Objective: To classify causal associations among drugs and adverse events by the identified drug safety signals from the US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS).Material and Methods: FAERS reports were collected for a period between 2004 and 2014. Empirical Bayes Geometric Mean was applied to model associations between drugs and adverse events. Identified signals were evaluated using Reporting Ratio values and Chi-Square test. Based on the identified drug-associated adverse events, we constructed a drug-drug association network, and applied a random walk algorithm to find drug communities with similar adverse event patterns. We developed 14 clusters for comparison with the 14 main groups in the first level of the Anatomical Therapeutic Chemical (ATC) classification system to evaluate relationships between the two classification systems.Results: The retrieved FAERS dataset included 981 drugs and 16,179 adverse events, from which we identified 63,083 significant drug-adverse event pairs. We found new potential safety signals when comparing the drug-adverse event pairs with information from relevant sources. Network analysis of the constructed drug communities revealed connections among drugs, adverse events, and ATC codes, suggesting that drug adverse events may be used to predict the ATC codes of unclassified drugs. For the ATC-classified drugs, the network analysis revealed potential relationships among the drugs by calculating the similarities of the adverse events.Conclusion: The generated drug groups and the drug-drug associations derived from the network analysis might have predictive potential for adverse events, as well as provide information for drug review and development.

BACKGROUND: Next-generation sequencing (NGS) technologies have provided researchers with vast possibilities in various biological and biomedical research areas. Efficient data mining strategies are in high demand for large scale comparative and evolutional studies to be performed on the large amounts of data derived from NGS projects. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining. METHODS: We report a novel procedure to analyse NGS data using topic modeling. It consists of four major procedures: NGS data retrieval, preprocessing, topic modeling, and data mining using Latent Dirichlet Allocation (LDA) topic outputs. The NGS data set of the Salmonella enterica strains were used as a case study to show the workflow of this procedure. The perplexity measurement of the topic numbers and the convergence efficiencies of Gibbs sampling were calculated and discussed for achieving the best result from the proposed procedure. RESULTS: The output topics by LDA algorithms could be treated as features of Salmonella strains to accurately describe the genetic diversity of fliC gene in various serotypes. The results of a two-way hierarchical clustering and data matrix analysis on LDA-derived matrices successfully classified Salmonella serotypes based on the NGS data. The implementation of topic modeling in NGS data analysis procedure provides a new way to elucidate genetic information from NGS data, and identify the gene-phenotype relationships and biomarkers, especially in the era of biological and medical big data. CONCLUSION: The implementation of topic modeling in NGS data analysis provides a new way to elucidate genetic information from NGS data, and identify the gene-phenotype relationships and biomarkers, especially in the era of biological and medical big data.

After publication of the original article [1] it was brought to our attention that the following was incorrectly placed under subheading '3.Classification analysis and comparison' of subsection 'Evaluation of topic modeling performance' of the 'Methods' section:Topic model-derived clustering method [33] was applied, in which LDA was utilized as a feature reduction approach for cluster analysis.The LDAderived topics were considered as the new features of datasets.The sampletopic matrix (Fig. 1(f)) was treated as a new representation of the original dataset.Based on the sample-topic matrix (topic number was chosen as 5 and 30, respectively), conventional clustering algorithms, such as k-means, was used for the clustering analysis.The number of clusters was set as 7 in the k-means method due to 7 different serotypes in the dataset.While in comparison, k-means algorithm was also applied on VSM matrix using Hamming Distance similarities.For further comparison, due to the dimension reduction of topic modeling approach, the traditional tool of PCA was used to reduce features (Numbers of 2, 5, 10 and 30 were randomly selected as the reduced features, respectively) of VSM matrix followed by the k-means cluster analysis.Moreover, clustering by only LDA referred as "highest probable topic assignment" [33] (5 and 30 topics were used) was also used for comparison.In "highest probable topic assignment", the LDA-derived topics were made as the clusters of the dataset.Then, each sample was assigned to the cluster (Topic) with the highest probability in the row of the sample-topic matrix.To interpret the clustering results obtained by the k-means algorithm, samples in each cluster were labeled as the dominant serotype of the samples in the cluster.

