Using R to provide statistical functionality for QSAR modeling in CDK A description of the integration of CDK and the R statistical environment

A previous article[1] discussed the use of the CDK framework as a cheminformatics backend to statistical software, allowing for easy access to chemical information for modeling purposes. However, access to statistical functionality from within the CDK framework is a very useful feature. Ideally, the CDK should be agnostic in terms of which backend provides the statistical functionality. This requires that an interface between the CDK and the statistical software be implemented. This article describes the design and implementation of the general statistical modeling interface for CDK and then describes in detail the interface between the CDK and R [2].