Accurate prediction of kinase-substrate networks using knowledge graphs.

Phosphorylation of specific substrates by protein kinases is a key control mechanism for vital cell-fate decisions and other cellular processes. However, discovering specific kinase-substrate relationships is time-consuming and often rather serendipitous. Computational predictions alleviate these challenges, but the current approaches suffer from limitations like restricted kinome coverage and inaccuracy. They also typically utilise only local features without reflecting broader interaction context. To address these limitations, we have developed an alternative predictive model. It uses statistical relational learning on top of phosphorylation networks interpreted as knowledge graphs, a simple yet robust model for representing networked knowledge. Compared to a representative selection of six existing systems, our model has the highest kinome coverage and produces biologically valid high-confidence predictions not possible with the other tools. Specifically, we have experimentally validated predictions of previously unknown phosphorylations by the LATS1, AKT1, PKA and MST2 kinases in human. Thus, our tool is useful for focusing phosphoproteomic experiments, and facilitates the discovery of new phosphorylation reactions. Our model can be accessed publicly via an easy-to-use web interface (LinkPhinder).

Facilitating prediction of adverse drug reactions by using knowledge graphs and multi-label learning models.

Timely identification of adverse drug reactions (ADRs) is highly important in the domains of public health and pharmacology. Early discovery of potential ADRs can limit their effect on patient lives and also make drug development pipelines more robust and efficient. Reliable in silico prediction of ADRs can be helpful in this context, and thus, it has been intensely studied. Recent works achieved promising results using machine learning. The presented work focuses on machine learning methods that use drug profiles for making predictions and use features from multiple data sources. We argue that despite promising results, existing works have limitations, especially regarding flexibility in experimenting with different data sets and/or predictive models. We suggest to address these limitations by generalization of the key principles used by the state of the art. Namely, we explore effects of: (1) using knowledge graphs-machine-readable interlinked representations of biomedical knowledge-as a convenient uniform representation of heterogeneous data; and (2) casting ADR prediction as a multi-label ranking problem. We present a specific way of using knowledge graphs to generate different feature sets and demonstrate favourable performance of selected off-the-shelf multi-label learning models in comparison with existing works. Our experiments suggest better suitability of certain multi-label learning methods for applications where ranking is preferred. The presented approach can be easily extended to other feature sources or machine learning methods, making it flexible for experiments tuned toward specific requirements of end users. Our work also provides a clearly defined and reproducible baseline for any future related experiments.

Using Drug Similarities for Discovery of Possible Adverse Reactions.

We propose a new computational method for discovery of possible adverse drug reactions. The method consists of two key steps. First we use openly available resources to semi-automatically compile a consolidated data set describing drugs and their features (e.g., chemical structure, related targets, indications or known adverse reaction). The data set is represented as a graph, which allows for definition of graph-based similarity metrics. The metrics can then be used for propagating known adverse reactions between similar drugs, which leads to weighted (i.e., ranked) predictions of previously unknown links between drugs and their possible side effects. We implemented the proposed method in the form of a software prototype and evaluated our approach by discarding known drug-side effect links from our data and checking whether our prototype is able to re-discover them. As this is an evaluation methodology used by several recent state of the art approaches, we could compare our results with them. Our approach scored best in all widely used metrics like precision, recall or the ratio of relevant predictions present among the top ranked results. The improvement was as much as 125.79% over the next best approach. For instance, the F1 score was 0.5606 (66.35% better than the next best method). Most importantly, in 95.32% of cases, the top five results contain at least one, but typically three correctly predicted side effect (36.05% better than the second best approach).