Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
. 2014 Sep;9(9):2147-63.
doi: 10.1038/nprot.2014.151. Epub 2014 Aug 14.

Similarity-based Modeling in Large-Scale Prediction of Drug-Drug Interactions

Affiliations
Free PMC article

Similarity-based Modeling in Large-Scale Prediction of Drug-Drug Interactions

Santiago Vilar et al. Nat Protoc. .
Free PMC article

Abstract

Drug-drug interactions (DDIs) are a major cause of adverse drug effects and a public health concern, as they increase hospital care expenses and reduce patients' quality of life. DDI detection is, therefore, an important objective in patient safety, one whose pursuit affects drug development and pharmacovigilance. In this article, we describe a protocol applicable on a large scale to predict novel DDIs based on similarity of drug interaction candidates to drugs involved in established DDIs. The method integrates a reference standard database of known DDIs with drug similarity information extracted from different sources, such as 2D and 3D molecular structure, interaction profile, target and side-effect similarities. The method is interpretable in that it generates drug interaction candidates that are traceable to pharmacological or clinical effects. We describe a protocol with applications in patient safety and preclinical toxicity screening. The time frame to implement this protocol is 5-7 h, with additional time potentially necessary, depending on the complexity of the reference standard DDI database and the similarity measures implemented.

Conflict of interest statement

COMPETING FINANCIAL INTERESTS: The authors declare no competing financial interests.

Figures

Figure 1
Figure 1
Overview of the protocol to develop the DDI predictor.
Figure 2
Figure 2
Example of some structural keys in the MACCS fingerprint for the drug diazepam. ‘Key position’ assigns a specific number to a particular chemical structural feature; ‘Key description’ describes the said structural feature; and ‘Key code’ assigns a value of ‘1’ when the structural feature is present in the drug being examined, and a value of ‘0’ when the structural feature is not present.
Figure 3
Figure 3
Different drug fingerprints codifying in bit positions drug interactions (IPFs), target information (target fingerprints) or adverse effects (ADE fingerprints).
Figure 4
Figure 4
Workflow of the different steps implicated in the generation of the matrix M1, containing the reference standard DDI database.
Figure 5
Figure 5
Workflow of the different steps implicated in the generation of the matrix M2 containing the 2D structural MACCS similarity information.
Figure 6
Figure 6
Workflow of the different steps implicated in the generation of the matrix M2 containing IPF similarity information.
Figure 7
Figure 7
Workflow of the different steps implicated in the generation of the matrix M2 containing target and ADE similarity information.
Figure 8
Figure 8
Workflow of the different steps implicated in the generation of the matrix M2 containing the 3D pharmacophoric similarity information.
Figure 9
Figure 9
Generation of the new set of potential DDIs (matrix M3).
Figure 10
Figure 10
DDI effect linkage: list of DDIs extracted from M3 are associated with the initial source in M1 and with the clinical or pharmacological effects caused by the interaction.
Figure 11
Figure 11
Integration of the five DDI scores into one unique score using PCA.
Figure 12
Figure 12
ROC curves showing the performance of the different DDI predictors in the DrugBank database (example provided in ANTICIPATED RESULTS with 9,454 true positives and 420,674 false positives). TPF, true positive fraction; FPF, false positive fraction.

Similar articles

See all similar articles

Cited by 31 articles

See all "Cited by" articles

Publication types

Feedback