Abstract Motivation: Anti-cancer drugs may elicit resistance or sensitivity through mechanisms which involve several genomic layers. Nevertheless, we have demonstrated that gene expression contains most of the predictive capacity compared to the remaining omic data types. Unfortunately, this comes at a price: gene expression biomarkers are often hard to interpret and show poor robustness. Results: To capture the best of both worlds, i.e. the accuracy of gene expression and the robustness of other genomic levels, such as mutations, copy-number or methylation, we developed Percolate, a computational approach which extracts the joint signal between gene expression and the other omic data types. We developed an out-of-sample extension of Percolate which allows predictions on unseen samples without the necessity to recompute the joint signal on all data. We employed Percolate to extract the joint signal between gene expression and either mutations, copy-number or methylation, and used the out-of sample extension to perform response prediction on unseen samples. We showed that the joint signal recapitulates, and sometimes exceeds, the predictive performance achieved with each data type individually. Importantly, molecular signatures created by Percolate do not require gene expression to be evaluated, rendering them suitable to clinical applications where only one data type is available. Availability: Percolate is available as a Python 3.7 package and the scripts to reproduce the results are available here .

MOTIVATION: In many high-dimensional prediction or classification tasks, complementary data on the features are available, e.g. prior biological knowledge on (epi)genetic markers. Here we consider tasks with numerical prior information that provide an insight into the importance (weight) and the direction (sign) of the feature effects, e.g. regression coefficients from previous studies. RESULTS: We propose an approach for integrating multiple sources of such prior information into penalized regression. If suitable co-data are available, this improves the predictive performance, as shown by simulation and application. AVAILABILITY AND IMPLEMENTATION: The proposed method is implemented in the R package transreg (https://github.com/lcsb-bds/transreg, https://cran.r-project.org/package=transreg).

BACKGROUND: High-dimensional prediction considers data with more variables than samples. Generic research goals are to find the best predictor or to select variables. Results may be improved by exploiting prior information in the form of co-data, providing complementary data not on the samples, but on the variables. We consider adaptive ridge penalised generalised linear and Cox models, in which the variable-specific ridge penalties are adapted to the co-data to give a priori more weight to more important variables. The R-package ecpc originally accommodated various and possibly multiple co-data sources, including categorical co-data, i.e. groups of variables, and continuous co-data. Continuous co-data, however, were handled by adaptive discretisation, potentially inefficiently modelling and losing information. As continuous co-data such as external p values or correlations often arise in practice, more generic co-data models are needed. RESULTS: Here, we present an extension to the method and software for generic co-data models, particularly for continuous co-data. At the basis lies a classical linear regression model, regressing prior variance weights on the co-data. Co-data variables are then estimated with empirical Bayes moment estimation. After placing the estimation procedure in the classical regression framework, extension to generalised additive and shape constrained co-data models is straightforward. Besides, we show how ridge penalties may be transformed to elastic net penalties. In simulation studies we first compare various co-data models for continuous co-data from the extension to the original method. Secondly, we compare variable selection performance to other variable selection methods. The extension is faster than the original method and shows improved prediction and variable selection performance for non-linear co-data relations. Moreover, we demonstrate use of the package in several genomics examples throughout the paper. CONCLUSIONS: The R-package ecpc accommodates linear, generalised additive and shape constrained additive co-data models for the purpose of improved high-dimensional prediction and variable selection. The extended version of the package as presented here (version number 3.1.1 and higher) is available on ( https://cran.r-project.org/web/packages/ecpc/ ).

MOTIVATION: Machine learning in the biomedical sciences should ideally provide predictive and interpretable models. When predicting outcomes from clinical or molecular features, applied researchers often want to know which features have effects, whether these effects are positive or negative and how strong these effects are. Regression analysis includes this information in the coefficients but typically renders less predictive models than more advanced machine learning techniques. RESULTS: Here, we propose an interpretable meta-learning approach for high-dimensional regression. The elastic net provides a compromise between estimating weak effects for many features and strong effects for some features. It has a mixing parameter to weight between ridge and lasso regularization. Instead of selecting one weighting by tuning, we combine multiple weightings by stacking. We do this in a way that increases predictivity without sacrificing interpretability. AVAILABILITY AND IMPLEMENTATION: The R package starnet is available on GitHub (https://github.com/rauschenberger/starnet) and CRAN (https://CRAN.R-project.org/package=starnet).

