Pathway analysis using random forests classification and regression

MOTIVATION: Although numerous methods have been developed to better capture biological information from microarray data, commonly used single gene-based methods neglect interactions among genes and leave room for other novel approaches. For example, most classification and regression methods for microarray data are based on the whole set of genes and have not made use of pathway information. Pathway-based analysis in microarray studies may lead to more informative and relevant knowledge for biological researchers. RESULTS: In this paper, we describe a pathway-based classification and regression method using Random Forests to analyze gene expression data. The proposed methods allow researchers to rank important pathways from externally available databases, discover important genes, find pathway-based outlying cases and make full use of a continuous outcome variable in the regression setting. We also compared Random Forests with other machine learning methods using several datasets and found that Random Forests classification error rates were either the lowest or the second-lowest. By combining pathway information and novel statistical methods, this procedure represents a promising computational strategy in dissecting pathways and can provide biological insight into the study of microarray data. AVAILABILITY: Source code written in R is available from http://bioinformatics.med.yale.edu/pathway-analysis/rf.htm.     

A top-r feature selection algorithm for microarray gene expression data

Most of the conventional feature selection algorithms have a drawback whereby a weakly ranked gene that could perform well in terms of classification accuracy with an appropriate subset of genes will be left out of the selection. Considering this shortcoming, we propose a feature selection algorithm in gene expression data analysis of sample classifications. The proposed algorithm first divides genes into subsets, the sizes of which are relatively small (roughly of size h), then selects informative smaller subsets of genes (of size r < h) from a subset and merges the chosen genes with another gene subset (of size r) to update the gene subset. We repeat this process until all subsets are merged into one informative subset. We illustrate the effectiveness of the proposed algorithm by analyzing three distinct gene expression data sets. Our method shows promising classification accuracy for all the test data sets. We also show the relevance of the selected genes in terms of their biological functions.     

Pathway Correlation Profile of Gene-Gene Co-Expression for Identifying Pathway Perturbation

Identifying perturbed or dysregulated pathways is critical to understanding the biological processes that change within an experiment. Previous methods identified important pathways that are significantly enriched among differentially expressed genes; however, these methods cannot account for small, coordinated changes in gene expression that amass across a whole pathway. In order to overcome this limitation, we use microarray gene expression data to identify pathway perturbation based on pathway correlation profiles. By identifying the distribution of gene-gene pair correlations within a pathway, we can rank the pathways based on the level of perturbation and dysregulation. We have shown this successfully for differences between two experimental conditions in Escherichia coli and changes within time series data in Saccharomyces cerevisiae, as well as two estrogen receptor response classes of breast cancer. Overall, our method made significant predictions as to the pathway perturbations that are involved in the experimental conditions.     

Biological pathway selection through nonlinear dimension reduction

In the analysis of high-throughput biological data, it is often believed that the biological units such as genes behave interactively by groups, that is, pathways in our context. It is conceivable that utilization of priorly available pathway knowledge would greatly facilitate both interpretation and estimation in statistical analysis of such high-dimensional biological data. In this article, we propose a 2-step procedure for the purpose of identifying pathways that are related to and influence the clinical phenotype. In the first step, a nonlinear dimension reduction method is proposed, which permits flexible within-pathway gene interactions as well as nonlinear pathway effects on the response. In the second step, a regularized model-based pathway ranking and selection procedure is developed that is built upon the summary features extracted from the first step. Simulations suggest that the new method performs favorably compared to the existing solutions. An analysis of a glioblastoma microarray data finds 4 pathways that have evidence of support from the biological literature.     

Incorporating prior knowledge of predictors into penalized classifiers with multiple penalty terms

MOTIVATION: In the context of sample (e.g. tumor) classifications with microarray gene expression data, many methods have been proposed. However, almost all the methods ignore existing biological knowledge and treat all the genes equally a priori. On the other hand, because some genes have been identified by previous studies to have biological functions or to be involved in pathways related to the outcome (e.g. cancer), incorporating this type of prior knowledge into a classifier can potentially improve both the predictive performance and interpretability of the resulting model. RESULTS: We propose a simple and general framework to incorporate such prior knowledge into building a penalized classifier. As two concrete examples, we apply the idea to two penalized classifiers, nearest shrunken centroids (also called PAM) and penalized partial least squares (PPLS). Instead of treating all the genes equally a priori as in standard penalized methods, we group the genes according to their functional associations based on existing biological knowledge or data, and adopt group-specific penalty terms and penalization parameters. Simulated and real data examples demonstrate that, if prior knowledge on gene grouping is indeed informative, our new methods perform better than the two standard penalized methods, yielding higher predictive accuracy and screening out more irrelevant genes.     

