Search tips
Search criteria

Results 1-25 (1382442)

Clipboard (0)

Related Articles

1.  Exploring matrix factorization techniques for significant genes identification of Alzheimer’s disease microarray gene expression data 
BMC Bioinformatics  2011;12(Suppl 5):S7.
The wide use of high-throughput DNA microarray technology provide an increasingly detailed view of human transcriptome from hundreds to thousands of genes. Although biomedical researchers typically design microarray experiments to explore specific biological contexts, the relationships between genes are hard to identified because they are complex and noisy high-dimensional data and are often hindered by low statistical power. The main challenge now is to extract valuable biological information from the colossal amount of data to gain insight into biological processes and the mechanisms of human disease. To overcome the challenge requires mathematical and computational methods that are versatile enough to capture the underlying biological features and simple enough to be applied efficiently to large datasets.
Unsupervised machine learning approaches provide new and efficient analysis of gene expression profiles. In our study, two unsupervised knowledge-based matrix factorization methods, independent component analysis (ICA) and nonnegative matrix factorization (NMF) are integrated to identify significant genes and related pathways in microarray gene expression dataset of Alzheimer’s disease. The advantage of these two approaches is they can be performed as a biclustering method by which genes and conditions can be clustered simultaneously. Furthermore, they can group genes into different categories for identifying related diagnostic pathways and regulatory networks. The difference between these two method lies in ICA assume statistical independence of the expression modes, while NMF need positivity constrains to generate localized gene expression profiles.
In our work, we performed FastICA and non-smooth NMF methods on DNA microarray gene expression data of Alzheimer’s disease respectively. The simulation results shows that both of the methods can clearly classify severe AD samples from control samples, and the biological analysis of the identified significant genes and their related pathways demonstrated that these genes play a prominent role in AD and relate the activation patterns to AD phenotypes. It is validated that the combination of these two methods is efficient.
Unsupervised matrix factorization methods provide efficient tools to analyze high-throughput microarray dataset. According to the facts that different unsupervised approaches explore correlations in the high-dimensional data space and identify relevant subspace base on different hypotheses, integrating these methods to explore the underlying biological information from microarray dataset is an efficient approach. By combining the significant genes identified by both ICA and NMF, the biological analysis shows great efficient for elucidating the molecular taxonomy of Alzheimer’s disease and enable better experimental design to further identify potential pathways and therapeutic targets of AD.
PMCID: PMC3203370  PMID: 21989140
2.  Mining gene expression data by interpreting principal components 
BMC Bioinformatics  2006;7:194.
There are many methods for analyzing microarray data that group together genes having similar patterns of expression over all conditions tested. However, in many instances the biologically important goal is to identify relatively small sets of genes that share coherent expression across only some conditions, rather than all or most conditions as required in traditional clustering; e.g. genes that are highly up-regulated and/or down-regulated similarly across only a subset of conditions. Equally important is the need to learn which conditions are the decisive ones in forming such gene sets of interest, and how they relate to diverse conditional covariates, such as disease diagnosis or prognosis.
We present a method for automatically identifying such candidate sets of biologically relevant genes using a combination of principal components analysis and information theoretic metrics. To enable easy use of our methods, we have developed a data analysis package that facilitates visualization and subsequent data mining of the independent sources of significant variation present in gene microarray expression datasets (or in any other similarly structured high-dimensional dataset). We applied these tools to two public datasets, and highlight sets of genes most affected by specific subsets of conditions (e.g. tissues, treatments, samples, etc.). Statistically significant associations for highlighted gene sets were shown via global analysis for Gene Ontology term enrichment. Together with covariate associations, the tool provides a basis for building testable hypotheses about the biological or experimental causes of observed variation.
We provide an unsupervised data mining technique for diverse microarray expression datasets that is distinct from major methods now in routine use. In test uses, this method, based on publicly available gene annotations, appears to identify numerous sets of biologically relevant genes. It has proven especially valuable in instances where there are many diverse conditions (10's to hundreds of different tissues or cell types), a situation in which many clustering and ordering algorithms become problematic. This approach also shows promise in other topic domains such as multi-spectral imaging datasets.
PMCID: PMC1501050  PMID: 16600052
3.  Investigating the Efficacy of Nonlinear Dimensionality Reduction Schemes in Classifying Gene- and Protein-Expression Studies 
The recent explosion in procurement and availability of high-dimensional gene- and protein-expression profile datasets for cancer diagnostics has necessitated the development of sophisticated machine learning tools with which to analyze them. While some investigators are focused on identifying informative genes and proteins that play a role in specific diseases, other researchers have attempted instead to use patients based on their expression profiles to prognosticate disease status. A major limitation in the ability to accurate classify these high-dimensional datasets stems from the ‘curse of dimensionality’, occurring in situations where the number of genes or peptides significantly exceeds the total number of patient samples. Previous attempts at dealing with this issue have mostly centered on the use of a dimensionality reduction (DR) scheme, Principal Component Analysis (PCA), to obtain a low-dimensional projection of the high-dimensional data. However, linear PCA and other linear DR methods, which rely on Euclidean distances to estimate object similarity, do not account for the inherent underlying nonlinear structure associated with most biomedical data. While some researchers have begun to explore nonlinear DR methods for computer vision problems such as face detection and recognition, to the best of our knowledge, few such attempts have been made for classification and visualization of high-dimensional biomedical data. The motivation behind this work is to identify the appropriate DR methods for analysis of high-dimensional gene- and protein-expression studies. Towards this end, we empirically and rigorously compare three nonlinear (Isomap, Locally Linear Embedding, Laplacian Eigenmaps) and three linear DR schemes (PCA, Linear Discriminant Analysis, Multidimensional Scaling) with the intent of determining a reduced subspace representation in which the individual object classes are more easily discriminable. Owing to the to the inherent nonlinear structure of gene- and protein-expression studies, our claim is that the nonlinear DR methods provide a more truthful low-dimensional representation of the data compared to the linear DR schemes. Evaluation of the DR schemes was done by (i) assessing the discriminability of two supervised classifiers (Support Vector Machine and C4.5 Decision Trees) in the different low-dimensional data embeddings and (ii) 5 cluster validity measures to evaluate the size, distance and tightness of object aggregates in the low-dimensional space. For each of the 7 evaluation measures considered, statistically significant improvement in the quality of the embeddings across 10 cancer datasets via the use of 3 nonlinear DR schemes over 3 linear DR techniques was observed. Similar trends were observed when linear and nonlinear DR was applied to the high-dimensional data following feature pruning to isolate the most informative features. Qualitative evaluation of the low-dimensional data embedding obtained via the 6 DR methods further suggests that the nonlinear schemes are better able to identify potential novel classes (e.g. cancer subtypes) within the data.
