Cis-Regulatory Code for Predicting Plant Cell-Type Transcriptional Response to High Salinity

Area of Science:

• Plant molecular biology
• Genomics
• Systems biology

Background:

• Multicellular organisms exhibit diverse cell types with specialized functions.
• Transcriptional regulatory programs dictate cell-type-specific environmental responses.
• Understanding plant cell-type specific responses to environmental stress is crucial.

Purpose of the Study:

• To identify cis-regulatory elements (pCREs) involved in high-salinity stress response in Arabidopsis root cell types.
• To develop machine learning models (cis-regulatory codes) for predicting salinity-responsive genes in specific cell types.
• To compare the predictive power of pCRE-based models with transcription factor binding data.

Main Methods:

• Identification of putative cis-regulatory elements (pCREs) in gene promoters responsive to high salinity.
• Application of machine learning to build cis-regulatory codes using pCREs.
• Comparison of model performance using organ-level versus cell-type-specific pCREs.

Main Results:

• pCRE-based models accurately predicted high-salinity stress responses in six Arabidopsis root cell types.
• These models outperformed predictions based on transcription factor binding data.
• Organ-level pCREs showed comparable predictive power to cell-type specific pCREs, highlighting complementary regulatory roles.

Conclusions:

• Cis-regulatory codes provide insights into the spatial transcriptional regulation of plant stress responses.
• The developed approach has broad applicability across species, especially those with limited regulatory data.
• This study advances understanding of how plants adapt to environmental challenges at the cellular level.