Computational Drug Repositioning by Ranking and Integrating Multiple Data Sources

Zhang, Ping; Agarwal, Pankaj; Obradović, Zoran

doi:10.1007/978-3-642-40994-3_37

Cited by 28 publications

(46 citation statements)

References 25 publications

(50 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Namely, clinical side-effects of each drug have been collected and stored in numerous databases. The side-effects provide a drug with a profile and allow for construction of pairwise side-effect similarities between two drugs using the Jaccard index [117]. -Protein similarity networks represent networks of proteins with similar sequences (PSeqS) [118] or structures (PStrS) [119].…”

Section: Biological Data and Network Representationmentioning

confidence: 99%

Methods for biological data integration: perspectives and challenges

Gligorijević¹,

Pržulj²

2015

J. R. Soc. Interface.

209

157

View full text Add to dashboard Cite

Rapid technological advances have led to the production of different types of biological data and enabled construction of complex networks with various types of interactions between diverse biological entities. Standard network data analysis methods were shown to be limited in dealing with such heterogeneous networked data and consequently, new methods for integrative data analyses have been proposed. The integrative methods can collectively mine multiple types of biological data and produce more holistic, systems-level biological insights. We survey recent methods for collective mining (integration) of various types of networked biological data. We compare different stateof-the-art methods for data integration and highlight their advantages and disadvantages in addressing important biological problems. We identify the important computational challenges of these methods and provide a general guideline for which methods are suited for specific biological problems, or specific data types. Moreover, we propose that recent non-negative matrix factorization-based approaches may become the integration methodology of choice, as they are well suited and accurate in dealing with heterogeneous data and have many opportunities for further development.

show abstract

Section: Biological Data and Network Representationmentioning

confidence: 99%

Methods for biological data integration: perspectives and challenges

Gligorijević¹,

Pržulj²

2015

J. R. Soc. Interface.

209

157

View full text Add to dashboard Cite

show abstract

“…7 They collected 1,007 approved drugs and their targets from DrugBank, 18 the chemical structure information of these drugs from PubChem 19 and the side effect information from SIDER. 20 The drugs were represented by a combination of 775 targets extracted from DrugBank and 881 substructures in PubChem.…”

Section: Data Setsmentioning

confidence: 99%

“…Following the previous studies, we balanced the data set such that it contained twice as many negative instances as positives. 4,7 Thus, in a k-fold cross validation run, we created k groups containing 2, 229/k positive instances and 2 × 2, 229/k negative instances that were randomly chosen among all negative instances. Each fold was used as the test set once, in which all the remaining folds were used to train the classifier.…”

Section: Prediction Accuracy Evaluationmentioning

confidence: 99%

“…These studies often combine various features including but not limited to chemical 2D fingerprint similarity, overlap or interaction network closeness of drug targets and correlation between drug side effects and build a machine learning model based on different algorithms, such as support vector machines, random forests and logistic regression classifiers. [4][5][6][7][8][9][10][11] The proposed models are then compared in a cross validation setting, in which a portion of the known drug-disease associations are hidden during training and used for the validation afterwards. The areas under reciver operating characteristic (ROC) curves in the cross validation analysis reported for these models range between 75-95%, suggesting that some of these models can accurately identify novel drugdisease associations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reproducible Drug Repurposing: When Similarity Does Not Suffice

Güney

2016

Biocomputing 2017

View full text Add to dashboard Cite

Repurposing existing drugs for new uses has attracted considerable attention over the past years. To identify potential candidates that could be repositioned for a new indication, many studies make use of chemical, target, and side effect similarity between drugs to train classifiers. Despite promising prediction accuracies of these supervised computational models, their use in practice, such as for rare diseases, is hindered by the assumption that there are already known and similar drugs for a given condition of interest. In this study, using publicly available data sets, we question the prediction accuracies of supervised approaches based on drug similarity when the drugs in the training and the test set are completely disjoint. We first build a Python platform to generate reproducible similaritybased drug repurposing models. Next, we show that, while a simple chemical, target, and side effect similarity based machine learning method can achieve good performance on the benchmark data set, the prediction performance drops sharply when the drugs in the folds of the cross validation are not overlapping and the similarity information within the training and test sets are used independently. These intriguing results suggest revisiting the assumptions underlying the validation scenarios of similarity-based methods and underline the need for unsupervised approaches to identify novel drug uses inside the unexplored pharmacological space. We make the digital notebook containing the Python code to replicate our analysis that involves the drug repurposing platform based on machine learning models and the proposed disjoint cross fold generation method freely available at github. com/emreg00/repurpose.

show abstract

“…It presents a promising avenue for identifying better and safer treatments without the full cost or time required for de novo drug development. Many algorithms have been proposed as the hypothesis generation tools for the drug repositioning process because of the huge number of drug-disease pairs [11,31,35]. At the same time, as the number of approved drugs is continuously increasing, Drug-Drug Interaction (DDI) has become a serious health and safety issue which draws great attention from both academia and industry.…”

Section: Introductionmentioning

confidence: 99%

Computational Drug Discovery with Dyadic Positive-Unlabeled Learning

Liu¹,

Qiu²,

Zhang³

et al. 2017

Proceedings of the 2017 SIAM International Conference on Data Mining

Self Cite

View full text Add to dashboard Cite

Computational Drug Discovery, which uses computational techniques to facilitate and improve the drug discovery process, has aroused considerable interests in recent years. Drug Repositioning (DR) and DrugDrug Interaction (DDI) prediction are two key problems in drug discovery and many computational techniques have been proposed for them in the last decade. Although these two problems have mostly been researched separately in the past, both DR and DDI can be formulated as the problem of detecting positive interactions between data entities (DR is between drug and disease, and DDI is between pairwise drugs). The challenge in both problems is that we can only observe a very small portion of positive interactions. In this paper, we propose a novel framework called Dyadic PositiveUnlabeled learning (DyPU) to solve the problem of detecting positive interactions. DyPU forces positive data pairs to rank higher than the average score of unlabeled data pairs. Moreover, we also derive the dual formulation of the proposed method with the rectifier scoring function and we show that the associated non-trivial proximal operator admits a closed form solution. Extensive experiments are conducted on real drug data sets and the results show that our method achieves superior performance comparing with the state-of-the-art.

show abstract

Computational Drug Repositioning by Ranking and Integrating Multiple Data Sources

Cited by 28 publications

References 25 publications

Methods for biological data integration: perspectives and challenges

Methods for biological data integration: perspectives and challenges

Reproducible Drug Repurposing: When Similarity Does Not Suffice

Computational Drug Discovery with Dyadic Positive-Unlabeled Learning

Contact Info

Product

Resources

About