PBMDR: A particle swarm optimization-based multifactor dimensionality reduction for the detection of multilocus...

Area of Science:

• Genetics
• Computational Biology
• Bioinformatics

Background:

• Identifying susceptibility genes for complex diseases is challenging.
• Multilocus interactions are crucial but difficult to detect.
• Existing methods for analyzing gene-gene interactions have limitations.

Purpose of the Study:

• To propose a novel particle swarm optimization-based multifactor dimensionality reduction (PBMDR) approach.
• To enhance the detection of multilocus interactions for complex diseases.
• To evaluate the performance of PBMDR against existing algorithms.

Main Methods:

• A particle swarm optimization (PSO)-based multifactor dimensionality reduction (MDR) approach was developed (PBMDR).
• Simulated genotype data from 26 SNPs in eight breast-cancer-related genes were used.
• PBMDR was compared with other global optimization algorithms, including PSO and chaotic PSOs.

Main Results:

• PBMDR demonstrated superior ability to explore and detect specific SNP-genotype combinations in simulated disease models.
• The PBMDR algorithm achieved higher accuracy and chi-square values compared to other tested algorithms.
• PBMDR outperformed existing global optimization algorithms in detecting multilocus interactions.

Conclusions:

• The proposed PBMDR algorithm is an effective tool for identifying multilocus interactions associated with complex diseases.
• PBMDR offers improved accuracy and power for genetic susceptibility gene discovery.
• This approach advances the field of genetic analysis for complex diseases.