MicroarrayDesigner: an online search tool and repository for near-optimal microarray experimental designs.

Area of Science:

• Bioinformatics
• Genomics
• Experimental Design

Background:

• Dual-channel microarray experiments are crucial for differential gene expression analysis across diverse biological contexts.
• Minimizing inference errors in microarray studies is critical for accurate biological interpretation.
• Existing tools for optimal microarray design lack scalability and comprehensive databases.

Purpose of the Study:

• To develop a scalable search tool for near-optimal dual-channel microarray designs.
• To create a continuously updated database of these optimal designs for broader accessibility.
• To provide biologists with readily available, high-quality experimental designs.

Main Methods:

• Implemented a greedy hill-climbing optimization strategy for efficient search of near-optimal designs.
• Developed an improved interwoven loop design construction algorithm for a computable class of designs.
• Established a web-accessible catalog of near-optimal microarray designs.

Main Results:

• The developed search method successfully and efficiently identifies near-optimal dual-channel microarray designs.
• The interwoven loop design algorithm provides a practical and generalizable class of optimal designs.
• A comprehensive and evolving database of optimal designs is now available online.

Conclusions:

• A novel search algorithm and database for near-optimal microarray designs have been successfully developed.
• The developed search tool and database are accessible online, facilitating wider adoption.
• Source code is available for academic use, promoting further research and development.