Hierarchical Gene Selection and Genetic Fuzzy System for Cancer Microarray Data Classification

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

Incorporating biological pathways via a Markov random field model in genome-wide association studies

Genome-wide association studies (GWAS) examine a large number of markers across the genome to identify associations between genetic variants and disease. Most published studies examine only single markers, which may be less informative than considering multiple markers and multiple genes jointly because genes may interact with each other to affect disease risk. Much knowledge has been accumulated in the literature on biological pathways and interactions. It is conceivable that appropriate incorporation of such prior knowledge may improve the likelihood of making genuine discoveries. Although a number of methods have been developed recently to prioritize genes using prior biological knowledge, such as pathways, most methods treat genes in a specific pathway as an exchangeable set without considering the topological structure of a pathway. However, how genes are related with each other in a pathway may be very informative to identify association signals. To make use of the connectivity information among genes in a pathway in GWAS analysis, we propose a Markov Random Field (MRF) model to incorporate pathway topology for association analysis. We show that the conditional distribution of our MRF model takes on a simple logistic regression form, and we propose an iterated conditional modes algorithm as well as a decision theoretic approach for statistical inference of each gene's association with disease. Simulation studies show that our proposed framework is more effective to identify genes associated with disease than a single gene-based method. We also illustrate the usefulness of our approach through its applications to a real data example.     

Identifying gene pathways associated with cancer characteristics via sparse statistical methods

We propose a statistical method for uncovering gene pathways that characterize cancer heterogeneity. To incorporate knowledge of the pathways into the model, we define a set of activities of pathways from microarray gene expression data based on the Sparse Probabilistic Principal Component Analysis (SPPCA). A pathway activity logistic regression model is then formulated for cancer phenotype. To select pathway activities related to binary cancer phenotypes, we use the elastic net for the parameter estimation and derive a model selection criterion for selecting tuning parameters included in the model estimation. Our proposed method can also reverse-engineer gene networks based on the identified multiple pathways that enables us to discover novel gene-gene associations relating with the cancer phenotypes. We illustrate the whole process of the proposed method through the analysis of breast cancer gene expression data.     

Identification of novel metastasis suppressor signaling pathways for breast cancer

Cancer lethality is mainly caused by metastasis. Therefore, understanding the nature of the genes involved in this process has become a priority. Given the heterogeneity of mutations in cancer cells, considerable focus has been directed toward characterizing metastasis genes in the context of relevant signaling pathways rather than treating genes as independent and equal entities. One signaling cascade implicated in the regulation of cell growth, invasion and metastasis is the MAP kinase pathway. Raf kinase inhibitory protein (RKIP) functions as an inhibitor of the MAP kinase pathway and is a metastasis suppressor in different cancer models. By utilizing statistical analysis of clinical data integrated with experimental validation, we recently identified components of the RKIP signaling pathway relevant to breast cancer metastasis. Using the RKIP pathway as an example, we show how prior biological knowledge can be efficiently combined with genome-wide patient data to identify gene regulatory mechanisms that control metastasis.     

Prediction of regulatory pathways using mRNA expression and protein interaction data: application to identification of galactose regulatory pathway

We propose a novel technique that constructs gene regulatory networks from DNA microarray data and gene-protein databases and then applies Mason rule to systematically search for the most dominant regulators of the network. The algorithm then recommends the identified dominant regulator genes as the best candidates for future knock-out experiments. Actively choosing the genes for knock-out experiments allows optimal perturbation of the pathway and therefore produces the most informative DNA microarray data for pathway identification purposes. This approach is more practically advantageous in analysis of large pathways where the time and cost of DNA microarray data experiments can be reduced using the proposed optimal experiment design. The proposed method was successfully tested on the galactose regulatory network.     

Along signal paths: an empirical gene set approach exploiting pathway topology

Gene set analysis using biological pathways has become a widely used statistical approach for gene expression analysis. A biological pathway can be represented through a graph where genes and their interactions are, respectively, nodes and edges of the graph. From a biological point of view only some portions of a pathway are expected to be altered; however, few methods using pathway topology have been proposed and none of them tries to identify the signal paths, within a pathway, mostly involved in the biological problem. Here, we present a novel algorithm for pathway analysis clipper, that tries to fill in this gap. clipper implements a two-step empirical approach based on the exploitation of graph decomposition into a junction tree to reconstruct the most relevant signal path. In the first step clipper selects significant pathways according to statistical tests on the means and the concentration matrices of the graphs derived from pathway topologies. Then, it identifies within these pathways the signal paths having the greatest association with a specific phenotype. We test our approach on simulated and two real expression datasets. Our results demonstrate the efficacy of clipper in the identification of signal transduction paths totally coherent with the biological problem.     