PMCID: PMC2562675  PMID: 18670041
Dimensionality reduction; bioinformatics; data clustering; data visualization; machine learning; manifold learning; nonlinear dimensionality reduction; gene expression; proteomics; prostate cancer; lung cancer; ovarian cancer; principal component analysis; linear discriminant analysis; multidimensional scaling; Isomap; locally linear embedding; laplacian eigenmaps; classification; support vector machine; decision trees; LLE; PCA
4.  Systematic image-driven analysis of the spatial Drosophila embryonic expression landscape 
We created innovative virtual representation for our large scale Drosophila insitu expression dataset. We aligned an elliptically shaped mesh comprised of small triangular regions to the outline of each embryo. Each triangle defines a unique location in the embryo and comparing corresponding triangles allows easy identification of similar expression patterns.The virtual representation was used to organize the expression landscape at stage 4-6. We identified regions with similar expression in the embryo and clustered genes with similar expression patterns.We created algorithms to mine the dataset for adjacent non-overlapping patterns and anti-correlated patterns. We were able to mine the dataset to identify co-expressed and putative interacting genes.Using co-expression we were able to assign putative functions to unknown genes.
Analyzing both temporal and spatial gene expression is essential for understanding development and regulatory networks of multicellular organisms. Interacting genes are commonly expressed in overlapping or adjacent domains. Thus, gene expression patterns can be used to assign putative gene functions and mined to infer candidates for networks.
We have generated a systematic two-dimensional mRNA expression atlas profiling embryonic development of Drosophila melanogaster (Tomancak et al, 2002, 2007). To date, we have collected over 70 000 images for over 6000 genes. To explore spatial relationships between gene expression patterns, we used a novel computational image-processing approach by converting expression patterns from the images into virtual representations (Figure 1). Using a custom-designed automated pipeline, for each image, we segmented and aligned the outline of the embryo to an elliptically shaped mesh, comprised of 311 small triangular regions each defining a unique location within the embryo. By comparing corresponding triangles, we produced a distance score to identify similar patterns. We generated those triangulated images (TIs) for our entire data set at all developmental stages and demonstrated that this representation can be used as for objective computationally defined description for expression in in situ hybridization images from various sources, including images from the literature.
We used the TIs to conduct a comprehensive analysis of the expression landscape. To this end, we created a novel approach to temporally sort and compact TIs to a non-redundant data set suitable for further computational processing. Although generally applicable for all developmental stages, for this study, we focused on developmental stages 4–6. For this stage range, we reduced the initial set of about 5800 TIs to 553 TIs containing 364 genes. Using this filtered data set, to discover how expression subdivides the embryo into regions, we clustered areas with similar expression and demonstrated that expression patterns divide the early embryo into distinct spatial regions resembling a fate map (Figure 3). To discover the range of unique expression patterns, we used affinity propagation clustering (Frey and Dueck, 2007) to group TIs with similar patterns and identified 39 clusters each representing a distinct pattern class. We integrated the remaining genes into the 39 clusters and studied the distribution of expression patterns and the relationships between the clusters.
The clustered expression patterns were used to identify putative positive and negative regulatory interactions. The similar TIs in each cluster not only grouped already known genes with related functions, but previously undescribed genes. A comparative analysis identified subtle differences between the genes within each expression cluster. To investigate these differences, we developed a novel Markov Random Field (MRF) segmentation algorithm to extract patterns. We then extended the MRF algorithm to detect shared expression boundaries, generate similarity measurements, and discriminate even faint/uncertain patterns between two TIs. This enabled us to identify more subtle partial expression pattern overlaps and adjacent non-overlapping patterns. For example, by conducting this analysis on the cluster containing the gene snail, we identified the previously known huckebein, which restricts snail expression (Reuter and Leptin, 1994), and zfh1, which interacts with tinman (Broihier et al, 1998; Su et al, 1999).
By studying the functions of known genes, we assigned putative developmental roles to each of the 39 clusters. Of the 1800 genes investigated, only half of them had previously assigned functions.
Representing expression patterns with geometric meshes facilitates the analysis of a complex process involving thousands of genes. This approach is complementary to the cellular resolution 3D atlas for the Drosophila embryo (Fowlkes et al, 2008). Our method can be used as a rapid, fully automated, high-throughput approach to obtain a map of co-expression, which will serve to select specific genes for detailed multiplex in-situ hybridization and confocal analysis for a fine-grain atlas. Our data are similar to the data in the literature, and research groups studying reporter constructs, mutant animals, or orthologs can easily produce in situ hybridizations. TIs can be readily created and provide representations that are both comparable to each other and our data set. We have demonstrated that our approach can be used for predicting relationships in regulatory and developmental pathways.
Discovery of temporal and spatial patterns of gene expression is essential for understanding the regulatory networks and development in multicellular organisms. We analyzed the images from our large-scale spatial expression data set of early Drosophila embryonic development and present a comprehensive computational image analysis of the expression landscape. For this study, we created an innovative virtual representation of embryonic expression patterns using an elliptically shaped mesh grid that allows us to make quantitative comparisons of gene expression using a common frame of reference. Demonstrating the power of our approach, we used gene co-expression to identify distinct expression domains in the early embryo; the result is surprisingly similar to the fate map determined using laser ablation. We also used a clustering strategy to find genes with similar patterns and developed new analysis tools to detect variation within consensus patterns, adjacent non-overlapping patterns, and anti-correlated patterns. Of the 1800 genes investigated, only half had previously assigned functions. The known genes suggest developmental roles for the clusters, and identification of related patterns predicts requirements for co-occurring biological functions.
PMCID: PMC2824522  PMID: 20087342
biological function; embryo; gene expression; in situ hybridization; Markov Random Field
5.  Network-based Survival Analysis Reveals Subnetwork Signatures for Predicting Outcomes of Ovarian Cancer Treatment 
PLoS Computational Biology  2013;9(3):e1002975.
Cox regression is commonly used to predict the outcome by the time to an event of interest and in addition, identify relevant features for survival analysis in cancer genomics. Due to the high-dimensionality of high-throughput genomic data, existing Cox models trained on any particular dataset usually generalize poorly to other independent datasets. In this paper, we propose a network-based Cox regression model called Net-Cox and applied Net-Cox for a large-scale survival analysis across multiple ovarian cancer datasets. Net-Cox integrates gene network information into the Cox's proportional hazard model to explore the co-expression or functional relation among high-dimensional gene expression features in the gene network. Net-Cox was applied to analyze three independent gene expression datasets including the TCGA ovarian cancer dataset and two other public ovarian cancer datasets. Net-Cox with the network information from gene co-expression or functional relations identified highly consistent signature genes across the three datasets, and because of the better generalization across the datasets, Net-Cox also consistently improved the accuracy of survival prediction over the Cox models regularized by or . This study focused on analyzing the death and recurrence outcomes in the treatment of ovarian carcinoma to identify signature genes that can more reliably predict the events. The signature genes comprise dense protein-protein interaction subnetworks, enriched by extracellular matrix receptors and modulators or by nuclear signaling components downstream of extracellular signal-regulated kinases. In the laboratory validation of the signature genes, a tumor array experiment by protein staining on an independent patient cohort from Mayo Clinic showed that the protein expression of the signature gene FBN1 is a biomarker significantly associated with the early recurrence after 12 months of the treatment in the ovarian cancer patients who are initially sensitive to chemotherapy. Net-Cox toolbox is available at
Author Summary
Network-based computational models are attracting increasing attention in studying cancer genomics because molecular networks provide valuable information on the functional organizations of molecules in cells. Survival analysis mostly with the Cox proportional hazard model is widely used to predict or correlate gene expressions with time to an event of interest (outcome) in cancer genomics. Surprisingly, network-based survival analysis has not received enough attention. In this paper, we studied resistance to chemotherapy in ovarian cancer with a network-based Cox model, called Net-Cox. The experiments confirm that networks representing gene co-expression or functional relations can be used to improve the accuracy and the robustness of survival prediction of outcome in ovarian cancer treatment. The study also revealed subnetwork signatures that are enriched by extracellular matrix receptors and modulators and the downstream nuclear signaling components of extracellular signal-regulators, respectively. In particular, FBN1, which was detected as a signature gene of high confidence by Net-Cox with network information, was validated as a biomarker for predicting early recurrence in platinum-sensitive ovarian cancer patients in laboratory.
