TransformPest-Atlas: Chemical identity, evidence provenance, and paired monitoring of pesticide transformation products

Description

TransformPest-Atlas provides analysis-ready derived data and validation code for evaluating pesticide transformation products across chemical identity, evidence provenance, toxicological evidence, and paired field monitoring. The dataset contains 9,587 exact-identity chemical entities, 45 reference-supported transformation-product entities, source-lineage corroboration records, 27 parent–transformation-product monitoring relationships derived from U.S. Geological Survey stream-monitoring data, and deterministic validation code. Raw third-party bulk data are not redistributed; source references, provenance information, and reuse notes are provided with the dataset.

Steps to reproduce

Download the dataset while preserving the directory structure. The analysis-ready tables are provided as CSV files under the data directory. Repository-level consistency can be verified with Python 3 and pandas by running: python code/validate_repository.py. The script checks the chemical-identity universe, reference-supported transformation-product inventory, source-lineage corroboration, paired-monitoring relationships and event totals, evidence registry, and ToxCast-derived tables. A successful validation terminates with “VALIDATION_STATUS=PASS”. File-level SHA256 checksums are provided in SHA256SUMS.txt. Definitions of variables and source provenance are documented in documentation/data_dictionary.md and documentation/source_references.csv.

Institutions

• Guangxi Subtropical Crops Research Institute

  Guangxi, Nanning

Categories

Licence