State-of-the-art rare variant association testing methods aggregate the contribution of rare variants in biologically relevant genomic regions to boost statistical power. However, testing single genes separately does not consider the complex interaction landscape of genes, nor the downstream effects of non-synonymous variants on protein structure and function. Here we present the NETwork Propagation-based Assessment of Genetic Events (NETPAGE), an integrative approach aimed at investigating the biological pathways through which rare variation results in complex disease phenotypes. We applied NETPAGE to sporadic, late-onset Alzheimer's disease (AD), using whole-genome sequencing from the AD Neuroimaging Initiative (ADNI) cohort, as well as whole-exome sequencing from the AD Sequencing Project (ADSP). NETPAGE is based on network propagation, a framework that models information flow on a graph and simulates the percolation of genetic variation through tissue-specific gene interaction networks. The result of network propagation is a set of smoothed gene scores that can be tested for association with disease status through sparse regression. The application of NETPAGE to AD enabled the identification of a set of connected genes whose smoothed variation profile was robustly associated to case-control status, based on gene interactions in the hippocampus. Additionally, smoothed scores significantly correlated with risk of conversion to AD in Mild Cognitive Impairment (MCI) subjects. Lastly, we investigated tissue-specific transcriptional dysregulation of the core genes in two independent RNA-seq datasets, as well as significant enrichments in terms of gene sets with known connections to AD. We present a framework that enables enhanced genetic association testing for a wide range of traits, diseases, and sample sizes.

MOTIVATION: High-throughput-sequencing (HTS) technologies are the method of choice for screening the human genome for rare sequence variants causing susceptibility to complex diseases. Unfortunately, preparation of samples for a large number of individuals is still very cost- and labor intensive. Thus, recently, screens for rare sequence variants were carried out in samples of pooled DNA, in which equimolar amounts of DNA from multiple individuals are mixed prior to sequencing with HTS. The resulting sequence data, however, poses a bioinformatics challenge: the discrimination of sequencing errors from real sequence variants present at a low frequency in the DNA pool. RESULTS: Our method vipR uses data from multiple DNA pools in order to compensate for differences in sequencing error rates along the sequenced region. More precisely, instead of aiming at discriminating sequence variants from sequencing errors, vipR identifies sequence positions that exhibit significantly different minor allele frequencies in at least two DNA pools using the Skellam distribution. The performance of vipR was compared with three other models on data from a targeted resequencing study of the TMEM132D locus in 600 individuals distributed over four DNA pools. Performance of the methods was computed on SNPs that were also genotyped individually using a MALDI-TOF technique. On a set of 82 sequence variants, vipR achieved an average sensitivity of 0.80 at an average specificity of 0.92, thus outperforming the reference methods by at least 0.17 in specificity at comparable sensitivity. AVAILABILITY: The code of vipR is freely available via: http://sourceforge.net/projects/htsvipr/ CONTACT: [email protected].