PMCID: PMC3605061  PMID: 23555212
6.  A modified hyperplane clustering algorithm allows for efficient and accurate clustering of extremely large datasets 
Bioinformatics  2009;25(9):1152-1157.
Motivation: As the number of publically available microarray experiments increases, the ability to analyze extremely large datasets across multiple experiments becomes critical. There is a requirement to develop algorithms which are fast and can cluster extremely large datasets without affecting the cluster quality. Clustering is an unsupervised exploratory technique applied to microarray data to find similar data structures or expression patterns. Because of the high input/output costs involved and large distance matrices calculated, most of the algomerative clustering algorithms fail on large datasets (30 000 + genes/200 + arrays). In this article, we propose a new two-stage algorithm which partitions the high-dimensional space associated with microarray data using hyperplanes. The first stage is based on the Balanced Iterative Reducing and Clustering using Hierarchies algorithm with the second stage being a conventional k-means clustering technique. This algorithm has been implemented in a software tool (HPCluster) designed to cluster gene expression data. We compared the clustering results using the two-stage hyperplane algorithm with the conventional k-means algorithm from other available programs. Because, the first stage traverses the data in a single scan, the performance and speed increases substantially. The data reduction accomplished in the first stage of the algorithm reduces the memory requirements allowing us to cluster 44 460 genes without failure and significantly decreases the time to complete when compared with popular k-means programs. The software was written in C# (.NET 1.1).
Availability: The program is freely available and can be downloaded from
Supplementary information: Supplementary data are available at Bioinformatics online.
PMCID: PMC2672630  PMID: 19261720
7.  Differences in Levels of Secreted Locus of Enterocyte Effacement Proteins between Human Disease-Associated and Bovine Escherichia coli O157 
Infection and Immunity  2001;69(8):5107-5114.
Ongoing extensive epidemiological studies of verotoxin-carrying Escherichia coli O157 (stx+ eae+) have shown this bacterial pathogen to be common in cattle herds in the United States and the United Kingdom. However, the incidence of disease in humans due to this pathogen is still very low. This study set out to investigate if there is a difference between strains isolated from human disease cases and those isolated from asymptomatic cattle which would account for the low disease incidence of such a ubiquitous organism. The work presented here has compared human disease strains from both sporadic and outbreak cases with a cross-section, as defined by pulsed-field gel electrophoresis, of E. coli O157 strains from cattle. Human (n = 22) and bovine (n = 31) strains were genotyped for carriage of the genes for Shiga-like toxin types 1, 2, and 2c; E. coli secreted protein genes espA, espB, and espP; the enterohemolysin gene; eae (intimin); ast (enteroaggregative E. coli stable toxin [EAST]); and genes for common E. coli adhesins. Strains were also phenotyped for hemolysin, EspP, Tir, and EspD expression as well as production of actin and cytoskeletal rearrangement associated with attaching and effacing (A/E) lesions on HeLa cells. The genotyping confirmed that there was little difference between the two groups, including carriage of stx2 and stx2c, which was similar in both sets. ast alleles were confirmed to all contain mutations that would prevent EAST expression. espP mutations were found only in cattle strains (5 of 30). Clear differences were observed in the expression of locus of enterocyte effacement (LEE)-encoded factors between strains and in different media. EspD, as an indicator of LEE4 (espA, -B, and -D) expression, and Tir levels in supernatants were measured. Virtually all strains from both sources could produce EspD in Luria-Bertani broth, although at very different levels. Standard trichloroacetic acid precipitation of secreted proteins from tissue culture medium produced detectable levels of EspD from the majority of strains of human origin (15 of 20) compared with only a few (4 of 20) bovine strains (P < 0.001), which is indicative of much higher levels of protein secretion from the human strains. Addition of bovine serum albumin carrier protein before precipitation and enhanced detection techniques confirmed that EspD could be detected after growth in tissue culture medium for all strains, but levels from strains of human origin were on average 90-fold higher than those from strains of bovine origin. In general, levels of secretion also correlated with ability to form A/E lesions on HeLa cells, with only the high-level protein secretors in tissue culture medium exhibiting a localized adherence phenotype. This research shows significant differences between human- and bovine-derived E. coli O157 (stx+ eae+) strains and their production of certain LEE-encoded virulence factors. These data support the recent finding of Kim et al. (J. Kim, J. Nietfeldt, and A. K. Benson, Proc. Natl. Acad. Sci. USA 96:13288–13293, 1999) proposing different E. coli O157 lineages in cattle and humans and extend the differential to the regulation of virulence factors. Potentially only a subset of E. coli O157 isolates (stx+ eae+) in cattle may be capable of causing severe disease in humans.
PMCID: PMC98606  PMID: 11447192
8.  Predictive gene lists for breast cancer prognosis: A topographic visualisation study 
The controversy surrounding the non-uniqueness of predictive gene lists (PGL) of small selected subsets of genes from very large potential candidates as available in DNA microarray experiments is now widely acknowledged [1]. Many of these studies have focused on constructing discriminative semi-parametric models and as such are also subject to the issue of random correlations of sparse model selection in high dimensional spaces. In this work we outline a different approach based around an unsupervised patient-specific nonlinear topographic projection in predictive gene lists.
We construct nonlinear topographic projection maps based on inter-patient gene-list relative dissimilarities. The Neuroscale, the Stochastic Neighbor Embedding(SNE) and the Locally Linear Embedding(LLE) techniques have been used to construct two-dimensional projective visualisation plots of 70 dimensional PGLs per patient, classifiers are also constructed to identify the prognosis indicator of each patient using the resulting projections from those visualisation techniques and investigate whether a-posteriori two prognosis groups are separable on the evidence of the gene lists.
A literature-proposed predictive gene list for breast cancer is benchmarked against a separate gene list using the above methods. Generalisation ability is investigated by using the mapping capability of Neuroscale to visualise the follow-up study, but based on the projections derived from the original dataset.
The results indicate that small subsets of patient-specific PGLs have insufficient prognostic dissimilarity to permit a distinction between two prognosis patients. Uncertainty and diversity across multiple gene expressions prevents unambiguous or even confident patient grouping. Comparative projections across different PGLs provide similar results.
