Pathway-Based Kernel Boosting for the Analysis of Genome-Wide Association Studies

Review of Genetic Variation as a Predictive Biomarker for Chronic Graft-Versus-Host-Disease After Allogeneic Stem Cell Transplantation

Chronic graft-versus-host disease (cGvHD) is one of the major complications of allogeneic stem cell transplantation (HSCT). cGvHD is an autoimmune-like disorder affecting multiple organs and involves a dermatological rash, tissue inflammation and fibrosis. The incidence of cGvHD has been reported to be as high as 30% to 60% and there are currently no reliable tools for predicting the occurrence of cGvHD. There is therefore an important unmet clinical need for predictive biomarkers. The present review summarizes the state of the art for genetic variation as a predictive biomarker for cGvHD. We discuss three different modes of action for genetic variation in transplantation: genetic associations, genetic matching, and pharmacogenetics. The results indicate that currently, there are no genetic polymorphisms or genetic tools that can be reliably used as validated biomarkers for predicting cGvHD. A number of recommendations for future studies can be drawn. The majority of studies to date have been under-powered and included too few patients and genetic markers. Like in all complex multifactorial diseases, large collaborative genome-level studies are now needed to achieve reliable and unbiased results. Some of the candidate genes, in particular, CTLA4, HSPE, IL1R1, CCR6, FGFR1OP, and IL10, and some non-HLA variants in the HLA gene region have been replicated to be associated with cGvHD risk in independent studies. These associations should now be confirmed in large well-characterized cohorts with fine mapping. Some patients develop cGvHD despite very extensive immunosuppression and other treatments, indicating that the current therapeutic regimens may not always be effective enough. Hence, more studies on pharmacogenetics are also required. Moreover, all of these studies should be adjusted for diagnostic and clinical features of cGvHD. We conclude that future studies should focus on modern genome-level tools, such as machine learning, polygenic risk scores and genome-wide association study-transcription meta-analyses, instead of focusing on just single variants. The risk of cGvHD may be related to the summary level of immunogenetic differences, or whole genome histocompatibility between each donor-recipient pair. As the number of genome-wide analyses in HSCT is increasing, we are approaching an era where there will be sufficient data to incorporate these approaches in the near future.

A Pathway-Based Kernel Boosting Method for Sample Classification Using Genomic Data

The analysis of cancer genomic data has long suffered "the curse of dimensionality." Sample sizes for most cancer genomic studies are a few hundreds at most while there are tens of thousands of genomic features studied. Various methods have been proposed to leverage prior biological knowledge, such as pathways, to more effectively analyze cancer genomic data. Most of the methods focus on testing marginal significance of the associations between pathways and clinical phenotypes. They can identify informative pathways but do not involve predictive modeling. In this article, we propose a Pathway-based Kernel Boosting (PKB) method for integrating gene pathway information for sample classification, where we use kernel functions calculated from each pathway as base learners and learn the weights through iterative optimization of the classification loss function. We apply PKB and several competing methods to three cancer studies with pathological and clinical information, including tumor grade, stage, tumor sites and metastasis status. Our results show that PKB outperforms other methods and identifies pathways relevant to the outcome variables.

Improving stability of prediction models based on correlated omics data by using network approaches

Building prediction models based on complex omics datasets such as transcriptomics, proteomics, metabolomics remains a challenge in bioinformatics and biostatistics. Regularized regression techniques are typically used to deal with the high dimensionality of these datasets. However, due to the presence of correlation in the datasets, it is difficult to select the best model and application of these methods yields unstable results. We propose a novel strategy for model selection where the obtained models also perform well in terms of overall predictability. Several three step approaches are considered, where the steps are 1) network construction, 2) clustering to empirically derive modules or pathways, and 3) building a prediction model incorporating the information on the modules. For the first step, we use weighted correlation networks and Gaussian graphical modelling. Identification of groups of features is performed by hierarchical clustering. The grouping information is included in the prediction model by using group-based variable selection or group-specific penalization. We compare the performance of our new approaches with standard regularized regression via simulations. Based on these results we provide recommendations for selecting a strategy for building a prediction model given the specific goal of the analysis and the sizes of the datasets. Finally we illustrate the advantages of our approach by application of the methodology to two problems, namely prediction of body mass index in the DIetary, Lifestyle, and Genetic determinants of Obesity and Metabolic syndrome study (DILGOM) and prediction of response of each breast cancer cell line to treatment with specific drugs using a breast cancer cell lines pharmacogenomics dataset.

An Update on Statistical Boosting in Biomedicine

Statistical boosting algorithms have triggered a lot of research during the last decade. They combine a powerful machine learning approach with classical statistical modelling, offering various practical advantages like automated variable selection and implicit regularization of effect estimates. They are extremely flexible, as the underlying base-learners (regression functions defining the type of effect for the explanatory variables) can be combined with any kind of loss function (target function to be optimized, defining the type of regression setting). In this review article, we highlight the most recent methodological developments on statistical boosting regarding variable selection, functional regression, and advanced time-to-event modelling. Additionally, we provide a short overview on relevant applications of statistical boosting in biomedicine.