Particle Swarm Optimization in Gene Expression Spectrum Clustering

The traditional clustering method of gene expression profile is affected by the number of iterations, its gene sequence marker value is in a negative range for a long time, and its clustering ability is poor. Therefore, the particle swarm optimization application analysis in gene expression profile clustering is proposed. Through DNA microarray experiment, gene expression spectrum data was obtained; it unified the expression value order of the data, optimized the particle swarm by improving the inertia weight and learning factor, extracted the data characteristics of gene expression spectrum, updated the clustering center by using particle code, and realized gene expression spectrum clustering. The experimental results show that, compared with the traditional gene spectrum clustering method, in the application of particle swarm optimization algorithm, with the increase of the number of iterations, the gene sequence marker value is always in a positive range, and the clustering ability is better.