English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The investigation of plant transcriptional regulation constitutes afundamental basis for crop breeding, where cis-regulatory elements (CREs), asthe key factor determining gene expression, have become the focus of cropgenetic improvement research. Deep learning techniques, leveraging theirexceptional capacity for high-dimensional feature extraction and nonlinearregulatory relationship modeling, have been extensively employed in this field.However, current methodologies present notable limitations: single CNN-basedarchitectures struggle to capture long-range regulatory interactions, whileexisting CNN-Transformer hybrid models demonstrate proneness to overfitting andinadequate generalization in cross-species prediction contexts. To addressthese challenges, this study proposes DeepPlantCRE, a deep-learning frameworkfor plant gene expression prediction and CRE Extraction. The model employs aTransformer-CNN hybrid architecture that achieves enhanced Accuracy, AUC-ROC,and F1-score metrics over existing baselines (DeepCRE and PhytoExpr), withimproved generalization performance and overfitting inhibiting. Cross-speciesvalidation experiments conducted on gene expression datasets from\textit{Gossypium}, \textit{Arabidopsis thaliana}, \textit{Solanumlycopersicum}, \textit{Sorghum bicolor}, and \textit{Arabidopsis thaliana}reveal that the model achieves peak prediction accuracy of 92.3\%, particularlyexcelling in complex genomic data analysis. Furthermore, interpretabilityinvestigations using DeepLIFT and Transcription Factor Motif Discovery from theimportance scores algorithm (TF-MoDISco) demonstrate that the derived motifsfrom our model exhibit high concordance with known transcription factor bindingsites (TFBSs) such as MYR2, TSO1 in JASPAR plant database, substantiating thepotential of biological interpretability and practical agricultural applicationof DeepPlantCRE.

DeepPlantCRE: A Transformer-CNN Hybrid Framework for Plant Gene Expression Modeling and Cross-Species Generalization