MOTIVATION: Cell lines and patient-derived xenografts (PDXs) have been used extensively to understand the molecular underpinnings of cancer. While core biological processes are typically conserved, these models also show important differences compared to human tumors, hampering the translation of findings from pre-clinical models to the human setting. In particular, employing drug response predictors generated on data derived from pre-clinical models to predict patient response remains a challenging task. As very large drug response datasets have been collected for pre-clinical models, and patient drug response data are often lacking, there is an urgent need for methods that efficiently transfer drug response predictors from pre-clinical models to the human setting. RESULTS: We show that cell lines and PDXs share common characteristics and processes with human tumors. We quantify this similarity and show that a regression model cannot simply be trained on cell lines or PDXs and then applied on tumors. We developed PRECISE, a novel methodology based on domain adaptation that captures the common information shared amongst pre-clinical models and human tumors in a consensus representation. Employing this representation, we train predictors of drug response on pre-clinical data and apply these predictors to stratify human tumors. We show that the resulting domain-invariant predictors show a small reduction in predictive performance in the pre-clinical domain but, importantly, reliably recover known associations between independent biomarkers and their companion drugs on human tumors. AVAILABILITY AND IMPLEMENTATION: PRECISE and the scripts for running our experiments are available on our GitHub page (https://github.com/NKI-CCB/PRECISE). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Our aim is to improve omics based prediction and feature selection using multiple sources of auxiliary information: co-data. Adaptive group regularized ridge regression (GRridge) was proposed to achieve this by estimating additional group-based penalty parameters through an empirical Bayes method at a low computational cost. We illustrate the GRridge method and software on RNA sequencing datasets. The method boosts the performance of an ordinary ridge regression and outperforms other classifiers. Post-hoc feature selection maintains the predictive ability of the classifier with far fewer markers. AVAILABILITY AND IMPLEMENTATION: GRridge is an R package that includes a vignette. It is freely available at ( https://bioconductor.org/packages/GRridge/ ). All information and R scripts used in this study, including those on retrieval and processing of the co-data, are available from http://github.com/markvdwiel/GRridgeCodata . CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Prediction in high dimensional settings is difficult due to the large number of variables relative to the sample size. We demonstrate how auxiliary 'co-data' can be used to improve the performance of a Random Forest in such a setting. RESULTS: Co-data are incorporated in the Random Forest by replacing the uniform sampling probabilities that are used to draw candidate variables by co-data moderated sampling probabilities. Co-data here are defined as any type information that is available on the variables of the primary data, but does not use its response labels. These moderated sampling probabilities are, inspired by empirical Bayes, learned from the data at hand. We demonstrate the co-data moderated Random Forest (CoRF) with two examples. In the first example we aim to predict the presence of a lymph node metastasis with gene expression data. We demonstrate how a set of external p-values, a gene signature, and the correlation between gene expression and DNA copy number can improve the predictive performance. In the second example we demonstrate how the prediction of cervical (pre-)cancer with methylation data can be improved by including the location of the probe relative to the known CpG islands, the number of CpG sites targeted by a probe, and a set of p-values from a related study. CONCLUSION: The proposed method is able to utilize auxiliary co-data to improve the performance of a Random Forest.

BACKGROUND: Testing for association between RNA-Seq and other genomic data is challenging due to high variability of the former and high dimensionality of the latter. RESULTS: Using the negative binomial distribution and a random-effects model, we develop an omnibus test that overcomes both difficulties. It may be conceptualised as a test of overall significance in regression analysis, where the response variable is overdispersed and the number of explanatory variables exceeds the sample size. CONCLUSIONS: The proposed test can detect genetic and epigenetic alterations that affect gene expression. It can examine complex regulatory mechanisms of gene expression. The R package globalSeq is available from Bioconductor.