OBJECTIVES: This article summarizes past and current data mining activities at the United States Food and Drug Administration (FDA). TARGET AUDIENCE: We address data miners in all sectors, anyone interested in the safety of products regulated by the FDA (predominantly medical products, food, veterinary products and nutrition, and tobacco products), and those interested in FDA activities. SCOPE: Topics include routine and developmental data mining activities, short descriptions of mined FDA data, advantages and challenges of data mining at the FDA, and future directions of data mining at the FDA.

BACKGROUND: Topic modelling is an active research field in machine learning. While mainly used to build models from unstructured textual data, it offers an effective means of data mining where samples represent documents, and different biological endpoints or omics data represent words. Latent Dirichlet Allocation (LDA) is the most commonly used topic modelling method across a wide number of technical fields. However, model development can be arduous and tedious, and requires burdensome and systematic sensitivity studies in order to find the best set of model parameters. Often, time-consuming subjective evaluations are needed to compare models. Currently, research has yielded no easy way to choose the proper number of topics in a model beyond a major iterative approach. METHODS AND RESULTS: Based on analysis of variation of statistical perplexity during topic modelling, a heuristic approach is proposed in this study to estimate the most appropriate number of topics. Specifically, the rate of perplexity change (RPC) as a function of numbers of topics is proposed as a suitable selector. We test the stability and effectiveness of the proposed method for three markedly different types of grounded-truth datasets: Salmonella next generation sequencing, pharmacological side effects, and textual abstracts on computational biology and bioinformatics (TCBB) from PubMed. CONCLUSION: The proposed RPC-based method is demonstrated to choose the best number of topics in three numerical experiments of widely different data types, and for databases of very different sizes. The work required was markedly less arduous than if full systematic sensitivity studies had been carried out with number of topics as a parameter. We understand that additional investigation is needed to substantiate the method's theoretical basis, and to establish its generalizability in terms of dataset characteristics.

BACKGROUND: The big data moniker is nowhere better deserved than to describe the ever-increasing prodigiousness and complexity of biological and medical datasets. New methods are needed to generate and test hypotheses, foster biological interpretation, and build validated predictors. Although multivariate techniques such as cluster analysis may allow researchers to identify groups, or clusters, of related variables, the accuracies and effectiveness of traditional clustering methods diminish for large and hyper dimensional datasets. Topic modeling is an active research field in machine learning and has been mainly used as an analytical tool to structure large textual corpora for data mining. Its ability to reduce high dimensionality to a small number of latent variables makes it suitable as a means for clustering or overcoming clustering difficulties in large biological and medical datasets. RESULTS: In this study, three topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, are proposed and tested on the cluster analysis of three large datasets: Salmonella pulsed-field gel electrophoresis (PFGE) dataset, lung cancer dataset, and breast cancer dataset, which represent various types of large biological or medical datasets. All three various methods are shown to improve the efficacy/effectiveness of clustering results on the three datasets in comparison to traditional methods. A preferable cluster analysis method emerged for each of the three datasets on the basis of replicating known biological truths. CONCLUSION: Topic modeling could be advantageously applied to the large datasets of biological or medical research. The three proposed topic model-derived clustering methods, highest probable topic assignment, feature selection and feature extraction, yield clustering improvements for the three different data types. Clusters more efficaciously represent truthful groupings and subgroupings in the data than traditional methods, suggesting that topic model-based methods could provide an analytic advancement in the analysis of large biological or medical datasets.

A class-imbalanced classifier is a decision rule to predict the class membership of new samples from an available data set where the class sizes differ considerably. When the class sizes are very different, most standard classification algorithms may favor the larger (majority) class resulting in poor accuracy in the minority class prediction. A class-imbalanced classifier typically modifies a standard classifier by a correction strategy or by incorporating a new strategy in the training phase to account for differential class sizes. This article reviews and evaluates some most important methods for class prediction of high-dimensional imbalanced data. The evaluation addresses the fundamental issues of the class-imbalanced classification problem: imbalance ratio, small disjuncts and overlap complexity, lack of data and feature selection. Four class-imbalanced classifiers are considered. The four classifiers include three standard classification algorithms each coupled with an ensemble correction strategy and one support vector machines (SVM)-based correction classifier. The three algorithms are (i) diagonal linear discriminant analysis (DLDA), (ii) random forests (RFs) and (ii) SVMs. The SVM-based correction classifier is SVM threshold adjustment (SVM-THR). A Monte-Carlo simulation and five genomic data sets were used to illustrate the analysis and address the issues. The SVM-ensemble classifier appears to perform the best when the class imbalance is not too severe. The SVM-THR performs well if the imbalance is severe and predictors are highly correlated. The DLDA with a feature selection can perform well without using the ensemble correction.