The random correlation effect to an arbitrary outcome induced by small subset selection from very high dimensional interrelated gene expression profiles leads to an outcome with associated uncertainty. This continuum and uncertainty precludes any attempts at constructing discriminative classifiers.
However a patient's gene expression profile could possibly be used in treatment planning, based on knowledge of other patients' responses.
We conclude that many of the patients involved in such medical studies are intrinsically unclassifiable on the basis of provided PGL evidence. This additional category of 'unclassifiable' should be accommodated within medical decision support systems if serious errors and unnecessary adjuvant therapy are to be avoided.
PMCID: PMC2375896  PMID: 18419801
9.  Independent component analysis of Alzheimer's DNA microarray gene expression data 
Gene microarray technology is an effective tool to investigate the simultaneous activity of multiple cellular pathways from hundreds to thousands of genes. However, because data in the colossal amounts generated by DNA microarray technology are usually complex, noisy, high-dimensional, and often hindered by low statistical power, their exploitation is difficult. To overcome these problems, two kinds of unsupervised analysis methods for microarray data: principal component analysis (PCA) and independent component analysis (ICA) have been developed to accomplish the task. PCA projects the data into a new space spanned by the principal components that are mutually orthonormal to each other. The constraint of mutual orthogonality and second-order statistics technique within PCA algorithms, however, may not be applied to the biological systems studied. Extracting and characterizing the most informative features of the biological signals, however, require higher-order statistics.
ICA is one of the unsupervised algorithms that can extract higher-order statistical structures from data and has been applied to DNA microarray gene expression data analysis. We performed FastICA method on DNA microarray gene expression data from Alzheimer's disease (AD) hippocampal tissue samples and consequential gene clustering. Experimental results showed that the ICA method can improve the clustering results of AD samples and identify significant genes. More than 50 significant genes with high expression levels in severe AD were extracted, representing immunity-related protein, metal-related protein, membrane protein, lipoprotein, neuropeptide, cytoskeleton protein, cellular binding protein, and ribosomal protein. Within the aforementioned categories, our method also found 37 significant genes with low expression levels. Moreover, it is worth noting that some oncogenes and phosphorylation-related proteins are expressed in low levels. In comparison to the PCA and support vector machine recursive feature elimination (SVM-RFE) methods, which are widely used in microarray data analysis, ICA can identify more AD-related genes. Furthermore, we have validated and identified many genes that are associated with AD pathogenesis.
We demonstrated that ICA exploits higher-order statistics to identify gene expression profiles as linear combinations of elementary expression patterns that lead to the construction of potential AD-related pathogenic pathways. Our computing results also validated that the ICA model outperformed PCA and the SVM-RFE method. This report shows that ICA as a microarray data analysis tool can help us to elucidate the molecular taxonomy of AD and other multifactorial and polygenic complex diseases.
PMCID: PMC2646728  PMID: 19173745
Molecular Microbiology  2011;81(3):734-750.
Enterohemorrhagic E. coli is a causative agent of gastrointestinal and diarrheal diseases. These pathogenic E. coli express a syringe like protein machine, known as the type III secretion system (T3SS), used for the injection of virulence factors into the cytosol of the host epithelial cell. Breaching the epithelial plasma membrane requires formation of a translocation pore that contains the secreted protein EspD. Here we demonstrate that the N-terminal segment of EspD, encompassing residues 1–171, contains two amphipathic domains spanning residues 24–41 and 66–83, with the latter of these helices being critical for EspD function. Fluorescence and circular dichroism analysis revealed that, in solution, His6-EspD1-171 adopts a native disordered structure; however, on binding anionic small unilamellar vesicles composed of phosphatidylserine, His6-EspD1-171 undergoes a pH depended conformational change that increases the α-helix content of this protein ~7-fold. This change coincides with insertion of the region circumscribing Trp47 into the hydrophobic core of the lipid bilayer. On the HeLa cell plasma membrane, His6-EspD1-171 forms a homodimer that is postulated to promote EspD-EspD oligomerization and pore formation. Complementation of ΔespD null mutant bacteria with an espDΔ66-83 gene showed that this protein was secreted but non-functional.
PMCID: PMC3254054  PMID: 21651628
Type III Secretion; T3SS; E. coli; EspD; enteropathogenic; enterohemorrhagic
11.  Using Pre-existing Microarray Datasets to Increase Experimental Power: Application to Insulin Resistance 
PLoS Computational Biology  2010;6(3):e1000718.
Although they have become a widely used experimental technique for identifying differentially expressed (DE) genes, DNA microarrays are notorious for generating noisy data. A common strategy for mitigating the effects of noise is to perform many experimental replicates. This approach is often costly and sometimes impossible given limited resources; thus, analytical methods are needed which increase accuracy at no additional cost. One inexpensive source of microarray replicates comes from prior work: to date, data from hundreds of thousands of microarray experiments are in the public domain. Although these data assay a wide range of conditions, they cannot be used directly to inform any particular experiment and are thus ignored by most DE gene methods. We present the SVD Augmented Gene expression Analysis Tool (SAGAT), a mathematically principled, data-driven approach for identifying DE genes. SAGAT increases the power of a microarray experiment by using observed coexpression relationships from publicly available microarray datasets to reduce uncertainty in individual genes' expression measurements. We tested the method on three well-replicated human microarray datasets and demonstrate that use of SAGAT increased effective sample sizes by as many as 2.72 arrays. We applied SAGAT to unpublished data from a microarray study investigating transcriptional responses to insulin resistance, resulting in a 50% increase in the number of significant genes detected. We evaluated 11 (58%) of these genes experimentally using qPCR, confirming the directions of expression change for all 11 and statistical significance for three. Use of SAGAT revealed coherent biological changes in three pathways: inflammation, differentiation, and fatty acid synthesis, furthering our molecular understanding of a type 2 diabetes risk factor. We envision SAGAT as a means to maximize the potential for biological discovery from subtle transcriptional responses, and we provide it as a freely available software package that is immediately applicable to any human microarray study.
Author Summary
Though the use of microarrays to identify differentially expressed (DE) genes has become commonplace, it is still not a trivial task. Microarray data are notorious for being noisy, and current DE gene methods do not fully utilize pre-existing biological knowledge to help control this noise. One such source of knowledge is the vast number of publicly available microarray datasets. To leverage this information, we have developed the SVD Augmented Gene expression Analysis Tool (SAGAT) for identifying DE genes. SAGAT extracts transcriptional modules from publicly available microarray data and integrates this information with a dataset of interest. We explore SAGAT's ability to improve DE gene identification on simulated data, and we validate the method on three highly replicated biological datasets. Finally, we demonstrate SAGAT's effectiveness on a novel human dataset investigating the transcriptional response to insulin resistance. Use of SAGAT leads to an increased number of insulin resistant candidate genes, and we validate a subset of these with qPCR. We provide SAGAT as an open source R package that is applicable to any human microarray study.