Identification of novel metastasis suppressor signaling pathways for breast cancer

Cancer lethality is mainly caused by metastasis. Therefore, understanding the nature of the genes involved in this process has become a priority. Given the heterogeneity of mutations in cancer cells, considerable focus has been directed toward characterizing metastasis genes in the context of relevant signaling pathways rather than treating genes as independent and equal entities. One signaling cascade implicated in the regulation of cell growth, invasion and metastasis is the MAP kinase pathway. Raf kinase inhibitory protein (RKIP) functions as an inhibitor of the MAP kinase pathway and is a metastasis suppressor in different cancer models. By utilizing statistical analysis of clinical data integrated with experimental validation, we recently identified components of the RKIP signaling pathway relevant to breast cancer metastasis. Using the RKIP pathway as an example, we show how prior biological knowledge can be efficiently combined with genome-wide patient data to identify gene regulatory mechanisms that control metastasis.     

Pathway recognition and augmentation by computational analysis of microarray expression data

MOTIVATION: We present a system, QPACA (Quantitative Pathway Analysis in Cancer) for analysis of biological data in the context of pathways. QPACA supports data visualization and both fine- and coarse-grained specifications, but, more importantly, addresses the problems of pathway recognition and pathway augmentation. RESULTS: Given a set of genes hypothesized to be part of a pathway or a coordinated process, QPACA is able to reliably distinguish true pathways from non-pathways using microarray expression data. Relying on the observation that only some of the experiments within a dataset are relevant to a specific biochemical pathway, QPACA automates selection of this subset using an optimization procedure. We present data on all human and yeast pathways found in the KEGG pathway database. In 117 out of 191 cases (61%), QPACA was able to correctly identify these positive cases as bona fide pathways with p-values measured using rigorous permutation analysis. Success in recognizing pathways was dependent on pathway size, with the largest quartile of pathways yielding 83% success. In cross-validation tests of pathway membership prediction, QPACA was able to yield enrichments for predicted pathway genes over random genes at rates of 2-fold or better the majority of the time, with rates of 10-fold or better 10-20% of the time. AVAILABILITY: The software is available for academic research use free of charge by email request. SUPPLEMENTARY INFORMATION: Data used in the paper may be downloaded from http://www.jainlab.org/downloads.html     

Graphical modeling for gene set analysis: A critical appraisal

Current demand for understanding the behavior of groups of related genes, combined with the greater availability of data, has led to an increased focus on statistical methods in gene set analysis. In this paper, we aim to perform a critical appraisal of the methodology based on graphical models developed in Massa et al. ( 2010 ) that uses pathway signaling networks as a starting point to develop statistically sound procedures for gene set analysis. We pay attention to the potential of the methodology with respect to the organizational aspects of dealing with such complex but highly informative starting structures, that is pathways. We focus on three themes: the translation of a biological pathway into a graph suitable for modeling, the role of shrinkage when more genes than samples are obtained, the evaluation of respondence of the statistical models to the biological expectations. To study the impact of shrinkage, two simulation studies will be run. To evaluate the biological expectation we will use data from a network with known behavior that offer the possibility of carrying out a realistic check of respondence of the model to changes in the experimental conditions.     

Pathway Analysis Based on a Genome-Wide Association Study of Polycystic Ovary Syndrome

Background

Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders in women of reproductive age, and it is affected by both environmental and genetic factors. Although the genetic component of PCOS is evident, studies aiming to identify susceptibility genes have shown controversial results. This study conducted a pathway-based analysis using a dataset obtained through a genome-wide association study (GWAS) to elucidate the biological pathways that contribute to PCOS susceptibility and the associated genes.

Methods

We used GWAS data on 636,797 autosomal single nucleotide polymorphisms (SNPs) from 1,221 individuals (432 PCOS patients and 789 controls) for analysis. A pathway analysis was conducted using meta-analysis gene-set enrichment of variant associations (MAGENTA). Top-ranking pathways or gene sets associated with PCOS were identified, and significant genes within the pathways were analyzed.

Results

The pathway analysis of the GWAS dataset identified significant pathways related to oocyte meiosis and the regulation of insulin secretion by acetylcholine and free fatty acids (all nominal gene-set enrichment analysis (GSEA) P-values < 0.05). In addition, INS, GNAQ, STXBP1, PLCB3, PLCB2, SMC3 and PLCZ1 were significant genes observed within the biological pathways (all gene P-values < 0.05).

Conclusions

By applying MAGENTA pathway analysis to PCOS GWAS data, we identified significant pathways and candidate genes involved in PCOS. Our findings may provide new leads for understanding the mechanisms underlying the development of PCOS.     