BACKGROUND: Lung cancer is the leading cause of cancer-related death worldwide. Tremendous research efforts have been devoted to improving treatment procedures, but the average five-year overall survival rates are still less than 20%. Many biomarkers have been identified for predicting survival; challenges arise, however, in translating the findings into clinical practice due to their inconsistency and irreproducibility. In this study, we proposed an approach by identifying predictive genes through pathways. RESULTS: The microarrays from Shedden et al. were used as the training set, and the log-rank test was performed to select potential signature genes. We focused on 24 cancer-related pathways from 4 biological databases. A scoring scheme was developed by the Cox hazard regression model, and patients were divided into two groups based on the medians. Subsequently, their predictability and generalizability were evaluated by the 2-fold cross-validation and a resampling test in 4 independent datasets, respectively. A set of 16 genes related to apoptosis execution was demonstrated to have good predictability as well as generalizability in more than 700 lung adenocarcinoma patients and was reproducible in 4 independent datasets. This signature set was shown to have superior performances compared to 6 other published signatures. Furthermore, the corresponding risk scores derived from the set were found to associate with the efficacy of the anti-cancer drug ZD-6474 targeting EGFR. CONCLUSIONS: In summary, we presented a new approach to identify reproducible survival predictors for lung adenocarcinoma, and the identified genes may serve as both prognostic and predictive biomarkers in the future.

BACKGROUND: Pulsed field gel electrophoresis (PFGE) is currently the most widely and routinely used method by the Centers for Disease Control and Prevention (CDC) and state health labs in the United States for Salmonella surveillance and outbreak tracking. Major drawbacks of commercially available PFGE analysis programs have been their difficulty in dealing with large datasets and the limited availability of analysis tools. There exists a need to develop new analytical tools for PFGE data mining in order to make full use of valuable data in large surveillance databases. RESULTS: In this study, a software package was developed consisting of five types of bioinformatics approaches exploring and implementing for the analysis and visualization of PFGE fingerprinting. The approaches include PFGE band standardization, Salmonella serotype prediction, hierarchical cluster analysis, distance matrix analysis and two-way hierarchical cluster analysis. PFGE band standardization makes it possible for cross-group large dataset analysis. The Salmonella serotype prediction approach allows users to predict serotypes of Salmonella isolates based on their PFGE patterns. The hierarchical cluster analysis approach could be used to clarify subtypes and phylogenetic relationships among groups of PFGE patterns. The distance matrix and two-way hierarchical cluster analysis tools allow users to directly visualize the similarities/dissimilarities of any two individual patterns and the inter- and intra-serotype relationships of two or more serotypes, and provide a summary of the overall relationships between user-selected serotypes as well as the distinguishable band markers of these serotypes. The functionalities of these tools were illustrated on PFGE fingerprinting data from PulseNet of CDC. CONCLUSIONS: The bioinformatics approaches included in the software package developed in this study were integrated with the PFGE database to enhance the data mining of PFGE fingerprints. Fast and accurate prediction makes it possible to elucidate Salmonella serotype information before conventional serological methods are pursued. The development of bioinformatics tools to distinguish the PFGE markers and serotype specific patterns will enhance PFGE data retrieval, interpretation and serotype identification and will likely accelerate source tracking to identify the Salmonella isolates implicated in foodborne diseases.

BACKGROUND: Advances in microbial genomics and bioinformatics are offering greater insights into the emergence and spread of foodborne pathogens in outbreak scenarios. The Food and Drug Administration (FDA) has developed a genomics tool, ArrayTrack™, which provides extensive functionalities to manage, analyze, and interpret genomic data for mammalian species. ArrayTrack™ has been widely adopted by the research community and used for pharmacogenomics data review in the FDA's Voluntary Genomics Data Submission program. RESULTS: ArrayTrack™ has been extended to manage and analyze genomics data from bacterial pathogens of human, animal, and food origin. It was populated with bioinformatics data from public databases such as NCBI, Swiss-Prot, KEGG Pathway, and Gene Ontology to facilitate pathogen detection and characterization. ArrayTrack™'s data processing and visualization tools were enhanced with analysis capabilities designed specifically for microbial genomics including flag-based hierarchical clustering analysis (HCA), flag concordance heat maps, and mixed scatter plots. These specific functionalities were evaluated on data generated from a custom Affymetrix array (FDA-ECSG) previously developed within the FDA. The FDA-ECSG array represents 32 complete genomes of Escherichia coli and Shigella. The new functions were also used to analyze microarray data focusing on antimicrobial resistance genes from Salmonella isolates in a poultry production environment using a universal antimicrobial resistance microarray developed by the United States Department of Agriculture (USDA). CONCLUSION: The application of ArrayTrack™ to different microarray platforms demonstrates its utility in microbial genomics research, and thus will improve the capabilities of the FDA to rapidly identify foodborne bacteria and their genetic traits (e.g., antimicrobial resistance, virulence, etc.) during outbreak investigations. ArrayTrack™ is free to use and available to public, private, and academic researchers at http://www.fda.gov/ArrayTrack.

