Can We Predict Gene Expression by Understanding Proximal Promoter Architecture?

Area of Science:

• Computational biology
• Genomics
• Developmental biology

Background:

• Gene expression is regulated by promoter architecture, including transcription factor binding.
• Predicting spatial expression patterns in complex organisms is crucial for understanding development.
• Existing methods show success in predicting expression breadth but struggle with tissue specificity.

Purpose of the Study:

• To review computational methods for predicting gene expression from promoter architecture.
• To highlight challenges and opportunities in predicting spatial expression patterns.
• To discuss the impact of new functional genomics data and potential applications.

Main Methods:

• Review of computational prediction models for gene expression.
• Analysis of successes and limitations in predicting expression breadth and specificity.
• Consideration of machine learning and data mining approaches.
• Evaluation of single-cell expression data and artificial promoter design.

Main Results:

• Computational predictions of gene expression from promoter architecture are advancing.
• Predicting the fraction of tissue types a gene is expressed in (breadth) has seen some success.
• Predicting precise tissue specificity of gene expression remains a significant challenge.

Conclusions:

• Progress in predicting tissue-specific gene expression can be achieved through machine learning and data mining.
• Emerging functional genomics datasets, particularly single-cell expression data, are expected to drive significant advancements.
• The design of artificial promoters represents a practical application of these predictive capabilities.