BACKGROUND: In genomics, hierarchical clustering (HC) is a popular method for grouping similar samples based on a distance measure. HC algorithms do not actually create clusters, but compute a hierarchical representation of the data set. Usually, a fixed height on the HC tree is used, and each contiguous branch of samples below that height is considered a separate cluster. Due to the fixed-height cutting, those clusters may not unravel significant functional coherence hidden deeper in the tree. Besides that, most existing approaches do not make use of available clinical information to guide cluster extraction from the HC. Thus, the identified subgroups may be difficult to interpret in relation to that information. RESULTS: We develop a novel framework for decomposing the HC tree into clusters by semi-supervised piecewise snipping. The framework, called guided piecewise snipping, utilizes both molecular data and clinical information to decompose the HC tree into clusters. It cuts the given HC tree at variable heights to find a partition (a set of non-overlapping clusters) which does not only represent a structure deemed to underlie the data from which HC tree is derived, but is also maximally consistent with the supplied clinical data. Moreover, the approach does not require the user to specify the number of clusters prior to the analysis. Extensive results on simulated and multiple medical data sets show that our approach consistently produces more meaningful clusters than the standard fixed-height cut and/or non-guided approaches. CONCLUSIONS: The guided piecewise snipping approach features several novelties and advantages over existing approaches. The proposed algorithm is generic, and can be combined with other algorithms that operate on detected clusters. This approach represents an advancement in several regards: (1) a piecewise tree snipping framework that efficiently extracts clusters by snipping the HC tree possibly at variable heights while preserving the HC tree structure; (2) a flexible implementation allowing a variety of data types for both building and snipping the HC tree, including patient follow-up data like survival as auxiliary information. The data sets and R code are provided as supplementary files. The proposed method is available from Bioconductor as the R-package HCsnip.

BACKGROUND: To determine which changes in the host cell genome are crucial for cervical carcinogenesis, a longitudinal in vitro model system of HPV-transformed keratinocytes was profiled in a genome-wide manner. Four cell lines affected with either HPV16 or HPV18 were assayed at 8 sequential time points for gene expression (mRNA) and gene copy number (DNA) using high-resolution microarrays. Available methods for temporal differential expression analysis are not designed for integrative genomic studies. RESULTS: Here, we present a method that allows for the identification of differential gene expression associated with DNA copy number changes over time. The temporal variation in gene expression is described by a generalized linear mixed model employing low-rank thin-plate splines. Model parameters are estimated with an empirical Bayes procedure, which exploits integrated nested Laplace approximation for fast computation. Iteratively, posteriors of hyperparameters and model parameters are estimated. The empirical Bayes procedure shrinks multiple dispersion-related parameters. Shrinkage leads to more stable estimates of the model parameters, better control of false positives and improvement of reproducibility. In addition, to make estimates of the DNA copy number more stable, model parameters are also estimated in a multivariate way using triplets of features, imposing a spatial prior for the copy number effect. CONCLUSION: With the proposed method for analysis of time-course multilevel molecular data, more profound insight may be gained through the identification of temporal differential expression induced by DNA copy number abnormalities. In particular, in the analysis of an integrative oncogenomics study with a time-course set-up our method finds genes previously reported to be involved in cervical carcinogenesis. Furthermore, the proposed method yields improvements in sensitivity, specificity and reproducibility compared to existing methods. Finally, the proposed method is able to handle count (RNAseq) data from time course experiments as is shown on a real data set.

BACKGROUND: Complex designs are common in (observational) clinical studies. Sequencing data for such studies are produced more and more often, implying challenges for the analysis, such as excess of zeros, presence of random effects and multi-parameter inference. Moreover, when sample sizes are small, inference is likely to be too liberal when, in a Bayesian setting, applying a non-appropriate prior or to lack power when not carefully borrowing information across features. RESULTS: We show on microRNA sequencing data from a clinical cancer study how our software ShrinkBayes tackles the aforementioned challenges. In addition, we illustrate its comparatively good performance on multi-parameter inference for groups using a data-based simulation. Finally, in the small sample size setting, we demonstrate its high power and improved FDR estimation by use of Gaussian mixture priors that include a point mass. CONCLUSION: ShrinkBayes is a versatile software package for the analysis of count-based sequencing data, which is particularly useful for studies with small sample sizes or complex designs.