Survival associated pathway identification with group <Emphasis Type="Italic">L</Emphasis><Subscript><Emphasis Type="Italic">p</Emphasis></Subscript> penalized global AUC maximization

It has been demonstrated that genes in a cell do not act independently. They interact with one another to complete certain biological processes or to implement certain molecular functions. How to incorporate biological pathways or functional groups into the model and identify survival associated gene pathways is still a challenging problem. In this paper, we propose a novel iterative gradient based method for survival analysis with group L _p penalized global AUC summary maximization. Unlike LASSO, L _p (p < 1) (with its special implementation entitled adaptive LASSO) is asymptotic unbiased and has oracle properties [1]. We first extend L _p for individual gene identification to group L _p penalty for pathway selection, and then develop a novel iterative gradient algorithm for penalized global AUC summary maximization (IGGAUCS). This method incorporates the genetic pathways into global AUC summary maximization and identifies survival associated pathways instead of individual genes. The tuning parameters are determined using 10-fold cross validation with training data only. The prediction performance is evaluated using test data. We apply the proposed method to survival outcome analysis with gene expression profile and identify multiple pathways simultaneously. Experimental results with simulation and gene expression data demonstrate that the proposed procedures can be used for identifying important biological pathways that are related to survival phenotype and for building a parsimonious model for predicting the survival times.     

Nonparametric pathway-based regression models for analysis of genomic data

High-throughout genomic data provide an opportunity for identifying pathways and genes that are related to various clinical phenotypes. Besides these genomic data, another valuable source of data is the biological knowledge about genes and pathways that might be related to the phenotypes of many complex diseases. Databases of such knowledge are often called the metadata. In microarray data analysis, such metadata are currently explored in post hoc ways by gene set enrichment analysis but have hardly been utilized in the modeling step. We propose to develop and evaluate a pathway-based gradient descent boosting procedure for nonparametric pathways-based regression (NPR) analysis to efficiently integrate genomic data and metadata. Such NPR models consider multiple pathways simultaneously and allow complex interactions among genes within the pathways and can be applied to identify pathways and genes that are related to variations of the phenotypes. These methods also provide an alternative to mediating the problem of a large number of potential interactions by limiting analysis to biologically plausible interactions between genes in related pathways. Our simulation studies indicate that the proposed boosting procedure can indeed identify relevant pathways. Application to a gene expression data set on breast cancer distant metastasis identified that Wnt, apoptosis, and cell cycle-regulated pathways are more likely related to the risk of distant metastasis among lymph-node-negative breast cancer patients. Results from analysis of other two breast cancer gene expression data sets indicate that the pathways of Metalloendopeptidases (MMPs) and MMP inhibitors, as well as cell proliferation, cell growth, and maintenance are important to breast cancer relapse and survival. We also observed that by incorporating the pathway information, we achieved better prediction for cancer recurrence.     

Pathway Miner: extracting gene association networks from molecular pathways for predicting the biological significance of gene expression microarray data

We have developed a web-based system (Pathway Miner) for visualizing gene expression profiles in the context of biological pathways. Pathway Miner catalogs genes based on their role in metabolic, cellular and regulatory pathways. A Fisher exact test is provided as an option to rank pathways. The genes are mapped onto pathways and gene product association networks are extracted for genes that co-occur in pathways. The networks can be filtered for analysis based on user-selected options. AVAILABILITY: Pathway Miner is a freely available web accessible tool at http://www.biorag.org/pathway.html     

The correlation of gene expression and co-regulated gene patterns in characteristic KEGG pathways

There is great interest in chromosome- and pathway-based techniques for genomics data analysis in the current work in order to understand the mechanism of disease. However, there are few studies addressing the abilities of machine learning methods in incorporating pathway information for analyzing microarray data. In this paper, we identified the characteristic pathways by combining the classification error rates of out-of-bag (OOB) in random forests with pathways information. At each characteristic pathway, the correlation of gene expression was studied and the co-regulated gene patterns in different biological conditions were mined by Mining Attribute Profile (MAP) algorithm. The discovered co-regulated gene patterns were clustered by the average-linkage hierarchical clustering technique. The results showed that the expression of genes at the same characteristic pathway were approximate. Furthermore, two characteristic pathways were discovered to present co-regulated gene patterns in which one contained 108 patterns and the other contained one pattern. The results of cluster analysis showed that the smallest similarity coefficient of clusters was more than 0.623, which indicated that the co-regulated patterns in different biological conditions were more approximate at the same characteristic pathway. The methods discussed in this paper can provide additional insight into the study of microarray data.