Supervised principal component analysis for gene set enrichment of microarray data with continuous or survival outcomes

MOTIVATION: Gene set analysis allows formal testing of subtle but coordinated changes in a group of genes, such as those defined by Gene Ontology (GO) or KEGG Pathway databases. We propose a new method for gene set analysis that is based on principal component analysis (PCA) of genes expression values in the gene set. PCA is an effective method for reducing high dimensionality and capture variations in gene expression values. However, one limitation with PCA is that the latent variable identified by the first PC may be unrelated to outcome. RESULTS: In the proposed supervised PCA (SPCA) model for gene set analysis, the PCs are estimated from a selected subset of genes that are associated with outcome. As outcome information is used in the gene selection step, this method is supervised, thus called the Supervised PCA model. Because of the gene selection step, test statistic in SPCA model can no longer be approximated well using t-distribution. We propose a two-component mixture distribution based on Gumbel exteme value distributions to account for the gene selection step. We show the proposed method compares favorably to currently available gene set analysis methods using simulated and real microarray data. SOFTWARE: The R code for the analysis used in this article are available upon request, we are currently working on implementing the proposed method in an R package.     

A weighted principal component analysis and its application to gene expression data

In this work, we introduce in the first part new developments in Principal Component Analysis (PCA) and in the second part a new method to select variables (genes in our application). Our focus is on problems where the values taken by each variable do not all have the same importance and where the data may be contaminated with noise and contain outliers, as is the case with microarray data. The usual PCA is not appropriate to deal with this kind of problems. In this context, we propose the use of a new correlation coefficient as an alternative to Pearson's. This leads to a so-called weighted PCA (WPCA). In order to illustrate the features of our WPCA and compare it with the usual PCA, we consider the problem of analyzing gene expression data sets. In the second part of this work, we propose a new PCA-based algorithm to iteratively select the most important genes in a microarray data set. We show that this algorithm produces better results when our WPCA is used instead of the usual PCA. Furthermore, by using Support Vector Machines, we show that it can compete with the Significance Analysis of Microarrays algorithm.     

Size correction in biology: how reliable are approaches based on (common) principal component analysis?

Morphological traits typically scale with the overall body size of an organism. A meaningful comparison of trait values among individuals or populations that differ in size therefore requires size correction. A frequently applied size correction method involves subjecting the set of n morphological traits of interest to (common) principal component analysis [(C)PCA], and treating the first principal component [(C)PC1] as a latent size variable. The remaining variation (PC2–PCn) is considered size-independent and interpreted biologically. I here analyze simulated data and natural datasets to demonstrate that this (C)PCA-based size correction generates systematic statistical artifacts. Artifacts arise even when all traits are tightly correlated with overall size, and they are particularly strong when the magnitude of variance is heterogeneous among the traits, and when the traits under study are few. (C)PCA-based approaches are therefore inappropriate for size correction and should be abandoned in favor of methods using univariate general linear models with an adequate independent body size metric as covariate. As I demonstrate, (C)PC1 extracted from a subset of traits, not themselves subjected to size correction, can provide such a size metric.     

Unifying measures of gene function and evolution

Recent genome analyses revealed intriguing correlations between variables characterizing the functioning of a gene, such as expression level (EL), connectivity of genetic and protein-protein interaction networks, and knockout effect, and variables describing gene evolution, such as sequence evolution rate (ER) and propensity for gene loss. Typically, variables within each of these classes are positively correlated, e.g. products of highly expressed genes also have a propensity to be involved in many protein-protein interactions, whereas variables between classes are negatively correlated, e.g. highly expressed genes, on average, evolve slower than weakly expressed genes. Here, we describe principal component (PC) analysis of seven genome-related variables and propose biological interpretations for the first three PCs. The first PC reflects a gene's 'importance', or the 'status' of a gene in the genomic community, with positive contributions from knockout lethality, EL, number of protein-protein interaction partners and the number of paralogues, and negative contributions from sequence ER and gene loss propensity. The next two PCs define a plane that seems to reflect the functional and evolutionary plasticity of a gene. Specifically, PC2 can be interpreted as a gene's 'adaptability' whereby genes with high adaptability readily duplicate, have many genetic interaction partners and tend to be non-essential. PC3 also might reflect the role of a gene in organismal adaptation albeit with a negative rather than a positive contribution of genetic interactions; we provisionally designate this PC 'reactivity'. The interpretation of PC2 and PC3 as measures of a gene's plasticity is compatible with the observation that genes with high values of these PCs tend to be expressed in a condition- or tissue-specific manner. Functional classes of genes substantially vary in status, adaptability and reactivity, with the highest status characteristic of the translation system and cytoskeletal proteins, highest adaptability seen in cellular processes and signalling genes, and top reactivity characteristic of metabolic enzymes.     