BACKGROUND: Before conducting a microarray experiment, one important issue that needs to be determined is the number of arrays required in order to have adequate power to identify differentially expressed genes. This paper discusses some crucial issues in the problem formulation, parameter specifications, and approaches that are commonly proposed for sample size estimation in microarray experiments. Common methods for sample size estimation are formulated as the minimum sample size necessary to achieve a specified sensitivity (proportion of detected truly differentially expressed genes) on average at a specified false discovery rate (FDR) level and specified expected proportion (pi1) of the true differentially expression genes in the array. Unfortunately, the probability of detecting the specified sensitivity in such a formulation can be low. We formulate the sample size problem as the number of arrays needed to achieve a specified sensitivity with 95% probability at the specified significance level. A permutation method using a small pilot dataset to estimate sample size is proposed. This method accounts for correlation and effect size heterogeneity among genes. RESULTS: A sample size estimate based on the common formulation, to achieve the desired sensitivity on average, can be calculated using a univariate method without taking the correlation among genes into consideration. This formulation of sample size problem is inadequate because the probability of detecting the specified sensitivity can be lower than 50%. On the other hand, the needed sample size calculated by the proposed permutation method will ensure detecting at least the desired sensitivity with 95% probability. The method is shown to perform well for a real example dataset using a small pilot dataset with 4-6 samples per group. CONCLUSIONS: We recommend that the sample size problem should be formulated to detect a specified proportion of differentially expressed genes with 95% probability. This formulation ensures finding the desired proportion of true positives with high probability. The proposed permutation method takes the correlation structure and effect size heterogeneity into consideration and works well using only a small pilot dataset.

Recent development of high-throughput technology has accelerated interest in the development of molecular biomarker classifiers for safety assessment, disease diagnostics and prognostics, and prediction of response for patient assignment. This article reviews and evaluates some important aspects and key issues in the development of biomarker classifiers. Development of a biomarker classifier for high-throughput data involves two components: (i) model building and (ii) performance assessment. This article focuses on feature selection in model building and cross validation for performance assessment. A 'frequency' approach to feature selection is presented and compared to the 'conventional' approach in terms of the predictive accuracy and stability of the selected feature set. The two approaches are compared based on four biomarker classifiers, each with a different feature selection method and well-known classification algorithm. In each of the four classifiers the feature predictor set selected by the frequency approach is more stable than the feature set selected by the conventional approach.

