Predictive design of sigma factor-specific promoters

Area of Science:

• Synthetic Biology
• Molecular Biology
• Bioengineering

Background:

• Engineering complex synthetic gene circuits requires precise control over gene expression using molecular building blocks.
• Automated design tools and tailored components are crucial for fine-tuning gene networks.
• Predictive tools for forward engineering promoter transcription initiation frequency (TIF) are currently lacking.

Purpose of the Study:

• To develop a predictive model for promoter transcription initiation frequency (TIF).
• To assess the orthogonality of sigma factor-specific promoters.
• To create an online tool for designing custom promoters for synthetic genetic systems.

Main Methods:

• Utilized promoter libraries from E. coli (sigma factor 70) and B. subtilis (sigma B, F, W).
• Trained a convolutional neural network (CNN) using high-throughput DNA sequencing data.
• Employed fluorescence-activated cell sorting (FACS) for promoter library analysis.

Main Results:

• Developed a CNN model capable of predicting promoter TIF with high accuracy.
• Demonstrated the model's ability to predict promoter orthogonality among different sigma factors.
• The model serves as the foundation for the online promoter design tool (ProD).

Conclusions:

• The developed prediction model enables accurate TIF and orthogonality prediction for promoter design.
• ProD facilitates the creation of tailored promoters for specific genetic engineering applications.
• This work advances the design and construction of complex synthetic gene circuits.