One of the hardest problems in the fight against cancer is pulling the results of hundreds of scientific studies and clinical trials together into a single holistic picture of the disease. This is exactly the problem that researchers at the Parker Institute for Cancer Immunotherapy (PICI) have tackled with CANDEL, the Cancer Data and Evidence Library.

Our goal was to build a unified database for Cancer Immunotherapy research, spanning complex and constantly evolving data, that we could use as a foundation for reproducible analysis tools. We are excited to work with Datomic due to its flexible information modeling, schema adaptability, API ergonomics, and immutability, which allows us to have reproducible analysis results.

Pier Federico Gherardini and Lacey Kitch of PICI

Built using both Clojure and Datomic, the goal of CANDEL is to enable researchers to predict which patients might respond to which combinations of immunotherapy treatments. A key part of CANDEL is a generalized mechanism to incorporate new results into a single data model that allows researchers to investigate different treatment approaches across multiple domains. This model and Datomic’s powerful querying capabilities uniquely empowers researchers to discover new treatments and insights via cross-study analysis.

In addition, CANDEL also provides a suite of tools to enable researchers to create custom CANDEL databases for exploratory work, prepare new data for inclusion, and process data in R to leverage machine learning and data visualization.

Cognitect is proud to have partnered with PICI on this effort.