Positive-Unlabeled Learning for inferring drug interactions based on heterogeneous attributes

Abstract

BackgroundInvestigating and understanding drug-drug interactions (DDIs) is important in improving the effectiveness of clinical care. DDIs can occur when two or more drugs are administered together. Experimentally based DDI detection methods require a large cost and time. Hence, there is a great interest in developing efficient and useful computational methods for inferring potential DDIs. Standard binary classifiers require both positives and negatives for training. In a DDI context, drug pairs that are known to interact can serve as positives for predictive methods. But, the negatives or drug pairs that have been confirmed to have no interaction are scarce. To address this lack of negatives, we introduce a Positive-Unlabeled Learning method for inferring potential DDIs.ResultsThe proposed method consists of three steps: i) application of Growing Self Organizing Maps to infer negatives from the unlabeled dataset; ii) using a pairwise similarity function to quantify the overlap between individual features of drugs and iii) using support vector machine classifier for inferring DDIs. We obtained 6036 DDIs from DrugBank database. Using the proposed approach, we inferred 589 drug pairs that are likely to not interact with each other; these drug pairs are used as representative data for the negative class in binary classification for DDI prediction. Moreover, we classify the predicted DDIs as Cytochrome P450 (CYP) enzyme-Dependent and CYP-Independent interactions invoking their locations on the Growing Self Organizing Map, due to the particular importance of these enzymes in clinically significant interaction effects. Further, we provide a case study on three predicted CYP-Dependent DDIs to evaluate the clinical relevance of this study.ConclusionOur proposed approach showed an absolute improvement in F1-score of 14 and 38% in comparison to the method that randomly selects unlabeled data points as likely negatives, depending on the choice of similarity function. We inferred 5300 possible CYP-Dependent DDIs and 592 CYP-Independent DDIs with the highest posterior probabilities. Our discoveries can be used to improve clinical care as well as the research outcomes of drug development.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-017-1546-7) contains supplementary material, which is available to authorized users.

Similarity-based machine learning support vector machine predictor of drug-drug interactions with improved accuracies

“…In clinical practices and investigations, drug combinations have been used or tested to achieve improved therapeutic effects . Reportedly, 20%‐30% of all adverse reactions to drugs were caused by drug‐drug interactions . Thus, detecting drug‐drug interactions (DDI) in the early stage of drug discovery enables the identification of therapeutically beneficial drug combinations and adverse drug interactions.…”

“…1,2 Reportedly, 20%-30% of all adverse reactions to drugs were caused by drug-drug interactions. 3 Thus, detecting drug-drug interactions (DDI) in the early stage of drug discovery enables the identification of therapeutically beneficial drug combinations and adverse drug interactions. DDI have been identified by in vivo and in vitro laboratory studies, as well as mechanistic modelling methods, like pharmacokinetic models of the drug levels within the body.…”

Dual graph convolutional neural network for predicting chemical networks

“…Statistical machine learning methods such as kernel methods have also been successfully applied to chemical property prediction [19][20][21]. In addition, statistical machine learning methods have been applied to predicting chemical networks such as metabolic reactions [22][23][24][25][26][27], drug-drug interactions [28][29][30][31] and beneficial drug combinations [32,33] by taking a pair of compounds as an input to a classifier.…”

GuiltyTargets: Prioritization of Novel Therapeutic Targets with Deep Network Representation Learning

“…More specifically, our method, GuiltyTargets, first maps a genome-wide protein-protein interaction network annotated with differential gene expression information into an Euclidean space using Gat2Vec. In that space, we then use positive-unlabeled (PU) machine learning [21][22][23][24] to learn a ranking of candidate targets. To the best of our knowledge, network representation learning as a data driven approach to implicitly learn relevant topological features from a network structure has not been used for target prioritization so far.…”