Dataset Information

Prediction of new drug indications based on clinical data and network modularity.

ABSTRACT: Drug repositioning is commonly done within the drug discovery process in order to adjust or expand the application line of an active molecule. Previous computational methods in this domain mainly focused on shared genes or correlations between genes to construct new drug-disease associations. We propose a method that can not only handle drugs or diseases with or without related genes but consider the network modularity. Our method firstly constructs a drug network and a disease network based on side effects and symptoms respectively. Because similar drugs imply similar diseases, we then cluster the two networks to identify drug and disease modules, and connect all possible drug-disease module pairs. Further, based on known drug-disease associations in CTD and using local connectivity of modules, we predict potential drug-disease associations. Our predictions are validated by testing their overlaps with drug indications reported in published literatures and CTD, and KEGG enrichment analysis are also made on their related genes. The experimental results demonstrate that our approach can complement the current computational approaches and its predictions can provide new clues for the candidate discovery of drug repositioning.

SUBMITTER: Yu L

PROVIDER: S-EPMC5039412 | biostudies-literature | 2016 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of new drug indications based on clinical data and network modularity.

Yu Liang L Ma Xiaoke X Zhang Long L Zhang Jing J Gao Lin L

Scientific reports 20160928

Drug repositioning is commonly done within the drug discovery process in order to adjust or expand the application line of an active molecule. Previous computational methods in this domain mainly focused on shared genes or correlations between genes to construct new drug-disease associations. We propose a method that can not only handle drugs or diseases with or without related genes but consider the network modularity. Our method firstly constructs a drug network and a disease network based on ...[more]

PMID: 27678071

Similar Datasets

Project description:Importance:National Cancer Institute Clinical Trial Network (NCTN) groups serve a vital role in identifying effective new antineoplastic regimens. However, the downstream clinical effect of their trials has not been systematically examined. Objective:To examine the association of NCTN trials with guideline care and new drug indications. Design, Setting, and Participants:This retrospective cohort study evaluated phase 3 SWOG Cancer Research Network clinical trials from January 1, 1980, through June 30, 2017. Only completed trials with published results were included. To be considered practice influential (PI), a trial must have been associated with guideline care through its inclusion in National Comprehensive Cancer Network (NCCN) clinical guidelines or US Food and Drug Administration (FDA) new drug approvals in favor of a recommended treatment. Data were analyzed from June 15, 2018, through March 29, 2019. Main Outcomes and Measures:Estimated overall rate of PI trials, as well as trends over time. The total federal investment supporting the set of trials was also determined. Results:In total, 182 trials consisting of 148 028 patients were studied. Eighty-two studies (45.1%; 95% CI, 37.7%-52.6%) were PI, of which 70 (38.5%) influenced NCCN guidelines, 6 (3.3%) influenced FDA new drug approvals, and 6 (3.3%) influenced both. The number of PI trials was 47 of 65 (72.3%) among those with positive findings and 35 of 117 (29.9%) among those with negative findings. Thus, 35 of 82 PI trials (42.7%) were based on studies with negative findings, with nearly half of these studies (17 of 35 [48.6%]) reaffirming standard of care compared with experimental therapy. The total federal investment spent in conducting the trials was $1.36 billion (2017 US dollars), a rate of $7.5 million per study or $16.6 million per PI trial. Conclusions and Relevance:Nearly half of all phase 3 trials by one of the NCTN's largest groups were associated with guideline care or new drug indications, including those with positive and negative findings. Compared with the costs of a new drug approval in pharmaceutical companies, typically estimated at more than $1 billion, the amount of federal funds invested to provide this valuable evidence was modest. These results suggest that the NCTN program contributes clinically meaningful, cost-efficient evidence to guide patient care.

Project description:BackgroundDrug sensitivity prediction and drug responsive biomarker selection on high-throughput genomic data is a critical step in drug discovery. Many computational methods have been developed to serve this purpose including several deep neural network models. However, the modular relations among genomic features have been largely ignored in these methods. To overcome this limitation, the role of the gene co-expression network on drug sensitivity prediction is investigated in this study.MethodsIn this paper, we first introduce a network-based method to identify representative features for drug response prediction by using the gene co-expression network. Then, two graph-based neural network models are proposed and both models integrate gene network information directly into neural network for outcome prediction. Next, we present a large-scale comparative study among the proposed network-based methods, canonical prediction algorithms (i.e., Elastic Net, Random Forest, Partial Least Squares Regression, and Support Vector Regression), and deep neural network models for drug sensitivity prediction. All the source code and processed datasets in this study are available at https://github.com/compbiolabucf/drug-sensitivity-prediction .ResultsIn the comparison of different feature selection methods and prediction methods on a non-small cell lung cancer (NSCLC) cell line RNA-seq gene expression dataset with 50 different drug treatments, we found that (1) the network-based feature selection method improves the prediction performance compared to Pearson correlation coefficients; (2) Random Forest outperforms all the other canonical prediction algorithms and deep neural network models; (3) the proposed graph-based neural network models show better prediction performance compared to deep neural network model; (4) the prediction performance is drug dependent and it may relate to the drug's mechanism of action.ConclusionsNetwork-based feature selection method and prediction models improve the performance of the drug response prediction. The relations between the genomic features are more robust and stable compared to the correlation between each individual genomic feature and the drug response in high dimension and low sample size genomic datasets.

Project description:As a common malignant tumor disease, thyroid cancer lacks effective preventive and therapeutic drugs. Thus, it is crucial to provide an effective drug selection method for thyroid cancer patients. The connectivity map (CMAP) project provides an experimental validated strategy to repurpose and optimize cancer drugs, the rationale behind which is to select drugs to reverse the gene expression variations induced by cancer. However, it has a few limitations. Firstly, CMAP was performed on cell lines, which are usually different from human tissues. Secondly, only gene expression information was considered, while the information about gene regulations and modules/pathways was more or less ignored. In this study, we first measured comprehensively the perturbations of thyroid cancer on a patient including variations at gene expression level, gene co-expression level and gene module level. After that, we provided a drug selection pipeline to reverse the perturbations based on drug signatures derived from tissue studies. We applied the analyses pipeline to the cancer genome atlas (TCGA) thyroid cancer data consisting of 56 normal and 500 cancer samples. As a result, we obtained 812 up-regulated and 213 down-regulated genes, whose functions are significantly enriched in extracellular matrix and receptor localization to synapses. In addition, a total of 33,778 significant differentiated co-expressed gene pairs were found, which form a larger module associated with impaired immune function and low immunity. Finally, we predicted drugs and gene perturbations that could reverse the gene expression and co-expression changes incurred by the development of thyroid cancer through the Fisher's exact test. Top predicted drugs included validated drugs like baclofen, nevirapine, glucocorticoid, formaldehyde and so on. Combining our analyses with literature mining, we inferred that the regulation of thyroid hormone secretion might be closely related to the inhibition of the proliferation of thyroid cancer cells.

Dataset Information

Prediction of new drug indications based on clinical data and network modularity.

Publications

Prediction of new drug indications based on clinical data and network modularity.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets