A multi-filter enhanced genetic ensemble system for gene selection and sample classification of microarray data

Background: Feature selection techniques are critical to the analysis of high dimensional datasets. This is especially true in gene selection from microarray data which are commonly with extremely high feature-to-sample ratio. In addition to the essential objectives such as to reduce data noise, to reduce data redundancy, to improve sample classification accuracy, and to improve model generalization property, feature selection also helps biologists to focus on the selected genes to further validate their biological hypotheses.
Results: In this paper we describe an improved hybrid system for gene selection. It is based on a recently proposed genetic ensemble (GE) system. To enhance the generalization property of the selected genes or gene subsets and to overcome the overfitting problem of the GE system, we devised a mapping strategy to fuse the goodness information of each gene provided by multiple filtering algorithms. This information is then used for initialization and mutation operation of the genetic ensemble system.
Conclusion: We used four benchmark microarray datasets (including both binary-class and multi-class classification problems) for concept proving and model evaluation. The experimental results indicate that the proposed multi-filter enhanced genetic ensemble (MF-GE) system is able to improve sample classification accuracy, generate more compact gene subset, and converge to the selection results more quickly. The MF-GE system is very flexible as various combinations of multiple filters and classifiers can be incorporated based on the data characteristics and the user preferences.

Simplifying gene expression microarray comparative analysis.

Gene Expression Comparative Analysis allows bioinformatics researchers to discover the conserved or specific functional regulation of genes. This is achieved through comparisons between quantitative gene expression measurements obtained in different species on different platforms to address a particular biological system. Comparisons are made more difficult due to the need to map orthologous genes between species, pre-processing of data (normalization) and post-analysis (statistical and correlation analysis). In this paper we introduce a web-based software package called EXP-PAC which provides on line interfaces for database construction and query of data, and makes use of a high performance computing platform of computer clusters to run gene sequence mapping and normalization methods in parallel. Thus, EXP-PAC facilitates the integration of gene expression data for comparative analysis and the online sharing, retrieval and visualization of complex multi-specific and multi-platform gene expression results.

A novel aggregate gene selection method for microarray data classification

This paper introduces a novel method for gene selection based on a modification of analytic hierarchy process (AHP). The modified AHP (MAHP) is able to deal with quantitative factors that are statistics of five individual gene ranking methods: two-sample t-test, entropy test, receiver operating characteristic curve, Wilcoxon test, and signal to noise ratio. The most prominent discriminant genes serve as inputs to a range of classifiers including linear discriminant analysis, k-nearest neighbors, probabilistic neural network, support vector machine, and multilayer perceptron. Gene subsets selected by MAHP are compared with those of four competing approaches: information gain, symmetrical uncertainty, Bhattacharyya distance and ReliefF. Four benchmark microarray datasets: diffuse large B-cell lymphoma, leukemia cancer, prostate and colon are utilized for experiments. As the number of samples in microarray data datasets are limited, the leave one out cross validation strategy is applied rather than the traditional cross validation. Experimental results demonstrate the significant dominance of the proposed MAHP against the competing methods in terms of both accuracy and stability. With a benefit of inexpensive computational cost, MAHP is useful for cancer diagnosis using DNA gene expression profiles in the real clinical practice.

Hierarchical gene selection and genetic fuzzy system for cancer microarray data classification

This paper introduces a novel approach to gene selection based on a substantial modification of analytic hierarchy process (AHP). The modified AHP systematically integrates outcomes of individual filter methods to select the most informative genes for microarray classification. Five individual ranking methods including t-test, entropy, receiver operating characteristic (ROC) curve, Wilcoxon and signal to noise ratio are employed to rank genes. These ranked genes are then considered as inputs for the modified AHP. Additionally, a method that uses fuzzy standard additive model (FSAM) for cancer classification based on genes selected by AHP is also proposed in this paper. Traditional FSAM learning is a hybrid process comprising unsupervised structure learning and supervised parameter tuning. Genetic algorithm (GA) is incorporated in-between unsupervised and supervised training to optimize the number of fuzzy rules. The integration of GA enables FSAM to deal with the high-dimensional-low-sample nature of microarray data and thus enhance the efficiency of the classification. Experiments are carried out on numerous microarray datasets. Results demonstrate the performance dominance of the AHP-based gene selection against the single ranking methods. Furthermore, the combination of AHP-FSAM shows a great accuracy in microarray data classification compared to various competing classifiers. The proposed approach therefore is useful for medical practitioners and clinicians as a decision support system that can be implemented in the real medical practice.

Comparative microarray analysis identifies commonalities in neuronal injury: evidence for oxidative stress, dysfunction of calcium signalling, and inhibition of autophagy-lysosomal pathway

Mitochondrial dysfunction, ubiquitin-proteasomal system impairment and excitotoxicity occur during the injury and death of neurons in neurodegenerative conditions. The aim of this work was to elucidate the cellular mechanisms that are universally altered by these conditions. Through overlapping expression profiles of rotenone-, lactacystin- and N-methyl-D-aspartate-treated cortical neurons, we have identified three affected biological processes that are commonly affected; oxidative stress, dysfunction of calcium signalling and inhibition of the autophagic-lysosomal pathway. These data provides many opportunities for therapeutic intervention in neurodegenerative conditions, where mitochondrial dysfunction, proteasomal inhibition and excitotoxicity are evident.

