Feature selection and classification for microarray data analysis: Evolutionary methods for identifying predictive genes

Background  

In the clinical context, samples assayed by microarray are often classified by cell line or tumour type and it is of interest to discover a set of genes that can be used as class predictors. The leukemia dataset of Golubet al.[1] and the NCI60 dataset of Rosset al.[2] present multiclass classification problems where three tumour types and nine cell lines respectively must be identified. We apply an evolutionary algorithm to identify the near-optimal set of predictive genes that classify the data. We also examine the initial gene selection step whereby the most informative genes are selected from the genes assayed.     

Identification of significant periodic genes in microarray gene expression data

Background  

One frequent application of microarray experiments is in the study of monitoring gene activities in a cell during cell cycle or cell division. A new challenge for analyzing the microarray experiments is to identify genes that are statistically significantly periodically expressed during the cell cycle. Such a challenge occurs due to the large number of genes that are simultaneously measured, a moderate to small number of measurements per gene taken at different time points, and high levels of non-normal random noises inherited in the data.     

Assessment of differential gene expression in human peripheral nerve injury

Background  

Microarray technology is a powerful methodology for identifying differentially expressed genes. However, when thousands of genes in a microarray data set are evaluated simultaneously by fold changes and significance tests, the probability of detecting false positives rises sharply. In this first microarray study of brachial plexus injury, we applied and compared the performance of two recently proposed algorithms for tackling this multiple testing problem, Significance Analysis of Microarrays (SAM) and Westfall and Young step down adjusted p values, as well as t-statistics and Welch statistics, in specifying differential gene expression under different biological States.     

Empirical Bayesian models for analysing molecular serotyping microarrays

Background  

Microarrays offer great potential as a platform for molecular diagnostics, testing clinical samples for the presence of numerous biomarkers in highly multiplexed assays. In this study applied to infectious diseases, data from a microarray designed for molecular serotyping of Streptococcus pneumoniae was used, identifying the presence of any one of 91 known pneumococcal serotypes from DNA extracts. This microarray incorporated oligonucleotide probes for all known capsular polysaccharide synthesis genes and required a statistical analysis of the microarray intensity data to determine which serotype, or combination of serotypes, were present within a sample based on the combination of genes detected.     

DFP: a Bioconductor package for fuzzy profile identification and gene reduction of microarray data

Background  

Expression profiling assays done by using DNA microarray technology generate enormous data sets that are not amenable to simple analysis. The greatest challenge in maximizing the use of this huge amount of data is to develop algorithms to interpret and interconnect results from different genes under different conditions. In this context, fuzzy logic can provide a systematic and unbiased way to both (i) find biologically significant insights relating to meaningful genes, thereby removing the need for expert knowledge in preliminary steps of microarray data analyses and (ii) reduce the cost and complexity of later applied machine learning techniques being able to achieve interpretable models.     

Phase analysis of circadian-related genes in two tissues

Background  

Recent circadian clock studies using gene expression microarray in two different tissues of mouse have revealed not all circadian-related genes are synchronized in phase or peak expression times across tissues in vivo. Instead, some circadian-related genes may be delayed by 4–8 hrs in peak expression in one tissue relative to the other. These interesting biological observations prompt a statistical question regarding how to distinguish the synchronized genes from genes that are systematically lagged in phase/peak expression time across two tissues.     

Information criterion-based clustering with order-restricted candidate profiles in short time-course microarray experiments

Background  

Time-course microarray experiments produce vector gene expression profiles across a series of time points. Clustering genes based on these profiles is important in discovering functional related and co-regulated genes. Early developed clustering algorithms do not take advantage of the ordering in a time-course study, explicit use of which should allow more sensitive detection of genes that display a consistent pattern over time. Peddada et al. [1] proposed a clustering algorithm that can incorporate the temporal ordering using order-restricted statistical inference. This algorithm is, however, very time-consuming and hence inapplicable to most microarray experiments that contain a large number of genes. Its computational burden also imposes difficulty to assess the clustering reliability, which is a very important measure when clustering noisy microarray data.     