PMCID: PMC2845644  PMID: 20361040
12.  Batch Mode Active Sampling based on Marginal Probability Distribution Matching 
Active Learning is a machine learning and data mining technique that selects the most informative samples for labeling and uses them as training data; it is especially useful when there are large amount of unlabeled data and labeling them is expensive. Recently, batch-mode active learning, where a set of samples are selected concurrently for labeling, based on their collective merit, has attracted a lot of attention. The objective of batch-mode active learning is to select a set of informative samples so that a classifier learned on these samples has good generalization performance on the unlabeled data. Most of the existing batch-mode active learning methodologies try to achieve this by selecting samples based on varied criteria. In this paper we propose a novel criterion which achieves good generalization performance of a classifier by specifically selecting a set of query samples that minimizes the difference in distribution between the labeled and the unlabeled data, after annotation. We explicitly measure this difference based on all candidate subsets of the unlabeled data and select the best subset. The proposed objective is an NP-hard integer programming optimization problem. We provide two optimization techniques to solve this problem. In the first one, the problem is transformed into a convex quadratic programming problem and in the second method the problem is transformed into a linear programming problem. Our empirical studies using publicly available UCI datasets and a biomedical image dataset demonstrate the effectiveness of the proposed approach in comparison with the state-of-the-art batch-mode active learning methods. We also present two extensions of the proposed approach, which incorporate uncertainty of the predicted labels of the unlabeled data and transfer learning in the proposed formulation. Our empirical studies on UCI datasets show that incorporation of uncertainty information improves performance at later iterations while our studies on 20 Newsgroups dataset show that transfer learning improves the performance of the classifier during initial iterations.
PMCID: PMC4191836  PMID: 25309807
Active learning; marginal probability distribution; Maximum Mean Discrepancy
13.  Pathway activity inference for multiclass disease classification through a mathematical programming optimisation framework 
BMC Bioinformatics  2014;15(1):390.
Applying machine learning methods on microarray gene expression profiles for disease classification problems is a popular method to derive biomarkers, i.e. sets of genes that can predict disease state or outcome. Traditional approaches where expression of genes were treated independently suffer from low prediction accuracy and difficulty of biological interpretation. Current research efforts focus on integrating information on protein interactions through biochemical pathway datasets with expression profiles to propose pathway-based classifiers that can enhance disease diagnosis and prognosis. As most of the pathway activity inference methods in literature are either unsupervised or applied on two-class datasets, there is good scope to address such limitations by proposing novel methodologies.
A supervised multiclass pathway activity inference method using optimisation techniques is reported. For each pathway expression dataset, patterns of its constituent genes are summarised into one composite feature, termed pathway activity, and a novel mathematical programming model is proposed to infer this feature as a weighted linear summation of expression of its constituent genes. Gene weights are determined by the optimisation model, in a way that the resulting pathway activity has the optimal discriminative power with regards to disease phenotypes. Classification is then performed on the resulting low-dimensional pathway activity profile.
The model was evaluated through a variety of published gene expression profiles that cover different types of disease. We show that not only does it improve classification accuracy, but it can also perform well in multiclass disease datasets, a limitation of other approaches from the literature. Desirable features of the model include the ability to control the maximum number of genes that may participate in determining pathway activity, which may be pre-specified by the user. Overall, this work highlights the potential of building pathway-based multi-phenotype classifiers for accurate disease diagnosis and prognosis problems.
Electronic supplementary material
The online version of this article (doi:10.1186/s12859-014-0390-2) contains supplementary material, which is available to authorized users.
PMCID: PMC4269079  PMID: 25475756
Disease classification; Microarray; Pathway activity; Mathematical programming; Optimisation
14.  A bi-ordering approach to linking gene expression with clinical annotations in gastric cancer 
BMC Bioinformatics  2010;11:477.
In the study of cancer genomics, gene expression microarrays, which measure thousands of genes in a single assay, provide abundant information for the investigation of interesting genes or biological pathways. However, in order to analyze the large number of noisy measurements in microarrays, effective and efficient bioinformatics techniques are needed to identify the associations between genes and relevant phenotypes. Moreover, systematic tests are needed to validate the statistical and biological significance of those discoveries.
In this paper, we develop a robust and efficient method for exploratory analysis of microarray data, which produces a number of different orderings (rankings) of both genes and samples (reflecting correlation among those genes and samples). The core algorithm is closely related to biclustering, and so we first compare its performance with several existing biclustering algorithms on two real datasets - gastric cancer and lymphoma datasets. We then show on the gastric cancer data that the sample orderings generated by our method are highly statistically significant with respect to the histological classification of samples by using the Jonckheere trend test, while the gene modules are biologically significant with respect to biological processes (from the Gene Ontology). In particular, some of the gene modules associated with biclusters are closely linked to gastric cancer tumorigenesis reported in previous literature, while others are potentially novel discoveries.
In conclusion, we have developed an effective and efficient method, Bi-Ordering Analysis, to detect informative patterns in gene expression microarrays by ranking genes and samples. In addition, a number of evaluation metrics were applied to assess both the statistical and biological significance of the resulting bi-orderings. The methodology was validated on gastric cancer and lymphoma datasets.
PMCID: PMC2949898  PMID: 20860844
15.  Recursive Cluster Elimination (RCE) for classification and feature selection from gene expression data 
BMC Bioinformatics  2007;8:144.
Classification studies using gene expression datasets are usually based on small numbers of samples and tens of thousands of genes. The selection of those genes that are important for distinguishing the different sample classes being compared, poses a challenging problem in high dimensional data analysis. We describe a new procedure for selecting significant genes as recursive cluster elimination (RCE) rather than recursive feature elimination (RFE). We have tested this algorithm on six datasets and compared its performance with that of two related classification procedures with RFE.
We have developed a novel method for selecting significant genes in comparative gene expression studies. This method, which we refer to as SVM-RCE, combines K-means, a clustering method, to identify correlated gene clusters, and Support Vector Machines (SVMs), a supervised machine learning classification method, to identify and score (rank) those gene clusters for the purpose of classification. K-means is used initially to group genes into clusters. Recursive cluster elimination (RCE) is then applied to iteratively remove those clusters of genes that contribute the least to the classification performance. SVM-RCE identifies the clusters of correlated genes that are most significantly differentially expressed between the sample classes. Utilization of gene clusters, rather than individual genes, enhances the supervised classification accuracy of the same data as compared to the accuracy when either SVM or Penalized Discriminant Analysis (PDA) with recursive feature elimination (SVM-RFE and PDA-RFE) are used to remove genes based on their individual discriminant weights.
SVM-RCE provides improved classification accuracy with complex microarray data sets when it is compared to the classification accuracy of the same datasets using either SVM-RFE or PDA-RFE. SVM-RCE identifies clusters of correlated genes that when considered together provide greater insight into the structure of the microarray data. Clustering genes for classification appears to result in some concomitant clustering of samples into subgroups.
Our present implementation of SVM-RCE groups genes using the correlation metric. The success of the SVM-RCE method in classification suggests that gene interaction networks or other biologically relevant metrics that group genes based on functional parameters might also be useful.
PMCID: PMC1877816  PMID: 17474999
16.  Confident difference criterion: a new Bayesian differentially expressed gene selection algorithm with applications 
BMC Bioinformatics  2015;16(1):245.
