CAFE MOCHA: An Integrated Platform for Discovering Clinically Relevant Molecular Changes in Cancer; an Example of Distant Metastasis and Recurrence-linked Classifiers in Head and Neck Squamous Cell Carcinoma

Abstract

BackgroundCAFE MOCHA (Clinical Association of Functionally Established MOlecular CHAnges) is an integrated GUI-driven computational and statistical framework to discover molecular signatures linked to a specific clinical attribute in a cancer type. We tested CAFE MOCHA in head and neck squamous cell carcinoma (HNSCC) for discovering a signature linked to distant metastasis and recurrence (MR) in 517 tumors from TCGA and validated the signature in 18 tumors from an independent cohort.\n\nMethodsThe platform integrates mutations and indels, gene expression, DNA methylation and copy number variations to discover a classifier first, predict an incoming tumour for the same by pulling defined class variables into a single framework that incorporates a coordinate geometry-based algorithm, called Complete Specificity Margin Based Clustering (CSMBC) with 100% specificity. CAFE MOCHA classifies an incoming tumour sample using either a matched normal or a built-in database of normal tissues. The application is packed and deployed using the install4j multi-platform installer.\n\nResultsWe tested CAFE MOCHA to discover a signature for distant metastasis and recurrence in HNSCC. The signature MR44 in HNSCC yielded 80% sensitivity and 100% specificity in the discovery stage and 100% sensitivity and 100% specificity in the validation stage.\n\nConclusionsCAFE MOCHA is a cancer type- and clinical attribute-agnostic computational and statistical framework to discover integrated molecular signature for a specific clinical attribute.\n\nCAFE MOCHA is available in GitHub (https://github.com/binaypanda/CAFEMOCHA).