PURE is an interpretable framework designed to construct Gene Regulatory Networks (GRNs) and identify key transcription factors (TFs) in plants.
By integrating co-expression patterns, sequence motifs, and orthology-projected in vivo binding evidence, PURE overcomes regulatory data sparsity in non-model species. It employs interpretability-first machine learning (CatBoost + SHAP) to decode transcriptomic programs, allowing researchers to prioritize high-confidence regulators governing stress responses, development, and specific metabolic pathways.