MOTIVATION: Gene class testing (GCT) or gene set analysis (GSA) is a statistical approach to determine whether some functionally predefined sets of genes express differently under different experimental conditions. Shortcomings of the Fisher's exact test for the overrepresentation analysis are illustrated by an example. Most alternative GSA methods are developed for data collected from two experimental conditions, and most is based on a univariate gene-by-gene test statistic or assume independence among genes in the gene set. A multivariate analysis of variance (MANOVA) approach is proposed for studies with two or more experimental conditions. RESULTS: When the number of genes in the gene set is greater than the number of samples, the sample covariance matrix is singular and ill-condition. The use of standard multivariate methods can result in biases in the analysis. The proposed MANOVA test uses a shrinkage covariance matrix estimator for the sample covariance matrix. The MANOVA test and six other GSA published methods, principal component analysis, SAM-GS, analysis of covariance, Global, GSEA and MaxMean, are evaluated using simulation. The MANOVA test appears to perform the best in terms of control of type I error and power under the models considered in the simulation. Several publicly available microarray datasets under two and three experimental conditions are analyzed for illustrations of GSA. Most methods, except for GSEA and MaxMean, generally are comparable in terms of power of identification of significant gene sets. AVAILABILITY: A free R-code to perform MANOVA test is available at http://mail.cmu.edu.tw/~catsai/research.htm. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Genome-wide association studies (GWAS) aim to identify genetic variants (usually single nucleotide polymorphisms [SNPs]) across the entire human genome that are associated with phenotypic traits such as disease status and drug response. Highly accurate and reproducible genotype calling are paramount since errors introduced by calling algorithms can lead to inflation of false associations between genotype and phenotype. Most genotype calling algorithms currently used for GWAS are based on multiple arrays. Because hundreds of gigabytes (GB) of raw data are generated from a GWAS, the samples are typically partitioned into batches containing subsets of the entire dataset for genotype calling. High call rates and accuracies have been achieved. However, the effects of batch size (i.e., number of chips analyzed together) and of batch composition (i.e., the choice of chips in a batch) on call rate and accuracy as well as the propagation of the effects into significantly associated SNPs identified have not been investigated. In this paper, we analyzed both the batch size and batch composition for effects on the genotype calling algorithm BRLMM using raw data of 270 HapMap samples analyzed with the Affymetrix Human Mapping 500 K array set. RESULTS: Using data from 270 HapMap samples interrogated with the Affymetrix Human Mapping 500 K array set, three different batch sizes and three different batch compositions were used for genotyping using the BRLMM algorithm. Comparative analysis of the calling results and the corresponding lists of significant SNPs identified through association analysis revealed that both batch size and composition affected genotype calling results and significantly associated SNPs. Batch size and batch composition effects were more severe on samples and SNPs with lower call rates than ones with higher call rates, and on heterozygous genotype calls compared to homozygous genotype calls. CONCLUSION: Batch size and composition affect the genotype calling results in GWAS using BRLMM. The larger the differences in batch sizes, the larger the effect. The more homogenous the samples in the batches, the more consistent the genotype calls. The inconsistency propagates to the lists of significantly associated SNPs identified in downstream association analysis. Thus, uniform and large batch sizes should be used to make genotype calls for GWAS. In addition, samples of high homogeneity should be placed into the same batch.

MOTIVATION: Gene class testing (GCT) is a statistical approach to determine whether some functionally predefined classes of genes express differently under two experimental conditions. GCT computes the P-value of each gene class based on the null distribution and the gene classes are ranked for importance in accordance with their P-values. Currently, two null hypotheses have been considered: the Q1 hypothesis tests the relative strength of association with the phenotypes among the gene classes, and the Q2 hypothesis assesses the statistical significance. These two hypotheses are related but not equivalent. METHOD: We investigate three one-sided and two two-sided test statistics under Q1 and Q2. The null distributions of gene classes under Q1 are generated by permuting gene labels and the null distributions under Q2 are generated by permuting samples. RESULTS: We applied the five statistics to a diabetes dataset with 143 gene classes and to a breast cancer dataset with 508 GO (Gene Ontology) terms. In each statistic, the null distributions of the gene classes under Q1 are different from those under Q2 in both datasets, and their rankings can be different too. We clarify the one-sided and two-sided hypotheses, and discuss some issues regarding the Q1 and Q2 hypotheses for gene class ranking in the GCT. Because Q1 does not deal with correlations among genes, we prefer test based on Q2. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Many researchers are concerned with the comparability and reliability of microarray gene expression data. Recent completion of the MicroArray Quality Control (MAQC) project provides a unique opportunity to assess reproducibility across multiple sites and the comparability across multiple platforms. The MAQC analysis presented for the conclusion of inter- and intra-platform comparability/reproducibility of microarray gene expression measurements is inadequate. We evaluate the reproducibility/comparability of the MAQC data for 12901 common genes in four titration samples generated from five high-density one-color microarray platforms and the TaqMan technology. We discuss some of the problems with the use of correlation coefficient as metric to evaluate the inter- and intra-platform reproducibility and the percent of overlapping genes (POG) as a measure for evaluation of a gene selection procedure by MAQC. RESULTS: A total of 293 arrays were used in the intra- and inter-platform analysis. A hierarchical cluster analysis shows distinct differences in the measured intensities among the five platforms. A number of genes show a small fold-change in one platform and a large fold-change in another platform, even though the correlations between platforms are high. An analysis of variance shows thirty percent of gene expressions of the samples show inconsistent patterns across the five platforms. We illustrated that POG does not reflect the accuracy of a selected gene list. A non-overlapping gene can be truly differentially expressed with a stringent cut, and an overlapping gene can be non-differentially expressed with non-stringent cutoff. In addition, POG is an unusable selection criterion. POG can increase or decrease irregularly as cutoff changes; there is no criterion to determine a cutoff so that POG is optimized. CONCLUSION: Using various statistical methods we demonstrate that there are differences in the intensities measured by different platforms and different sites within platform. Within each platform, the patterns of expression are generally consistent, but there is site-by-site variability. Evaluation of data analysis methods for use in regulatory decision should take no treatment effect into consideration, when there is no treatment effect, "a fold-change cutoff with a non-stringent p-value cutoff" could result in 100% false positive error selection.

