Dimensionality Reduction in Gene Expression Data Sets

Abstract

Dimensionality reduction is used in microarray data analysis to enhance prediction quality, reduce computing time, and construct more robust models. In addition, the algorithm learning performance involves an expressive number of attributes (genes) relative to the classes (samples). Therefore, in this paper, we conducted a detailed comparison of two reduction methods, attribute selection and principal component analysis, to analyze gene expression data sets. Both reduction methods were employed in the pre-processing stage and then evaluated experimentally. Furthermore, we introduced a combination of consistency-based subset evaluation (CSE) and minimum redundancy maximum relevance (mRMR), which we referred to as CSE-mRMR, to improve classification efficiency. The results indicated a significant increase in classifier hit rates with both methods, compared to using all attributes. By employing cross-validation, attribute selection outperformed PCA consistently across classifiers and datasets, and CSE-mRMR demonstrated good classification performance in the data sets. Taken together, the literature and current results suggest that the attribute selection may be relevant in the analysis and future prediction of gene expression data sets.

Cancer Classification from Gene Expression Using Ensemble Learning with an Influential Feature Selection Technique

“…Akhand et al [12] used minimum Redundancy Maximum Relevance (mRMR) as a feature selection technique and then employed ANN on four benchmark datasets for cancer classification. Souza et al [5] attempted to compare and contrast two reduction methods-attribute selection and principal component analysis-to provide the most comprehensive comparison while analyzing gene expression datasets.…”

“…Therefore, the gain of information was employed to reduce the number of irrelevant genes and eliminate the high dimensionality problem. Dimensionality reduction is also applicable for reducing computing time, constructing a robust model, and increasing the model's prediction quality [5].…”

Reduksi Dimensi pada Klasifikasi Data Microarray Menggunakan Minimum Redundancy Maximum Relevance dan Random Forest

“…[3] MRMR dirancang untuk menganalisis kualitas dan memberikan kinerja prediktif terbaik pada subset variabel data (atribut kelas). [5] Tujuan utama dari MRMR yaitu untuk memilih fitur terbaik dan mengurangi fitur yang sama. Metode ini menangani setiap fitur secara terpisah dari kumpulan beberapa data dan menggunakan informasi timbal balik di antara fitur-fitur yang ada, dimana dapat juga digunakan untuk mengukur tingkat kesamaan antara dua fitur.…”

Performance Analysis of Dimensionality Reduction Techniques in Cancer Detection using Microarray Data

“…Combination of consistency based subset evaluation and minimum redundancy maximum evaluation methods gives good classification performance accuracy. Here using PCA gives better accuracy over attribute selection method [10]. PCA is LMBP) algorithm and it is concluded that SVM gives 94.98% accuracy & LMBP gives 96.07% accuracy [11].…”