Gene Selection and Classification in Microarray Datasets using a Hybrid Approach of PCC-BPSO/GA with Multi Classifiers

Abstract

In this study, a three-phase hybrid approach is proposed for the selection and classification of high dimensional microarray data. The method uses Pearson's Correlation Coefficient (PCC) in combination with Binary Particle Swarm Optimization (BPSO) or Genetic Algorithm (GA) along with various classifiers, thereby forming a PCC-BPSO/GA-multi classifiers approach. As such, five various classifiers are employed in the final stage of the classification. It was noticed that the PCC filter showed a remarkable improvement in the classification accuracy when it was combined with BPSO or GA. This positive impact was seen to be varied for different datasets based on the final applied classifier. The performance of various combination of the hybrid technique was compared in terms of accuracy and number of selected genes. In addition to the fact that BPSO is working faster than GA, it was noticed that BPSO has better performance than GA when it is combined with PCC feature selection.

HDG-select: A novel GUI based application for gene selection and classification in high dimensional datasets

“…Furthermore, a comparison of the results obtained by the HDG-select application to those reported in literature suggesting that the filter-GBPSO-SVM algorithm is more efficient than PCC-BPSO-SVM and PCC-GA-SVM algorithm [ 11 ] when it comes to the selection and classification of attributed genes in high dimensional datasets, as shown in Fig 8 and S3 Table .…”

“…The characteristics of the datasets are given in Tables 1 and 2 . The first six datasets in Table 1 are in csv format, which were pre-processed and previously used for gene expression analysis [ 11 ]. The Leukemia cancer dataset was achieved from [ 44 ].…”

“…This is because each filter is based on different assumptions related to the mean, median and variance which can be found in the high dimensional datasets. Because the filtration power is different for each filter, the combination of them might yield a better selection performance [ 11 , 51 , 52 ]. The TT filter was applied to microarray genes [ 53 ] and it was seen that the filter shows a strong scalability when the number of genes is high [ 54 ].…”

“…Many computational methods are failed to extract a small subset of attributed genes in high dimensional datasets because of the presence of various correlations and redundancy among the genes. Interestingly, studies in the field of cancer informatics have shown a splendid contribution of data mining and machine learning to find the attributed genes [ 8 – 11 ]. It is however proved that machine learning can perform well in cancer classification, it is yet required further improvement and robustness in terms of efficiency and computational cost, especially when high dimensional datasets are investigated.…”

Gene Microarray Cancer Classification using Correlation Based Feature Selection Algorithm and Rules Classifiers

“…The Pearson's Correlation Coefficient (PCC) [18] is used in combination with Binary Particle Swarm Optimization (BPSO) or Genetic Algorithm (GA) along with various classifiers for the selection and classification of high dimensional microarray data. It was noticed that the PCC filter showed a remarkable improvement in the classification accuracy when it was combined with BPSO or GA. Also, the results show that the BPSO is working faster and has better performance than GA.…”