Recently, the Bayesian method becomes more popular for analyzing high dimensional gene expression data as it allows us to borrow information across different genes and provides powerful estimators for evaluating gene expression levels. It is crucial to develop a simple but efficient gene selection algorithm for detecting differentially expressed (DE) genes based on the Bayesian estimators.
In this paper, by extending the two-criterion idea of Chen et al. (Chen M-H, Ibrahim JG, Chi Y-Y. A new class of mixture models for differential gene expression in DNA microarray data. J Stat Plan Inference. 2008;138:387–404), we propose two new gene selection algorithms for general Bayesian models and name these new methods as the confident difference criterion methods. One is based on the standardized differences between two mean expression values among genes; the other adds the differences between two variances to it. The proposed confident difference criterion methods first evaluate the posterior probability of a gene having different gene expressions between competitive samples and then declare a gene to be DE if the posterior probability is large. The theoretical connection between the proposed first method based on the means and the Bayes factor approach proposed by Yu et al. (Yu F, Chen M-H, Kuo L. Detecting differentially expressed genes using alibrated Bayes factors. Statistica Sinica. 2008;18:783–802) is established under the normal-normal-model with equal variances between two samples. The empirical performance of the proposed methods is examined and compared to those of several existing methods via several simulations. The results from these simulation studies show that the proposed confident difference criterion methods outperform the existing methods when comparing gene expressions across different conditions for both microarray studies and sequence-based high-throughput studies. A real dataset is used to further demonstrate the proposed methodology. In the real data application, the confident difference criterion methods successfully identified more clinically important DE genes than the other methods.
The confident difference criterion method proposed in this paper provides a new efficient approach for both microarray studies and sequence-based high-throughput studies to identify differentially expressed genes.
Electronic supplementary material
The online version of this article (doi:10.1186/s12859-015-0664-3) contains supplementary material, which is available to authorized users.
PMCID: PMC4527130  PMID: 26250443
Bayesian; Differential expression; Microarray; Next-generation sequencing
17.  Sparse and Compositionally Robust Inference of Microbial Ecological Networks 
PLoS Computational Biology  2015;11(5):e1004226.
16S ribosomal RNA (rRNA) gene and other environmental sequencing techniques provide snapshots of microbial communities, revealing phylogeny and the abundances of microbial populations across diverse ecosystems. While changes in microbial community structure are demonstrably associated with certain environmental conditions (from metabolic and immunological health in mammals to ecological stability in soils and oceans), identification of underlying mechanisms requires new statistical tools, as these datasets present several technical challenges. First, the abundances of microbial operational taxonomic units (OTUs) from amplicon-based datasets are compositional. Counts are normalized to the total number of counts in the sample. Thus, microbial abundances are not independent, and traditional statistical metrics (e.g., correlation) for the detection of OTU-OTU relationships can lead to spurious results. Secondly, microbial sequencing-based studies typically measure hundreds of OTUs on only tens to hundreds of samples; thus, inference of OTU-OTU association networks is severely under-powered, and additional information (or assumptions) are required for accurate inference. Here, we present SPIEC-EASI (SParse InversE Covariance Estimation for Ecological Association Inference), a statistical method for the inference of microbial ecological networks from amplicon sequencing datasets that addresses both of these issues. SPIEC-EASI combines data transformations developed for compositional data analysis with a graphical model inference framework that assumes the underlying ecological association network is sparse. To reconstruct the network, SPIEC-EASI relies on algorithms for sparse neighborhood and inverse covariance selection. To provide a synthetic benchmark in the absence of an experimentally validated gold-standard network, SPIEC-EASI is accompanied by a set of computational tools to generate OTU count data from a set of diverse underlying network topologies. SPIEC-EASI outperforms state-of-the-art methods to recover edges and network properties on synthetic data under a variety of scenarios. SPIEC-EASI also reproducibly predicts previously unknown microbial associations using data from the American Gut project.
Author Summary
Genomic survey of microbes by 16S rRNA gene sequencing and metagenomics has inspired appreciation for the role of complex communities in diverse ecosystems. However, due to the unique properties of community composition data, standard data analysis tools are likely to produce statistical artifacts. For a typical experiment studying microbial ecosystems these artifacts can lead to erroneous conclusions about patterns of associations between microbial taxa. We developed a new procedure that seeks to infer ecological associations between microbial populations, by 1) taking advantage of the proportionality invariance of relative abundance data and 2) making assumptions about the underlying network structure when the number of taxa in the dataset is larger than the number of sampled communities. Additionally, we employed a novel tool to generate biologically plausible synthetic data and objectively benchmark current association inference tools. Finally, we tested our procedures on a large-scale 16S rRNA gene sequencing dataset sampled from the human gut.
PMCID: PMC4423992  PMID: 25950956
18.  Diagnostic prediction of complex diseases using phase-only correlation based on virtual sample template 
BMC Bioinformatics  2013;14(Suppl 8):S11.
Complex diseases induce perturbations to interaction and regulation networks in living systems, resulting in dynamic equilibrium states that differ for different diseases and also normal states. Thus identifying gene expression patterns corresponding to different equilibrium states is of great benefit to the diagnosis and treatment of complex diseases. However, it remains a major challenge to deal with the high dimensionality and small size of available complex disease gene expression datasets currently used for discovering gene expression patterns.
Here we present a phase-only correlation (POC) based classification method for recognizing the type of complex diseases. First, a virtual sample template is constructed for each subclass by averaging all samples of each subclass in a training dataset. Then the label of a test sample is determined by measuring the similarity between the test sample and each template. This novel method can detect the similarity of overall patterns emerged from the differentially expressed genes or proteins while ignoring small mismatches.
The experimental results obtained on seven publicly available complex disease datasets including microarray and protein array data demonstrate that the proposed POC-based disease classification method is effective and robust for diagnosing complex diseases with regard to the number of initially selected features, and its recognition accuracy is better than or comparable to other state-of-the-art machine learning methods. In addition, the proposed method does not require parameter tuning and data scaling, which can effectively reduce the occurrence of over-fitting and bias.
PMCID: PMC3654928  PMID: 23815677
19.  Finding biomarkers in non-model species: literature mining of transcription factors involved in bovine embryo development 
BioData Mining  2012;5:12.
Since processes in well-known model organisms have specific features different from those in Bos taurus, the organism under study, a good way to describe gene regulation in ruminant embryos would be a species-specific consideration of closely related species to cattle, sheep and pig. However, as highlighted by a recent report, gene dictionaries in pig are smaller than in cattle, bringing a risk to reduce the gene resources to be mined (and so for sheep dictionaries). Bioinformatics approaches that allow an integration of available information on gene function in model organisms, taking into account their specificity, are thus needed. Besides these closely related and biologically relevant species, there is indeed much more knowledge of (i) trophoblast proliferation and differentiation or (ii) embryogenesis in human and mouse species, which provides opportunities for reconstructing proliferation and/or differentiation processes in other mammalian embryos, including ruminants. The necessary knowledge can be obtained partly from (i) stem cell or cancer research to supply useful information on molecular agents or molecular interactions at work in cell proliferation and (ii) mouse embryogenesis to supply useful information on embryo differentiation. However, the total number of publications for all these topics and species is great and their manual processing would be tedious and time consuming. This is why we used text mining for automated text analysis and automated knowledge extraction. To evaluate the quality of this “mining”, we took advantage of studies that reported gene expression profiles during the elongation of bovine embryos and defined a list of transcription factors (or TF, n = 64) that we used as biological “gold standard”. When successful, the “mining” approach would identify them all, as well as novel ones.