Transcriptional profile of Pseudomonas syringae pv. phaseolicola NPS3121 in response to tissue extracts from a susceptible Phaseolus vulgaris L. cultivar

Background  

Pseudomonas syringae pv. phaseolicola is a Gram-negative plant-pathogenic bacterium that causes "halo blight" disease of beans (Phaseolus vulgaris L.). This disease affects both foliage and pods, and is a major problem in temperate areas of the world. Although several bacterial genes have been determined as participants in pathogenesis, the overall process still remains poorly understood, mainly because the identity and function of many of the genes are largely unknown. In this work, a genomic library of P. syringae pv. phaseolicola NPS3121 was constructed and PCR amplification of individual fragments was carried out in order to print a DNA microarray. This microarray was used to identify genes that are differentially expressed when bean leaf extracts, pod extracts or apoplastic fluid were added to the growth medium.     

Confirmation of human protein interaction data by human expression data

Background  

With microarray technology the expression of thousands of genes can be measured simultaneously. It is well known that the expression levels of genes of interacting proteins are correlated significantly more strongly in Saccharomyces cerevisiae than those of proteins that are not interacting. The objective of this work is to investigate whether this observation extends to the human genome.     

Stromal upregulation of lateral epithelial adhesions: Gene expression analysis of signalling pathways in prostate epithelium

Background  

Stromal signalling increases the lateral cell adhesions of prostate epithelial cells grown in 3D culture. The aim of this study was to use microarray analysis to identify significant epithelial signalling pathways and genes in this process.     

Rapid identification of bacterial pathogens using a PCR- and microarray-based assay

Background  

During the course of a bacterial infection, the rapid identification of the causative agent(s) is necessary for the determination of effective treatment options. We have developed a method based on a modified broad-range PCR and an oligonucleotide microarray for the simultaneous detection and identification of 12 bacterial pathogens at the species level. The broad-range PCR primer mixture was designed using conserved regions of the bacterial topoisomerase genes gyrB and parE. The primer design allowed the use of a novel DNA amplification method, which produced labeled, single-stranded DNA suitable for microarray hybridization. The probes on the microarray were designed from the alignments of species- or genus-specific variable regions of the gyrB and parE genes flanked by the primers. We included mecA-specific primers and probes in the same assay to indicate the presence of methicillin resistance in the bacterial species. The feasibility of this assay in routine diagnostic testing was evaluated using 146 blood culture positive and 40 blood culture negative samples.     

Comparison of evolutionary algorithms in gene regulatory network model inference

Background  

The evolution of high throughput technologies that measure gene expression levels has created a data base for inferring GRNs (a process also known as reverse engineering of GRNs). However, the nature of these data has made this process very difficult. At the moment, several methods of discovering qualitative causal relationships between genes with high accuracy from microarray data exist, but large scale quantitative analysis on real biological datasets cannot be performed, to date, as existing approaches are not suitable for real microarray data which are noisy and insufficient.     

Information propagation within the Genetic Network of Saccharomyces cerevisiae

Background  

A gene network's capacity to process information, so as to bind past events to future actions, depends on its structure and logic. From previous and new microarray measurements in Saccharomyces cerevisiae following gene deletions and overexpressions, we identify a core gene regulatory network (GRN) of functional interactions between 328 genes and the transfer functions of each gene. Inferred connections are verified by gene enrichment.     

A methodology for global validation of microarray experiments

Background  

DNA microarrays are popular tools for measuring gene expression of biological samples. This ever increasing popularity is ensuring that a large number of microarray studies are conducted, many of which with data publicly available for mining by other investigators. Under most circumstances, validation of differential expression of genes is performed on a gene to gene basis. Thus, it is not possible to generalize validation results to the remaining majority of non-validated genes or to evaluate the overall quality of these studies.     

Improving gene set analysis of microarray data by SAM-GS

Background  

Gene-set analysis evaluates the expression of biological pathways, or a priori defined gene sets, rather than that of individual genes, in association with a binary phenotype, and is of great biologic interest in many DNA microarray studies. Gene Set Enrichment Analysis (GSEA) has been applied widely as a tool for gene-set analyses. We describe here some critical problems with GSEA and propose an alternative method by extending the individual-gene analysis method, Significance Analysis of Microarray (SAM), to gene-set analyses (SAM-GS).