03_combine_final_predictors_simple.R
   Combines selected predictors from the MYT and MLT workflows.

4. 04_present_counties_env_processing_simple.R
   Processes present county-level environmental data for downstream analyses.

5. 05_met_optimization_models_FINAL_from_analysis.R
   Compares MET optimization scenarios using historical trial data.

6. 06_present_future_clustering_and_alluvial.R
   Clusters present and future environments separately and creates alluvial-ready comparison outputs.

The future climate extraction workflow is located in:

• Scripts/future_climate_projections/

---

Data

Data/raw/

Contains the main raw inputs used in the project:

• MYT_metadata.xlsx
• MYT_yield.xlsx
• MLT_metadata.xlsx
• MLT_yield.xlsx
• county_acerage.xlsx

Data/processed/

Contains processed and analysis-ready files, including:

• Wmatrix_MYT.csv
• Wmatrix_MLT.csv
• selected_EC_MYT.csv
• selected_EC_MLT.csv
• W_present.csv
• future_matrix_Mar_Jun.csv
• future_matrix_Apr_Jul.csv
• Scenarios_summary.csv

Detailed notes for these folders are available in:

• Data/raw/README.md
• Data/processed/README.md

---

Recommended run order

Run the main workflow in this order:

1. 01_multiyear_env_data_and_predictor_selection_simple.R
2. 02_multilocation_env_data_and_predictor_selection_simple.R
3. 03_combine_final_predictors_simple.R
4. 04_present_counties_env_processing_simple.R
5. future climate extraction workflow in Scripts/future_climate_projections/
6. 06_present_future_clustering_and_alluvial.R
7. 05_met_optimization_models_FINAL_from_analysis.R

---

Documentation

Additional workflow details are available in:

• Scripts/README_env_data_processing.md
• Scripts/future_climate_projections/README_future_climate_projections.md
• Data/raw/README.md
• Data/processed/README.md

---

Reproducibility notes

• File names in the scripts should match the files stored in Data/raw/ and Data/processed/.
• Environment names should be standardized before joining present and future datasets.
• Present and future environments are clustered separately in the comparison workflow.
• Some scripts depend on outputs generated by earlier steps.
• Processed future climate data should be exported in reproducible formats such as .csv for downstream analyses.

---

Intended use

This repository is designed as a research workflow archive for:

• reproducible analysis
• manuscript support
• figure and table generation
• future extension of cucumber MET optimization under changing environments

---

Citation

If you use or adapt this workflow, please cite the associated study once available.

---

License

This repository is released under the MIT License. See the LICENSE file for details.

---

Contact

For questions about the workflow or repository structure, please contact Kashish Grover (grvrkashish@gmail.com).