PIMKL: Pathway-Induced Multiple Kernel Learning

Abstract

Reliable identification of molecular biomarkers is essential for accurate patient stratification. While state-of-the-art machine learning approaches for sample classification continue to push boundaries in terms of performance, most of these methods are not able to integrate different data types and lack generalization power, limiting their application in a clinical setting. Furthermore, many methods behave as black boxes, and we have very little understanding about the mechanisms that lead to the prediction. While opaqueness concerning machine behavior might not be a problem in deterministic domains, in health care, providing explanations about the molecular factors and phenotypes that are driving the classification is crucial to build trust in the performance of the predictive system. We propose Pathway-Induced Multiple Kernel Learning (PIMKL), a methodology to reliably classify samples that can also help gain insights into the molecular mechanisms that underlie the classification. PIMKL exploits prior knowledge in the form of a molecular interaction network and annotated gene sets, by optimizing a mixture of pathway-induced kernels using a Multiple Kernel Learning (MKL) algorithm, an approach that has demonstrated excellent performance in different machine learning applications. After optimizing the combination of kernels to predict a specific phenotype, the model provides a stable molecular signature that can be interpreted in the light of the ingested prior knowledge and that can be used in transfer learning tasks.

Predicting clinical drug response from model systems by non-linear subspace-based transfer learning

“…Learning the similarity matrix, e.g. using multiple kernel learning 61 or deep learning methods such as variational auto-encoders 62 , may also help increase performance.…”

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

“…It is conceivable that the genes interact with each other through an underlying interaction network and intuitively, genes in the hub should get more weights compared to genes on the periphery. With the network information available, it is possible to build more sensible kernel functions as base learners [17]. Third, the pathway databases only cover a subset of the input genes.…”

“…The kernel weights are estimated through optimization and can be considered as a measure of pathway importance. Multiple kernel methods have been used to integrate multi-pathway information or multi-omics data sets and have achieved state-of-the-art performance in predictions of various outcomes [15,16,17].…”

Highly Accurate Cancer Phenotype Prediction with AKLIMATE, a Stacked Kernel Learner Integrating Multimodal Genomic Data and Pathway Knowledge

“…Multiple Kernel Learning (MKL) approaches [3][59] [74] [27] [66] [50] can incorporate heterogeneous data by mapping each set of features through a kernel function and learning a linear combination of the kernel representations. Each kernel represents distinct samplesample similarities providing flexible and powerful transformations to access either explicit or implicit feature combinations.…”

“…To prevent overfitting, MKL methods generally include a regularization term -e.g., L 1 sparsity-inducing norm on the kernel weights [59] or the elastic net [74]. Prior knowledge can also be integrated by constructing individual kernels from a pathway's member features within each data type [29] [27] [66] [78] [50]. Indeed, MKL methods with prior knowledge integration [29] [14] have won several Dialogue on Reverse-Engineering Assessment and Methods (DREAM) [70] challenges, including a predecessor of the approach described here [78] [29].…”