Principal component analysis for clustering gene expression data

MOTIVATION: There is a great need to develop analytical methodology to analyze and to exploit the information contained in gene expression data. Because of the large number of genes and the complexity of biological networks, clustering is a useful exploratory technique for analysis of gene expression data. Other classical techniques, such as principal component analysis (PCA), have also been applied to analyze gene expression data. Using different data analysis techniques and different clustering algorithms to analyze the same data set can lead to very different conclusions. Our goal is to study the effectiveness of principal components (PCs) in capturing cluster structure. Specifically, using both real and synthetic gene expression data sets, we compared the quality of clusters obtained from the original data to the quality of clusters obtained after projecting onto subsets of the principal component axes. RESULTS: Our empirical study showed that clustering with the PCs instead of the original variables does not necessarily improve, and often degrades, cluster quality. In particular, the first few PCs (which contain most of the variation in the data) do not necessarily capture most of the cluster structure. We also showed that clustering with PCs has different impact on different algorithms and different similarity metrics. Overall, we would not recommend PCA before clustering except in special circumstances.     

Principal component analysis based methods in bioinformatics studies

In analysis of bioinformatics data, a unique challenge arises from the high dimensionality of measurements. Without loss of generality, we use genomic study with gene expression measurements as a representative example but note that analysis techniques discussed in this article are also applicable to other types of bioinformatics studies. Principal component analysis (PCA) is a classic dimension reduction approach. It constructs linear combinations of gene expressions, called principal components (PCs). The PCs are orthogonal to each other, can effectively explain variation of gene expressions, and may have a much lower dimensionality. PCA is computationally simple and can be realized using many existing software packages. This article consists of the following parts. First, we review the standard PCA technique and their applications in bioinformatics data analysis. Second, we describe recent 'non-standard' applications of PCA, including accommodating interactions among genes, pathways and network modules and conducting PCA with estimating equations as opposed to gene expressions. Third, we introduce several recently proposed PCA-based techniques, including the supervised PCA, sparse PCA and functional PCA. The supervised PCA and sparse PCA have been shown to have better empirical performance than the standard PCA. The functional PCA can analyze time-course gene expression data. Last, we raise the awareness of several critical but unsolved problems related to PCA. The goal of this article is to make bioinformatics researchers aware of the PCA technique and more importantly its most recent development, so that this simple yet effective dimension reduction technique can be better employed in bioinformatics data analysis.     

Maximizing the Power of Principal-Component Analysis of Correlated Phenotypes in Genome-wide Association Studies

Many human traits are highly correlated. This correlation can be leveraged to improve the power of genetic association tests to identify markers associated with one or more of the traits. Principal component analysis (PCA) is a useful tool that has been widely used for the multivariate analysis of correlated variables. PCA is usually applied as a dimension reduction method: the few top principal components (PCs) explaining most of total trait variance are tested for association with a predictor of interest, and the remaining components are not analyzed. In this study we review the theoretical basis of PCA and describe the behavior of PCA when testing for association between a SNP and correlated traits. We then use simulation to compare the power of various PCA-based strategies when analyzing up to 100 correlated traits. We show that contrary to widespread practice, testing only the top PCs often has low power, whereas combining signal across all PCs can have greater power. This power gain is primarily due to increased power to detect genetic variants with opposite effects on positively correlated traits and variants that are exclusively associated with a single trait. Relative to other methods, the combined-PC approach has close to optimal power in all scenarios considered while offering more flexibility and more robustness to potential confounders. Finally, we apply the proposed PCA strategy to the genome-wide association study of five correlated coagulation traits where we identify two candidate SNPs that were not found by the standard approach.     