Loss of samples in the tissue microarray technique: comparison between slides using adhesive tape and silanized slides

The tissue microarray (TMA) technique allows multiple tissue samples in a single block. Commercial adhesive tape is used to avoid the loss of tissue samples during the immunostaining process. Few reports exist in the literature comparing the use of these adhesive tapes to other adhesive techniques. The objective of this study was to compare loss of sections adhered to slides using commercial adhesive tapes versus using silanized only slides. TMA was constructed with varying tissues using a fixed-base device (Beecher Instruments), placing 108 cylinders of 1 mm diameter in duplicate, spaced 1.2 mm apart. Section of 4 mu m were cut from the TMA block and adhered to 30 silanized slides and 30 commercial glass slides using adhesive tape, according to manufacturer's recommendations. Vimentin immunoexpression was evaluated by immunohistochemistry. Antigenic recovery was realized in citrate buffer using a microwave oven. Cylinder loss in the immunohistochemical process was quantified and expressed as: total (>80%), almost complete (75-79%), or partial (50-74%). The commercial adhesive tape group presented lesser total loss (1.1 versus 6.4%), almost complete loss (2.2 versus 3.5%), and partial loss (2.1 versus 3.8%) than the silanized slide group (ANOVA, P < 0.05). The sum of total and almost complete losses in the silanized slide group was 9.9%, greater than the losses in slides using commercial adhesive tapes (3.3%) and less than reported and considered acceptable in the literature (10-30%). In conclusion, the use of silanized only slides presents very satisfactory results, requires less training, and reduces costs significantly, thus justifying their use in research.

Genomic DNA microarray comparison of gene expression patterns in Paracoccidioides brasiliensis mycelia and yeasts in vitro

Paracoccidioides brasiliensis is a thermally dimorphic fungus, and causes the most prevalent systemic mycosis in Latin America. Infection is initiated by inhalation of conidia or mycelial fragments by the host, followed by further differentiation into the yeast form. Information regarding gene expression by either form has rarely been addressed with respect to multiple time points of growth in culture. Here, we report on the construction of a genomic DNA microarray, covering approximately 25% of the genome of the organism, and its utilization in identifying genes and gene expression patterns during growth in vitro. Cloned, amplified inserts from randomly sheared genomic DNA (gDNA) and known control genes were printed onto glass slides to generate a microarray of over 12 000 elements. To examine gene expression, mRNA was extracted and amplified from mycelial or yeast cultures grown in semi-defined medium for 5, 8 and 14 days. Principal components analysis and hierarchical clustering indicated that yeast gene expression profiles differed greatly from those of mycelia, especially at earlier time points, and that mycelial gene expression changed less than gene expression in yeasts over time. Genes upregulated in yeasts were found to encode proteins shown to be involved in methionine/cysteine metabolism, respiratory and metabolic processes (of sugars, amino acids, proteins and lipids), transporters (small peptides, sugars, ions and toxins), regulatory proteins and transcription factors. Mycelial genes involved in processes such as cell division, protein catabolism, nucleotide biosynthesis and toxin and sugar transport showed differential expression. Sequenced clones were compared with Histoplasma capsulatum and Coccidioides posadasii genome sequences to assess potentially common pathways across species, such as sulfur and lipid metabolism, amino acid transporters, transcription factors and genes possibly related to virulence. We also analysed gene expression with time in culture and found that while transposable elements and components of respiratory pathways tended to increase in expression with time, genes encoding ribosomal structural proteins and protein catabolism tended to sharply decrease in expression over time, particularly in yeast. These findings expand our knowledge of the different morphological forms of P. brasiliensis during growth in culture.

Microarray analysis of cytochrome P450 mediated insecticide resistance in Drosophila

Insecticide resistance in laboratory selected Drosophila strains has been associated with upregulation of a range of different cytochrome P450s, however in recent field isolates of D. melanogaster resistance to DDT and other compounds is conferred by one P450 gene, Cyp6g1. Using microarray analysis of all Drosophila P450 genes, here we show that different P450 genes such as Cyp12d1 and Cyp6a8 can also be selected using DDT in the laboratory. We also show, however, that a homolog of Cyp6g1 is over-expressed in a field resistant strain of D. simulans. In order to determine why Cyp6g1 is so widely selected in the field we examine the pattern of cross-resistance of both resistant strains and transgenic flies over-expressing Cyp6g1 alone. We show that all three DDT selected P450s can confer resistance to the neonicotinoid imidacloprid but that Cyp6a8 confers no cross-resistance to malathion. Transgenic flies over-expressing Cyp6g1 also show cross-resistance to other neonicotinoids such as acetamiprid and nitenpyram. We suggest that the broad level of cross-resistance shown by Cyp6g1 may have facilitated its selection as a resistance gene in natural Drosophila populations. (C) 2003 Elsevier B.V. Ltd. All rights reserved.

Production of DNA microarray and expression analysis of genes from Xylella fastidiosa in different culture media

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)