To gain knowledge on molecular-genetic regulations in a non model organism, we offer an approach based on literature-mining and score arrangement of data from model organisms. This approach was applied to identify novel transcription factors during bovine blastocyst elongation, a process that is not observed in rodents and primates. As a result, searching through human and mouse corpuses, we identified numerous bovine homologs, among which 11 to 14% of transcription factors including the gold standard TF as well as novel TF potentially important to gene regulation in ruminant embryo development. The scripts of the workflow are written in Perl and available on demand. They require data input coming from all various databases for any kind of biological issue once the data has been prepared according to keywords for the studied topic and species; we can provide data sample to illustrate the use and functionality of the workflow.
To do so, we created a workflow that allowed the pipeline processing of literature data and biological data, extracted from Web of Science (WoS) or PubMed but also from Gene Expression Omnibus (GEO), Gene Ontology (GO), Uniprot, HomoloGene, TcoF-DB and TFe (TF encyclopedia). First, the human and mouse homologs of the bovine proteins were selected, filtered by text corpora and arranged by score functions. The score functions were based on the gene name frequencies in corpora. Then, transcription factors were identified using TcoF-DB and double-checked using TFe to characterise TF groups and families. Thus, among a search space of 18,670 bovine homologs, 489 were identified as transcription factors. Among them, 243 were absent from the high-throughput data available at the time of the study. They thus stand so far for putative TF acting during bovine embryo elongation, but might be retrieved from a recent RNA sequencing dataset (Mamo et al. , 2012). Beyond the 246 TF that appeared expressed in bovine elongating tissues, we restricted our interpretation to those occurring within a list of 50 top-ranked genes. Among the transcription factors identified therein, half belonged to the gold standard (ASCL2, c-FOS, ETS2, GATA3, HAND1) and half did not (ESR1, HES1, ID2, NANOG, PHB2, TP53, STAT3).
A workflow providing search for transcription factors acting in bovine elongation was developed. The model assumed that proteins sharing the same protein domains in closely related species had the same protein functionalities, even if they were differently regulated among species or involved in somewhat different pathways. Under this assumption, we merged the information on different mammalian species from different databases (literature and biology) and proposed 489 TF as potential participants of embryo proliferation and differentiation, with (i) a recall of 95% with regard to a biological gold standard defined in 2011 and (ii) an extension of more than 3 times the gold standard of TF detected so far in elongating tissues. The working capacity of the workflow was supported by the manual expertise of the biologists on the results. The workflow can serve as a new kind of bioinformatics tool to work on fused data sources and can thus be useful in studies of a wide range of biological processes.
PMCID: PMC3563503  PMID: 22931563
20.  Feature Selection and Classification of MAQC-II Breast Cancer and Multiple Myeloma Microarray Gene Expression Data 
PLoS ONE  2009;4(12):e8250.
Microarray data has a high dimension of variables but available datasets usually have only a small number of samples, thereby making the study of such datasets interesting and challenging. In the task of analyzing microarray data for the purpose of, e.g., predicting gene-disease association, feature selection is very important because it provides a way to handle the high dimensionality by exploiting information redundancy induced by associations among genetic markers. Judicious feature selection in microarray data analysis can result in significant reduction of cost while maintaining or improving the classification or prediction accuracy of learning machines that are employed to sort out the datasets. In this paper, we propose a gene selection method called Recursive Feature Addition (RFA), which combines supervised learning and statistical similarity measures. We compare our method with the following gene selection methods:
Support Vector Machine Recursive Feature Elimination (SVMRFE)Leave-One-Out Calculation Sequential Forward Selection (LOOCSFS)Gradient based Leave-one-out Gene Selection (GLGS)
To evaluate the performance of these gene selection methods, we employ several popular learning classifiers on the MicroArray Quality Control phase II on predictive modeling (MAQC-II) breast cancer dataset and the MAQC-II multiple myeloma dataset. Experimental results show that gene selection is strictly paired with learning classifier. Overall, our approach outperforms other compared methods. The biological functional analysis based on the MAQC-II breast cancer dataset convinced us to apply our method for phenotype prediction. Additionally, learning classifiers also play important roles in the classification of microarray data and our experimental results indicate that the Nearest Mean Scale Classifier (NMSC) is a good choice due to its prediction reliability and its stability across the three performance measurements: Testing accuracy, MCC values, and AUC errors.
PMCID: PMC2789385  PMID: 20011240
21.  Analyzing miRNA co-expression networks to explore TF-miRNA regulation 
BMC Bioinformatics  2009;10:163.
Current microRNA (miRNA) research in progress has engendered rapid accumulation of expression data evolving from microarray experiments. Such experiments are generally performed over different tissues belonging to a specific species of metazoan. For disease diagnosis, microarray probes are also prepared with tissues taken from similar organs of different candidates of an organism. Expression data of miRNAs are frequently mapped to co-expression networks to study the functions of miRNAs, their regulation on genes and to explore the complex regulatory network that might exist between Transcription Factors (TFs), genes and miRNAs. These directions of research relating miRNAs are still not fully explored, and therefore, construction of reliable and compatible methods for mining miRNA co-expression networks has become an emerging area. This paper introduces a novel method for mining the miRNA co-expression networks in order to obtain co-expressed miRNAs under the hypothesis that these might be regulated by common TFs.
Three co-expression networks, configured from one patient-specific, one tissue-specific and a stem cell-based miRNA expression data, are studied for analyzing the proposed methodology. A novel compactness measure is introduced. The results establish the statistical significance of the sets of miRNAs evolved and the efficacy of the self-pruning phase employed by the proposed method. All these datasets yield similar network patterns and produce coherent groups of miRNAs. The existence of common TFs, regulating these groups of miRNAs, is empirically tested. The results found are very promising. A novel visual validation method is also proposed that reflects the homogeneity as well as statistical properties of the grouped miRNAs. This visual validation method provides a promising and statistically significant graphical tool for expression analysis.
A heuristic mining methodology that resembles a clustering motivation is proposed in this paper. However, there remains a basic difference between the mining method and a clustering approach. The heuristic approach can produce priority modules (PM) from an miRNA co-expression network, by employing a self-pruning phase, which are analyzed for statistical and biological significance. The mining algorithm minimizes the space/time complexity of the analysis, and also handles noise in the data. In addition, the mining method reveals promising results in the unsupervised analysis of TF-miRNA regulation.
PMCID: PMC2707367  PMID: 19476620
22.  Identifying genetic marker sets associated with phenotypes via an efficient adaptive score test 
Biostatistics (Oxford, England)  2012;13(4):776-790.