Exploring pleiotropy using principal components

A standard multivariate principal components (PCs) method was utilized to identify clusters of variables that may be controlled by a common gene or genes (pleiotropy). Heritability estimates were obtained and linkage analyses performed on six individual traits (total cholesterol (Chol), high and low density lipoproteins, triglycerides (TG), body mass index (BMI), and systolic blood pressure (SBP)) and on each PC to compare our ability to identify major gene effects. Using the simulated data from Genetic Analysis Workshop 13 (Cohort 1 and 2 data for year 11), the quantitative traits were first adjusted for age, sex, and smoking (cigarettes per day). Adjusted variables were standardized and PCs calculated followed by orthogonal transformation (varimax rotation). Rotated PCs were then subjected to heritability and quantitative multipoint linkage analysis. The first three PCs explained 73% of the total phenotypic variance. Heritability estimates were above 0.60 for all three PCs. We performed linkage analyses on the PCs as well as the individual traits. The majority of pleiotropic and trait-specific genes were not identified. Standard PCs analysis methods did not facilitate the identification of pleiotropic genes affecting the six traits examined in the simulated data set. In addition, genes contributing 20% of the variance in traits with over 0.60 heritability estimates could not be identified in this simulated data set using traditional quantitative trait linkage analyses. Lack of identification of pleiotropic and trait-specific genes in some cases may reflect their low contribution to the traits/PCs examined or more importantly, characteristics of the sample group analyzed, and not simply a failure of the PC approach itself.     

Synergism and Rules from Combination of Baicalin,Jasminoidin and Desoxycholic acid in Refined Qing Kai Ling for Treat Ischemic Stroke Mice Model

Refined Qing-Kai-Ling (QKL), a modified Chinese medicine, consists of three main ingredients (Baicalin, Jasminoidin and Desoxycholic acid), plays a synergistic effect on the treatment of the acute stage of ischemic stroke. However, the rules of the combination and synergism are still unknown. Based on the ischemic stroke mice model, all different kinds of combination of Baicalin, Jasminoidin, and Desoxycholic acid were investigated by the methods of neurological examination, microarray, and genomics analysis. As a result, it confirmed that the combination of three drugs offered a better therapeutical effect on ischemic stroke than monotherapy of each drug. Additionally, we used Ingenuity pathway Analysis (IPA) and principal component analysis (PCA) to extract the dominant information of expression changes in 373 ischemia-related genes. The results suggested that 5 principal components (PC1-5) could account for more than 95% energy in the gene data. Moreover, 3 clusters (PC1, PC2+PC5, and PC3+PC4) were addressed with cluster analysis. Furthermore, we matched PCs on the drug-target networks, the findings demonstrated that Baicalin related with PC1 that played the leading role in the combination; Jasminoidin related with PC2+PC5 that played a compensatory role; while Desoxycholic acid had the least performance alone which could relate with PC3+PC4 that played a compatible role. These manifestations were accorded with the principle of herbal formulae of Traditional Chinese Medicine (TCM), emperor-minister-adjuvant-courier. In conclusion, we firstly provided scientific evidence to the classic theory of TCM formulae, an initiating holistic viewpoint of combination therapy of TCM. This study also illustrated that PCA might be an applicable method to analyze the complicated data of drug combination.     

A web-based tool for principal component and significance analysis of microarray data

We have developed a program for microarray data analysis, which features the false discovery rate for testing statistical significance and the principal component analysis using the singular value decomposition method for detecting the global trends of gene-expression patterns. Additional features include analysis of variance with multiple methods for error variance adjustment, correction of cross-channel correlation for two-color microarrays, identification of genes specific to each cluster of tissue samples, biplot of tissues and corresponding tissue-specific genes, clustering of genes that are correlated with each principal component (PC), three-dimensional graphics based on virtual reality modeling language and sharing of PC between different experiments. The software also supports parameter adjustment, gene search and graphical output of results. The software is implemented as a web tool and thus the speed of analysis does not depend on the power of a client computer. AVAILABILITY: The tool can be used on-line or downloaded at http://lgsun.grc.nia.nih.gov/ANOVA/     