MOTIVATION: In life sciences, interpretability of machine learning models is as important as their prediction accuracy. Linear models are probably the most frequently used methods for assessing feature relevance, despite their relative inflexibility. However, in the past years effective estimators of feature relevance have been derived for highly complex or non-parametric models such as support vector machines and RandomForest (RF) models. Recently, it has been observed that RF models are biased in such a way that categorical variables with a large number of categories are preferred. RESULTS: In this work, we introduce a heuristic for normalizing feature importance measures that can correct the feature importance bias. The method is based on repeated permutations of the outcome vector for estimating the distribution of measured importance for each variable in a non-informative setting. The P-value of the observed importance provides a corrected measure of feature importance. We apply our method to simulated data and demonstrate that (i) non-informative predictors do not receive significant P-values, (ii) informative variables can successfully be recovered among non-informative variables and (iii) P-values computed with permutation importance (PIMP) are very helpful for deciding the significance of variables, and therefore improve model interpretability. Furthermore, PIMP was used to correct RF-based importance measures for two real-world case studies. We propose an improved RF model that uses the significant variables with respect to the PIMP measure and show that its prediction accuracy is superior to that of other existing models. AVAILABILITY: R code for the method presented in this article is available at http://www.mpi-inf.mpg.de/ approximately altmann/download/PIMP.R CONTACT: [email protected], [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: As there exists no cure or vaccine for the infection with human immunodeficiency virus (HIV), the standard approach to treating HIV patients is to repeatedly administer different combinations of several antiretroviral drugs. Because of the large number of possible drug combinations, manually finding a successful regimen becomes practically impossible. This presents a major challenge for HIV treatment. The application of machine learning methods for predicting virological responses to potential therapies is a possible approach to solving this problem. However, due to evolving trends in treating HIV patients the available clinical datasets have a highly unbalanced representation, which might negatively affect the usefulness of derived statistical models. RESULTS: This article presents an approach that tackles the problem of predicting virological response to combination therapies by learning a separate logistic regression model for each therapy. The models are fitted by using not only the data from the target therapy but also the information from similar therapies. For this purpose, we introduce and evaluate two different measures of therapy similarity. The models are also able to incorporate phenotypic knowledge on the therapy outcomes through a Gaussian prior. With our approach we balance the uneven therapy representation in the datasets and produce higher quality models for therapies with very few training samples. According to the results from the computational experiments our therapy similarity model performs significantly better than training separate models for each therapy by using solely their examples. Furthermore, the model's performance is as good as an approach that encodes therapy information in the input feature space with the advantage of delivering better results for therapies with very few training samples. AVAILABILITY: Code of the efficient logistic regression is available from http://www.mpi-inf.mpg.de/%7Ejasmina/fastLogistic.zip.

MOTIVATION: Optimizing HIV therapies is crucial since the virus rapidly develops mutations to evade drug pressure. Recent studies have shown that genotypic information might not be sufficient for the design of therapies and that other clinical and demographical factors may play a role in therapy failure. This study is designed to assess the improvement in prediction achieved when such information is taken into account. We use these factors to generate a prediction engine using a variety of machine learning methods and to determine which clinical conditions are most misleading in terms of predicting the outcome of a therapy. RESULTS: Three different machine learning techniques were used: generative-discriminative method, regression with derived evolutionary features, and regression with a mixture of effects. All three methods had similar performances with an area under the receiver operating characteristic curve (AUC) of 0.77. A set of three similar engines limited to genotypic information only achieved an AUC of 0.75. A straightforward combination of the three engines consistently improves the prediction, with significantly better prediction when the full set of features is employed. The combined engine improves on predictions obtained from an online state-of-the-art resistance interpretation system. Moreover, engines tend to disagree more on the outcome of failure therapies than regarding successful ones. Careful analysis of the differences between the engines revealed those mutations and drugs most closely associated with uncertainty of the therapy outcome. AVAILABILITY: The combined prediction engine will be available from July 2008, see http://engine.euresist.org.

Abstract Summary: In genetics, many evolutionary pathways can be modeled by the ordered accumulation of permanent changes. Mixture models of mutagenetic trees have been used to describe disease progression in cancer and in HIV. In cancer, progression is modeled by the accumulation of chromosomal gains and losses in tumor cells; in HIV, the accumulation of drug resistance-associated mutations in the viral genome is known to be associated with disease progression. From such evolutionary models, genetic progression scores can be derived that assign measures for the disease state to single patients. Rtreemix is an R package for estimating mixture models of evolutionary pathways from observed cross-sectional data and for estimating associated genetic progression scores. The package also provides extended functionality for estimating confidence intervals for estimated model parameters and for evaluating the stability of the estimated evolutionary mixture models. Availability: Rtreemix is an R package that is freely available from the Bioconductor project at http://www.bioconductor.org and runs on Linux and Windows. Contact: [email protected]