BACKGROUND: A microarray study may select different differentially expressed gene sets because of different selection criteria. For example, the fold-change and p-value are two commonly known criteria to select differentially expressed genes under two experimental conditions. These two selection criteria often result in incompatible selected gene sets. Also, in a two-factor, say, treatment by time experiment, the investigator may be interested in one gene list that responds to both treatment and time effects. RESULTS: We propose three layer ranking algorithms, point-admissible, line-admissible (convex), and Pareto, to provide a preference gene list from multiple gene lists generated by different ranking criteria. Using the public colon data as an example, the layer ranking algorithms are applied to the three univariate ranking criteria, fold-change, p-value, and frequency of selections by the SVM-RFE classifier. A simulation experiment shows that for experiments with small or moderate sample sizes (less than 20 per group) and detecting a 4-fold change or less, the two-dimensional (p-value and fold-change) convex layer ranking selects differentially expressed genes with generally lower FDR and higher power than the standard p-value ranking. Three applications are presented. The first application illustrates a use of the layer rankings to potentially improve predictive accuracy. The second application illustrates an application to a two-factor experiment involving two dose levels and two time points. The layer rankings are applied to selecting differentially expressed genes relating to the dose and time effects. In the third application, the layer rankings are applied to a benchmark data set consisting of three dilution concentrations to provide a ranking system from a long list of differentially expressed genes generated from the three dilution concentrations. CONCLUSION: The layer ranking algorithms are useful to help investigators in selecting the most promising genes from multiple gene lists generated by different filter, normalization, or analysis methods for various objectives.

MOTIVATION: One major area of interest in analyzing oligonucleotide gene array data is identifying differentially expressed genes. A challenge to biostatisticians is to develop an approach to summarizing probe-level information that adequately reflects the true expression level while accounting for probe variation, chip variation and interaction effects. Various statistical tools, such as MAS and RMA, have been developed to address this issue. In these approaches, the probe level expression data are summarized into gene level data, which are then used for downstream statistical analysis. Since probe variation is often larger than chip variation and there is also a potential interaction effect between probe affinity and treatment effect, strategies such as a gene level analysis, may not be optimal. In this study, we propose a procedure to analyze probe level data for selecting differentially expressed genes under two treatment conditions (groups) with a small number of replicates. The probe level discrepancy between two groups can be measured by a difference of the percentiles of probe perfect-match (PM) ranks or of probe PM weighted ranks. The difference is then compared with a pre-specified threshold to determine differentially expressed genes. The probe level approach takes into account non-homogenous treatment effects and reduces possible cross-hybridization effects across a set of probes. RESULTS: The proposed approach is compared with MAS and RMA using two benchmark gene array datasets. Positive predictivity and sensitivity are used for evaluation. Results show the proposed approach has higher positive predictivity and higher sensitivity. AVAILABILITY: Available on request from the authors. CONTACT: [email protected].

MOTIVATION: Microarray experiments often involve hundreds or thousands of genes. In a typical experiment, only a fraction of genes are expected to be differentially expressed; in addition, the measured intensities among different genes may be correlated. Depending on the experimental objectives, sample size calculations can be based on one of the three specified measures: sensitivity, true discovery and accuracy rates. The sample size problem is formulated as: the number of arrays needed in order to achieve the desired fraction of the specified measure at the desired family-wise power at the given type I error and (standardized) effect size. RESULTS: We present a general approach for estimating sample size under independent and equally correlated models using binomial and beta-binomial models, respectively. The sample sizes needed for a two-sample z-test are computed; the computed theoretical numbers agree well with the Monte Carlo simulation results. But, under more general correlation structures, the beta-binomial model can underestimate the needed samples by about 1-5 arrays. CONTACT: [email protected].

