The Predictive Oncology Model and Data Clearinghouse (MoDaC) is a data repository and model clearinghouse developed to transition resources to the broader research community. These resources consist of predictive oncology datasets and mathematical models (such as machine learning and deep learning models) developed within NCI and in collaborative programs, including the NCI-DOE Collaboration projects and the Accelerating Therapeutics for Opportunities in Medicine (ATOM) Consortium. Annotated datasets and models stored in the repository are publicly available and can be searched against their metadata and downloaded.

An account is not needed to perform searches of the repository and view the metadata associated with a collection. Users who create an account can download data. Downloads can be performed asynchronously to a Globus endpoint or to an AWS S3 bucket, or synchronously to the user's computer. Users can monitor the progress of their asynchronous downloads at any point during the transfer.

MoDaC allows users to associate a Document Object Identifier with a stored asset and provides a shareable link for citations.

MoDaC provides support for evaluating MoDaC models deployed to the NCI on-premises environment. Users can evaluate these models through the MoDaC web interface or programmatically using the MoDaC REST API suite. For more information on this capability, contact us.