Limits...
A network-based classification model for deriving novel drug-disease associations and assessing their molecular actions.

Oh M, Ahn J, Yoon Y - PLoS ONE (2014)

Bottom Line: The growing number and variety of genetic network datasets increases the feasibility of understanding how drugs and diseases are associated at the molecular level.These quantified scores were used as features for the prediction of novel drug-disease associations.Our classifiers using Random Forest, Multilayer Perceptron and C4.5 showed a high specificity and sensitivity (AUC score of 0.855, 0.828 and 0.797 respectively) in predicting novel drug indications, and displayed a better performance than other methods with limited drug and disease properties.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Engineering, Gachon University, Seongnam, Korea.

ABSTRACT
The growing number and variety of genetic network datasets increases the feasibility of understanding how drugs and diseases are associated at the molecular level. Properly selected features of the network representations of existing drug-disease associations can be used to infer novel indications of existing drugs. To find new drug-disease associations, we generated an integrative genetic network using combinations of interactions, including protein-protein interactions and gene regulatory network datasets. Within this network, network adjacencies of drug-drug and disease-disease were quantified using a scored path between target sets of them. Furthermore, the common topological module of drugs or diseases was extracted, and thereby the distance between topological drug-module and disease (or disease-module and drug) was quantified. These quantified scores were used as features for the prediction of novel drug-disease associations. Our classifiers using Random Forest, Multilayer Perceptron and C4.5 showed a high specificity and sensitivity (AUC score of 0.855, 0.828 and 0.797 respectively) in predicting novel drug indications, and displayed a better performance than other methods with limited drug and disease properties. Our predictions and current clinical trials overlap significantly across the different phases of drug development. We also identified and visualized the topological modules of predicted drug indications for certain types of cancers, and for Alzheimer's disease. Within the network, those modules show potential pathways that illustrate the mechanisms of new drug indications, including propranolol as a potential anticancer agent and telmisartan as treatment for Alzheimer's disease.

Show MeSH

Related in: MedlinePlus

Ten-fold cross-validation.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4214731&req=5

pone-0111668-g004: Ten-fold cross-validation.

Mentions: For performance testing, independent 10-fold cross-validations were conducted ten times. The training set in each 10-fold cross-validation consisted of a positive set of true drug-disease associations and a negative set of randomly generated drug-disease pairs. Each training set was arbitrarily separated into 10 parts (trained on nine of them and tested on the remaining one), and the process was repeated ten times for cross-validation. This procedure was applied to all of the ten different training sets, and a random negative set was respectively generated for each of them. The resulting AUC scores of the ten 10-fold cross-validations were averaged. We used C4.5, Multilayer Perceptron and Random Forest, as implemented in Weka v3.6 [18]. The 10-fold cross-validation results with the 10 training data sets are shown in Figure 4 and Table S2. The highest AUC (area under the ROC) is 0.855.


A network-based classification model for deriving novel drug-disease associations and assessing their molecular actions.

Oh M, Ahn J, Yoon Y - PLoS ONE (2014)

Ten-fold cross-validation.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4214731&req=5

pone-0111668-g004: Ten-fold cross-validation.
Mentions: For performance testing, independent 10-fold cross-validations were conducted ten times. The training set in each 10-fold cross-validation consisted of a positive set of true drug-disease associations and a negative set of randomly generated drug-disease pairs. Each training set was arbitrarily separated into 10 parts (trained on nine of them and tested on the remaining one), and the process was repeated ten times for cross-validation. This procedure was applied to all of the ten different training sets, and a random negative set was respectively generated for each of them. The resulting AUC scores of the ten 10-fold cross-validations were averaged. We used C4.5, Multilayer Perceptron and Random Forest, as implemented in Weka v3.6 [18]. The 10-fold cross-validation results with the 10 training data sets are shown in Figure 4 and Table S2. The highest AUC (area under the ROC) is 0.855.

Bottom Line: The growing number and variety of genetic network datasets increases the feasibility of understanding how drugs and diseases are associated at the molecular level.These quantified scores were used as features for the prediction of novel drug-disease associations.Our classifiers using Random Forest, Multilayer Perceptron and C4.5 showed a high specificity and sensitivity (AUC score of 0.855, 0.828 and 0.797 respectively) in predicting novel drug indications, and displayed a better performance than other methods with limited drug and disease properties.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer Engineering, Gachon University, Seongnam, Korea.

ABSTRACT
The growing number and variety of genetic network datasets increases the feasibility of understanding how drugs and diseases are associated at the molecular level. Properly selected features of the network representations of existing drug-disease associations can be used to infer novel indications of existing drugs. To find new drug-disease associations, we generated an integrative genetic network using combinations of interactions, including protein-protein interactions and gene regulatory network datasets. Within this network, network adjacencies of drug-drug and disease-disease were quantified using a scored path between target sets of them. Furthermore, the common topological module of drugs or diseases was extracted, and thereby the distance between topological drug-module and disease (or disease-module and drug) was quantified. These quantified scores were used as features for the prediction of novel drug-disease associations. Our classifiers using Random Forest, Multilayer Perceptron and C4.5 showed a high specificity and sensitivity (AUC score of 0.855, 0.828 and 0.797 respectively) in predicting novel drug indications, and displayed a better performance than other methods with limited drug and disease properties. Our predictions and current clinical trials overlap significantly across the different phases of drug development. We also identified and visualized the topological modules of predicted drug indications for certain types of cancers, and for Alzheimer's disease. Within the network, those modules show potential pathways that illustrate the mechanisms of new drug indications, including propranolol as a potential anticancer agent and telmisartan as treatment for Alzheimer's disease.

Show MeSH
Related in: MedlinePlus