A simple approach to guide factor retention decisions when applying principal component analysis to biomechanical data

The use of principal component analysis (PCA) as a multivariate statistical approach to reduce complex biomechanical data-sets is growing. With its increased application in biomechanics, there has been a concurrent divergence in the use of criteria to determine how much the data is reduced (i.e. how many principal factors are retained). This short communication presents power equations to support the use of a parallel analysis (PA) criterion as a quantitative and transparent method for determining how many factors to retain when conducting a PCA. Monte Carlo simulation was used to carry out PCA on random data-sets of varying dimension. This process mimicked the PA procedure that would be required to determine principal component (PC) retention for any independent study in which the data-set dimensions fell within the range tested here. A surface was plotted for each of the first eight PCs, expressing the expected outcome of a PA as a function of the dimensions of a data-set. A power relationship was used to fit the surface, facilitating the prediction of the expected outcome of a PA as a function of the dimensions of a data-set. Coefficients used to fit the surface and facilitate prediction are reported. These equations enable the PA to be freely adopted as a criterion to inform PC retention. A transparent and quantifiable criterion to determine how many PCs to retain will enhance the ability to compare and contrast between studies.     

A principal component analysis of the dynamics of subdomains and binding sites in human serum albumin

The conformational dynamics of human serum albumin (HSA) was investigated by principal component analysis (PCA) applied to three molecular dynamics trajectories of 200 ns each. The overlap of the essential subspaces spanned by the first 10 principal components (PC) of different trajectories was about 0.3 showing that the PCA based on a trajectory length of 200 ns is not completely convergent for this protein. The contributions of the relative motion of subdomains and of the subdomains (internal) distortion to the first 10 PCs were found to be comparable. Based on the distribution of the first 3 PC, 10 protein conformers are identified showing relative root mean square deviations (RMSD) between 2.3 and 4.6 Å. The main PCs are found to be delocalized over the whole protein structure indicating that the motions of different protein subdomains are coupled. This coupling is considered as being related to the allosteric effects observed upon ligand binding to HSA. On the other hand, the first PC of one of the three trajectories describes a conformational transition of the protein domain I that is close to that experimentally observed upon myristate binding. This is a theoretical support for the older hypothesis stating that changes of the protein onformation favorable to binding can precede the ligand complexation. A detailed all atoms PCA performed on the primary Sites 1 and 2 confirms the multiconformational character of the HSA binding sites as well as the significant coupling of their motions. © 2013 Wiley Periodicals, Inc. Biopolymers 101: 561–572, 2014.     

A novel data mining method to identify assay-specific signatures in functional genomic studies

Background:  

The highly dimensional data produced by functional genomic (FG) studies makes it difficult to visualize relationships between gene products and experimental conditions (i.e., assays). Although dimensionality reduction methods such as principal component analysis (PCA) have been very useful, their application to identify assay-specific signatures has been limited by the lack of appropriate methodologies. This article proposes a new and powerful PCA-based method for the identification of assay-specific gene signatures in FG studies.     

A PCA-based method for ancestral informative markers selection in structured populations

Identification of population structure can help trace population histories and identify disease genes. Structured association (SA) is a commonly used approach for population structure identification and association mapping. A major issue with SA is that its performance greatly depends on the informativeness and the numbers of ancestral informative markers (AIMs). Present major AIM selection methods mostly require prior individual ancestry information, which is usually not available or uncertain in practice. To address this potential weakness, we herein develop a novel approach for AIM selection based on principle component analysis (PCA), which does not require prior ancestry information of study subjects. Our simulation and real genetic data analysis results suggest that, with equivalent AIMs, PCA-based selected AIMs can significantly increase the accuracy of inferred individual ancestries compared with traditionally randomly selected AIMs. Our method can easily be applied to whole genome data to select a set of highly informative AIMs in population structure, which can then be used to identify potential population structure and correct possible statistical biases caused by population stratification.     

A transition-constrained discrete hidden Markov model for automatic sleep staging

Background

Principal component analysis (PCA) has been widely employed for automatic neuronal spike sorting. Calculating principal components (PCs) is computationally expensive, and requires complex numerical operations and large memory resources. Substantial hardware resources are therefore needed for hardware implementations of PCA. General Hebbian algorithm (GHA) has been proposed for calculating PCs of neuronal spikes in our previous work, which eliminates the needs of computationally expensive covariance analysis and eigenvalue decomposition in conventional PCA algorithms. However, large memory resources are still inherently required for storing a large volume of aligned spikes for training PCs. The large size memory will consume large hardware resources and contribute significant power dissipation, which make GHA difficult to be implemented in portable or implantable multi-channel recording micro-systems.

Method

In this paper, we present a new algorithm for PCA-based spike sorting based on GHA, namely stream-based Hebbian eigenfilter, which eliminates the inherent memory requirements of GHA while keeping the accuracy of spike sorting by utilizing the pseudo-stationarity of neuronal spikes. Because of the reduction of large hardware storage requirements, the proposed algorithm can lead to ultra-low hardware resources and power consumption of hardware implementations, which is critical for the future multi-channel micro-systems. Both clinical and synthetic neural recording data sets were employed for evaluating the accuracy of the stream-based Hebbian eigenfilter. The performance of spike sorting using stream-based eigenfilter and the computational complexity of the eigenfilter were rigorously evaluated and compared with conventional PCA algorithms. Field programmable logic arrays (FPGAs) were employed to implement the proposed algorithm, evaluate the hardware implementations and demonstrate the reduction in both power consumption and hardware memories achieved by the streaming computing

Results and discussion

Results demonstrate that the stream-based eigenfilter can achieve the same accuracy and is 10 times more computationally efficient when compared with conventional PCA algorithms. Hardware evaluations show that 90.3% logic resources, 95.1% power consumption and 86.8% computing latency can be reduced by the stream-based eigenfilter when compared with PCA hardware. By utilizing the streaming method, 92% memory resources and 67% power consumption can be saved when compared with the direct implementation of GHA.

Conclusion

Stream-based Hebbian eigenfilter presents a novel approach to enable real-time spike sorting with reduced computational complexity and hardware costs. This new design can be further utilized for multi-channel neuro-physiological experiments or chronic implants.     

Mining gene expression profiles: an integrated implementation of kernel principal component analysis and singular value decomposition

The detection of genes that show similar profiles under different experimental conditions is often an initial step in inferring the biological significance of such genes. Visualization tools are used to identify genes with similar profiles in microarray studies. Given the large number of genes recorded in microarray experiments, gene expression data are generally displayed on a low dimensional plot, based on linear methods. However, microarray data show nonlinearity, due to high-order terms of interaction between genes, so alternative approaches, such as kernel methods, may be more appropriate. We introduce a technique that combines kernel principal component analysis (KPCA) and Biplot to visualize gene expression profiles. Our approach relies on the singular value decomposition of the input matrix and incorporates an additional step that involves KPCA. The main properties of our method are the extraction of nonlinear features and the preservation of the input variables (genes) in the output display. We apply this algorithm to colon tumor, leukemia and lymphoma datasets. Our approach reveals the underlying structure of the gene expression profiles and provides a more intuitive understanding of the gene and sample association.     

A Class-Information-Based Penalized Matrix Decomposition for Identifying Plants Core Genes Responding to Abiotic Stresses

In terms of making genes expression data more interpretable and comprehensible, there exists a significant superiority on sparse methods. Many sparse methods, such as penalized matrix decomposition (PMD) and sparse principal component analysis (SPCA), have been applied to extract plants core genes. Supervised algorithms, especially the support vector machine-recursive feature elimination (SVM-RFE) method, always have good performance in gene selection. In this paper, we draw into class information via the total scatter matrix and put forward a class-information-based penalized matrix decomposition (CIPMD) method to improve the gene identification performance of PMD-based method. Firstly, the total scatter matrix is obtained based on different samples of the gene expression data. Secondly, a new data matrix is constructed by decomposing the total scatter matrix. Thirdly, the new data matrix is decomposed by PMD to obtain the sparse eigensamples. Finally, the core genes are identified according to the nonzero entries in eigensamples. The results on simulation data show that CIPMD method can reach higher identification accuracies than the conventional gene identification methods. Moreover, the results on real gene expression data demonstrate that CIPMD method can identify more core genes closely related to the abiotic stresses than the other methods.