BACKGROUND: The acceptance of microarray technology in regulatory decision-making is being challenged by the existence of various platforms and data analysis methods. A recent report (E. Marshall, Science, 306, 630-631, 2004), by extensively citing the study of Tan et al. (Nucleic Acids Res., 31, 5676-5684, 2003), portrays a disturbingly negative picture of the cross-platform comparability, and, hence, the reliability of microarray technology. RESULTS: We reanalyzed Tan's dataset and found that the intra-platform consistency was low, indicating a problem in experimental procedures from which the dataset was generated. Furthermore, by using three gene selection methods (i.e., p-value ranking, fold-change ranking, and Significance Analysis of Microarrays (SAM)) on the same dataset we found that p-value ranking (the method emphasized by Tan et al.) results in much lower cross-platform concordance compared to fold-change ranking or SAM. Therefore, the low cross-platform concordance reported in Tan's study appears to be mainly due to a combination of low intra-platform consistency and a poor choice of data analysis procedures, instead of inherent technical differences among different platforms, as suggested by Tan et al. and Marshall. CONCLUSION: Our results illustrate the importance of establishing calibrated RNA samples and reference datasets to objectively assess the performance of different microarray platforms and the proficiency of individual laboratories as well as the merits of various data analysis procedures. Thus, we are progressively coordinating the MAQC project, a community-wide effort for microarray quality control.

BACKGROUND: Microarray-based measurement of mRNA abundance assumes a linear relationship between the fluorescence intensity and the dye concentration. In reality, however, the calibration curve can be nonlinear. RESULTS: By scanning a microarray scanner calibration slide containing known concentrations of fluorescent dyes under 18 PMT gains, we were able to evaluate the differences in calibration characteristics of Cy5 and Cy3. First, the calibration curve for the same dye under the same PMT gain is nonlinear at both the high and low intensity ends. Second, the degree of nonlinearity of the calibration curve depends on the PMT gain. Third, the two PMTs (for Cy5 and Cy3) behave differently even under the same gain. Fourth, the background intensity for the Cy3 channel is higher than that for the Cy5 channel. The impact of such characteristics on the accuracy and reproducibility of measured mRNA abundance and the calculated ratios was demonstrated. Combined with simulation results, we provided explanations to the existence of ratio underestimation, intensity-dependence of ratio bias, and anti-correlation of ratios in dye-swap replicates. We further demonstrated that although Lowess normalization effectively eliminates the intensity-dependence of ratio bias, the systematic deviation from true ratios largely remained. A method of calculating ratios based on concentrations estimated from the calibration curves was proposed for correcting ratio bias. CONCLUSION: It is preferable to scan microarray slides at fixed, optimal gain settings under which the linearity between concentration and intensity is maximized. Although normalization methods improve reproducibility of microarray measurements, they appear less effective in improving accuracy.

MOTIVATION: A microarray experiment is a multi-step process, and each step is a potential source of variation. There are two major sources of variation: biological variation and technical variation. This study presents a variance-components approach to investigating animal-to-animal, between-array, within-array and day-to-day variations for two data sets. The first data set involved estimation of technical variances for pooled control and pooled treated RNA samples. The variance components included between-array, and two nested within-array variances: between-section (the upper- and lower-sections of the array are replicates) and within-section (two adjacent spots of the same gene are printed within each section). The second experiment was conducted on four different weeks. Each week there were reference and test samples with a dye-flip replicate in two hybridization days. The variance components included week-to-week, animal-to-animal and between-array and within-array variances. RESULTS: We applied the linear mixed-effects model to quantify different sources of variation. In the first data set, we found that the between-array variance is greater than the between-section variance, which, in turn, is greater than the within-section variance. In the second data set, for the reference samples, the week-to-week variance is larger than the between-array variance, which, in turn, is slightly larger than the within-array variance. For the test samples, the week-to-week variance has the largest variation. The animal-to-animal variance is slightly larger than the between-array and within-array variances. However, in a gene-by-gene analysis, the animal-to-animal variance is smaller than the between-array variance in four out of five housekeeping genes. In summary, the largest variation observed is the week-to-week effect. Another important source of variability is the animal-to-animal variation. Finally, we describe the use of variance-component estimates to determine optimal numbers of animals, arrays per animal and sections per array in planning microarray experiments.