In recent years, genome-wide association studies (GWAS) and gene-expression profiling have generated a large number of valuable datasets for assessing how genetic variations are related to disease outcomes. With such datasets, it is often of interest to assess the overall effect of a set of genetic markers, assembled based on biological knowledge. Genetic marker-set analyses have been advocated as more reliable and powerful approaches compared with the traditional marginal approaches (Curtis and others, 2005. Pathways to the analysis of microarray data. TRENDS in Biotechnology 23, 429–435; Efroni and others, 2007. Identification of key processes underlying cancer phenotypes using biologic pathway analysis. PLoS One 2, 425). Procedures for testing the overall effect of a marker-set have been actively studied in recent years. For example, score tests derived under an Empirical Bayes (EB) framework (Liu and others, 2007. Semiparametric regression of multidimensional genetic pathway data: least-squares kernel machines and linear mixed models. Biometrics 63, 1079–1088; Liu and others, 2008. Estimation and testing for the effect of a genetic pathway on a disease outcome using logistic kernel machine regression via logistic mixed models. BMC bioinformatics 9, 292–2; Wu and others, 2010. Powerful SNP-set analysis for case-control genome-wide association studies. American Journal of Human Genetics 86, 929) have been proposed as powerful alternatives to the standard Rao score test (Rao, 1948. Large sample tests of statistical hypotheses concerning several parameters with applications to problems of estimation. Mathematical Proceedings of the Cambridge Philosophical Society, 44, 50–57). The advantages of these EB-based tests are most apparent when the markers are correlated, due to the reduction in the degrees of freedom. In this paper, we propose an adaptive score test which up- or down-weights the contributions from each member of the marker-set based on the Z-scores of their effects. Such an adaptive procedure gains power over the existing procedures when the signal is sparse and the correlation among the markers is weak. By combining evidence from both the EB-based score test and the adaptive test, we further construct an omnibus test that attains good power in most settings. The null distributions of the proposed test statistics can be approximated well either via simple perturbation procedures or via distributional approximations. Through extensive simulation studies, we demonstrate that the proposed procedures perform well in finite samples. We apply the tests to a breast cancer genetic study to assess the overall effect of the FGFR2 gene on breast cancer risk.
PMCID: PMC3440238  PMID: 22734045
Adaptive procedures; Empirical Bayes; GWAS; Pathway analysis; Score test; SNP sets
23.  Discovering biclusters in gene expression data based on high-dimensional linear geometries 
BMC Bioinformatics  2008;9:209.
In DNA microarray experiments, discovering groups of genes that share similar transcriptional characteristics is instrumental in functional annotation, tissue classification and motif identification. However, in many situations a subset of genes only exhibits consistent pattern over a subset of conditions. Conventional clustering algorithms that deal with the entire row or column in an expression matrix would therefore fail to detect these useful patterns in the data. Recently, biclustering has been proposed to detect a subset of genes exhibiting consistent pattern over a subset of conditions. However, most existing biclustering algorithms are based on searching for sub-matrices within a data matrix by optimizing certain heuristically defined merit functions. Moreover, most of these algorithms can only detect a restricted set of bicluster patterns.
In this paper, we present a novel geometric perspective for the biclustering problem. The biclustering process is interpreted as the detection of linear geometries in a high dimensional data space. Such a new perspective views biclusters with different patterns as hyperplanes in a high dimensional space, and allows us to handle different types of linear patterns simultaneously by matching a specific set of linear geometries. This geometric viewpoint also inspires us to propose a generic bicluster pattern, i.e. the linear coherent model that unifies the seemingly incompatible additive and multiplicative bicluster models. As a particular realization of our framework, we have implemented a Hough transform-based hyperplane detection algorithm. The experimental results on human lymphoma gene expression dataset show that our algorithm can find biologically significant subsets of genes.
We have proposed a novel geometric interpretation of the biclustering problem. We have shown that many common types of bicluster are just different spatial arrangements of hyperplanes in a high dimensional data space. An implementation of the geometric framework using the Fast Hough transform for hyperplane detection can be used to discover biologically significant subsets of genes under subsets of conditions for microarray data analysis.
PMCID: PMC2386490  PMID: 18433477
24.  Improving accuracy for cancer classification with a new algorithm for genes selection 
BMC Bioinformatics  2012;13:298.
Even though the classification of cancer tissue samples based on gene expression data has advanced considerably in recent years, it faces great challenges to improve accuracy. One of the challenges is to establish an effective method that can select a parsimonious set of relevant genes. So far, most methods for gene selection in literature focus on screening individual or pairs of genes without considering the possible interactions among genes. Here we introduce a new computational method named the Binary Matrix Shuffling Filter (BMSF). It not only overcomes the difficulty associated with the search schemes of traditional wrapper methods and overfitting problem in large dimensional search space but also takes potential gene interactions into account during gene selection. This method, coupled with Support Vector Machine (SVM) for implementation, often selects very small number of genes for easy model interpretability.
We applied our method to 9 two-class gene expression datasets involving human cancers. During the gene selection process, the set of genes to be kept in the model was recursively refined and repeatedly updated according to the effect of a given gene on the contributions of other genes in reference to their usefulness in cancer classification. The small number of informative genes selected from each dataset leads to significantly improved leave-one-out (LOOCV) classification accuracy across all 9 datasets for multiple classifiers. Our method also exhibits broad generalization in the genes selected since multiple commonly used classifiers achieved either equivalent or much higher LOOCV accuracy than those reported in literature.
Evaluation of a gene’s contribution to binary cancer classification is better to be considered after adjusting for the joint effect of a large number of other genes. A computationally efficient search scheme was provided to perform effective search in the extensive feature space that includes possible interactions of many genes. Performance of the algorithm applied to 9 datasets suggests that it is possible to improve the accuracy of cancer classification by a big margin when joint effects of many genes are considered.
PMCID: PMC3562261  PMID: 23148517
25.  Incorporating rich background knowledge for gene named entity classification and recognition 
BMC Bioinformatics  2009;10:223.
Gene named entity classification and recognition are crucial preliminary steps of text mining in biomedical literature. Machine learning based methods have been used in this area with great success. In most state-of-the-art systems, elaborately designed lexical features, such as words, n-grams, and morphology patterns, have played a central part. However, this type of feature tends to cause extreme sparseness in feature space. As a result, out-of-vocabulary (OOV) terms in the training data are not modeled well due to lack of information.
We propose a general framework for gene named entity representation, called feature coupling generalization (FCG). The basic idea is to generate higher level features using term frequency and co-occurrence information of highly indicative features in huge amount of unlabeled data. We examine its performance in a named entity classification task, which is designed to remove non-gene entries in a large dictionary derived from online resources. The results show that new features generated by FCG outperform lexical features by 5.97 F-score and 10.85 for OOV terms. Also in this framework each extension yields significant improvements and the sparse lexical features can be transformed into both a lower dimensional and more informative representation. A forward maximum match method based on the refined dictionary produces an F-score of 86.2 on BioCreative 2 GM test set. Then we combined the dictionary with a conditional random field (CRF) based gene mention tagger, achieving an F-score of 89.05, which improves the performance of the CRF-based tagger by 4.46 with little impact on the efficiency of the recognition system. A demo of the NER system is available at .
PMCID: PMC2725142  PMID: 19615051

Results 1-25 (1382442)