SUMMARY: DNA copy number and mRNA expression are commonly used data types in cancer studies. Available software for integrative analysis arbitrarily fixes the parametric form of the association between the two molecular levels and hence offers no opportunities for modelling it. We present a new tool for flexible modelling of this association. PLRS uses a wide class of interpretable models including popular ones and incorporates prior biological knowledge. It is capable to identify the gene-specific type of relationship between gene copy number and mRNA expression. Moreover, it tests the strength of the association and provides confidence intervals. We illustrate PLRS using glioblastoma data from The Cancer Genome Atlas. AVAILABILITY AND IMPLEMENTATION: PLRS is implemented as an R package and available from Bioconductor (as of version 2.12; http://bioconductor.org). Additional code for parallel computations is available as Supplementary Material. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: An increasing number of genomic studies interrogating more than one molecular level is published. Bioinformatics follows biological practice, and recent years have seen a surge in methodology for the integrative analysis of genomic data. Often such analyses require knowledge of which elements of one platform link to those of another. Although important, many integrative analyses do not or insufficiently detail the matching of the platforms. RESULTS: We describe, illustrate and discuss six matching procedures. They are implemented in the R-package sigaR (available from Bioconductor). The principles underlying the presented matching procedures are generic, and can be combined to form new matching approaches or be applied to the matching of other platforms. Illustration of the matching procedures on a variety of data sets reveals how the procedures differ in the use of the available data, and may even lead to different results for individual genes. CONCLUSIONS: Matching of data from multiple genomics platforms is an important preprocessing step for many integrative bioinformatic analysis, for which we present six generic procedures, both old and new. They have been implemented in the R-package sigaR, available from Bioconductor.

Analysis of DNA copy number profiles requires methods tailored to the specific nature of these data. The number of available data analysis methods has grown enormously in the last 5 years. We discuss the typical characteristics of DNA copy number data, as measured by microarray technology and review the extensive literature on preprocessing methods such as segmentation and calling. Subsequently, the focus narrows to applications of DNA copy number in cancer, in particular, several downstream analyses of multi-sample data sets such as testing, clustering and classification. Finally, we look ahead: what should we prepare for and which methodology-related topics may deserve attention in the near future?

BACKGROUND: Combining clinical and molecular data types may potentially improve prediction accuracy of a classifier. However, currently there is a shortage of effective and efficient statistical and bioinformatic tools for true integrative data analysis. Existing integrative classifiers have two main disadvantages: First, coarse combination may lead to subtle contributions of one data type to be overshadowed by more obvious contributions of the other. Second, the need to measure both data types for all patients may be both unpractical and (cost) inefficient. RESULTS: We introduce a novel classification method, a stepwise classifier, which takes advantage of the distinct classification power of clinical data and high-dimensional molecular data. We apply classification algorithms to two data types independently, starting with the traditional clinical risk factors. We only turn to relatively expensive molecular data when the uncertainty of prediction result from clinical data exceeds a predefined limit. Experimental results show that our approach is adaptive: the proportion of samples that needs to be re-classified using molecular data depends on how much we expect the predictive accuracy to increase when re-classifying those samples. CONCLUSIONS: Our method renders a more cost-efficient classifier that is at least as good, and sometimes better, than one based on clinical or molecular data alone. Hence our approach is not just a classifier that minimizes a particular loss function. Instead, it aims to be cost-efficient by avoiding molecular tests for a potentially large subgroup of individuals; moreover, for these individuals a test result would be quickly available, which may lead to reduced waiting times (for diagnosis) and hence lower the patients distress. Stepwise classification is implemented in R-package stepwiseCM and available at the Bioconductor website.

