STW-MD: a novel spatio-temporal weighting and multi-step decision tree method for considering spatial heterogeneity...

Area of Science:

• Neuroscience
• Genomics
• Computational Biology

Background:

• Gene expression in brain development and disease is complex and dynamic.
• Previous research often overlooked spatio-temporal dynamics and focused on limited regions or stages.
• Understanding these dynamics is crucial for deciphering normal brain development and abnormal conditions like Alzheimer's disease (AD).

Purpose of the Study:

• To develop a novel framework integrating spatio-temporal information for comprehensive analysis of brain gene expression.
• To identify potential determinants of response in brain development and AD progression.
• To address the gap in understanding gene expression dynamics across different brain regions and developmental stages.

Main Methods:

• A novel two-step framework utilizing spatial-temporal information weighting.
• Application of multi-step decision trees for differential gene analysis.
• Analysis of gene expression data from both a brain development dataset (fetal to adulthood) and an Alzheimer's disease (AD) dataset (early, middle, late stages).

Main Results:

• The proposed framework effectively leverages spatial similarity and temporal dependence.
• It facilitates differential gene analysis in heterogeneous brain regions.
• Identified gene classes and elucidated their impact on normal brain development and AD progression across diverse regions and stages.

Conclusions:

• The framework offers a more complete understanding of spatio-temporal gene expression in the brain.
• It provides valuable insights into the molecular mechanisms underlying normal brain development and AD.
• Highlights the importance of considering spatio-temporal dynamics in neurodevelopmental and neurodegenerative research.