BACKGROUND: Determining a suitable sample size is an important step in the planning of microarray experiments. Increasing the number of arrays gives more statistical power, but adds to the total cost of the experiment. Several approaches for sample size determination have been developed for expression array studies, but so far none has been proposed for array comparative genomic hybridization (aCGH). RESULTS: Here we explore power calculations for aCGH experiments comparing two groups. In a pilot experiment CGHpower estimates the biological diversity between groups and provides a statistical framework for estimating average power as a function of sample size. As the method requires pilot data, it can be used either in the planning stage of larger studies or in estimating the power achieved in past experiments. CONCLUSIONS: The proposed method relies on certain assumptions. According to our evaluation with public and simulated data sets, they do not always hold true. Violation of the assumptions typically leads to unreliable sample size estimates. Despite its limitations, this method is, at least to our knowledge, the only one currently available for performing sample size calculations in the context of aCGH. Moreover, the implementation of the method provides diagnostic plots that allow critical assessment of the assumptions on which it is based and hence on the feasibility and reliability of the sample size calculations in each case.The CGHpower web application and the program outputs from evaluation data sets can be freely accessed at http://www.cangem.org/cghpower/

MOTIVATION: Many high-resolution array comparative genomic hybridization tumor profiles contain a wave bias, which makes accurate detection of breakpoints in such profiles more difficult. RESULTS: An efficient and highly effective algorithm that largely removes the wave bias from tumor profiles by regressing the tumor profile data on data of profiles from the clinical genetics practice. Results are illustrated on two independent datasets. The algorithm is shown to be robust against the presence of true copy number aberrations. Moreover, the smoothed profiles are able to recapitulate the aberration location and signal for simulated tumor profiles. AVAILABILITY: Easy-to-use R scripts, user instructions and data are available from http://www.few.vu.nl/~mavdwiel/nowaves.html. SUPPLEMENTARY INFORMATION: Supplementary information are available at Bioinformatics online.

BACKGROUND: We consider effects of dependence among variables of high-dimensional data in multiple hypothesis testing problems, in particular the False Discovery Rate (FDR) control procedures. Recent simulation studies consider only simple correlation structures among variables, which is hardly inspired by real data features. Our aim is to systematically study effects of several network features like sparsity and correlation strength by imposing dependence structures among variables using random correlation matrices. RESULTS: We study the robustness against dependence of several FDR procedures that are popular in microarray studies, such as Benjamin-Hochberg FDR, Storey's q-value, SAM and resampling based FDR procedures. False Non-discovery Rates and estimates of the number of null hypotheses are computed from those methods and compared. Our simulation study shows that methods such as SAM and the q-value do not adequately control the FDR to the level claimed under dependence conditions. On the other hand, the adaptive Benjamini-Hochberg procedure seems to be most robust while remaining conservative. Finally, the estimates of the number of true null hypotheses under various dependence conditions are variable. CONCLUSION: We discuss a new method for efficient guided simulation of dependent data, which satisfy imposed network constraints as conditional independence structures. Our simulation set-up allows for a structural study of the effect of dependencies on multiple testing criterions and is useful for testing a potentially new method on pi0 or FDR estimation in a dependency context.

UNLABELLED: CGHcall achieves high calling accuracy for array CGH data by effective use of breakpoint information from segmentation and by inclusion of several biological concepts that are ignored by existing algorithms. The algorithm is validated for simulated and verified real array CGH data. By incorporating more than three classes, CGHcall improves detection of single copy gains and amplifications. Moreover, it allows effective inclusion of chromosome arm information. AVAILABILITY: An R-package (GUI), a manual and an example data set are available at http://www.few.vu.nl/~mavdwiel/CGHcall.html. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Summary: We compute P-values, based on the Wilcoxon test with ties, to compare two conditions with array comparative genomic hybridization data, and we provide a simple interface to export and plot these P-values. Availability: CGHMultiArray is freely available at http://www.win.tue.nl/~markvdw/CGHMultiArray.html Contact: [email protected] Supplementary information: Programs, the manual and supplementary information are available on the website.