3015 Background: The clinical care of oncology patients is routinely informed by tumor and inherited genetic profiles. This is accomplished by molecular pathologists synthesizing the growing body of clinical guidelines and scientific evidence that associates cancer genome alterations and therapeutic response, and applying that knowledge during case reviews. Many academic medical centers formalize this process in the form of molecular tumor boards. As the number of cases for review and literature continue to increase, there is opportunity to leverage clinical interpretation algorithms to computationally prioritize molecular features and both enhance and automate the sample contextualization process. Here, we present the Molecular Oncology Almanac (MOAlmanac) to enable the rapid assessment of tumor actionability. Methods: Molecular Oncology Almanac is an open source clinical interpretation algorithm and paired knowledge base for precision cancer medicine. It is used to rapidly characterize and identify genomic features related to therapeutic sensitivity and resistance and of prognostic relevance. This is performed by assessing not only individual genomic features (e.g. somatic variants, copy number alterations, germline variants, and fusions) but also interactions between these events as well as secondary features such as mutational burden, mutational signatures, MSI status, and aneuploidy. MOAlmanac summarizes all clinically relevant findings into a web-based actionability report. The underlying knowledge base can be accessed through our API endpoints and web browser, and entries may be recommended through either Github or our browser extension. In addition, we developed a cloud-based web portal on top of the Terra framework to increase accessibility. Results: A total of 32,108 samples from 30,607 patients across 66 cancer types received targeted sequencing to characterize somatic variants, copy number alterations, and fusions from PROFILE’s Oncopanel and were evaluated with MOAlmanac. Based on Oncopanel’s tier 1 and tier 2 criteria for clinical actionability, we observed that 8,285 samples (26%, 0 - 69% by cancer type) of patients harbored at least one alteration suggesting therapeutic sensitivity based on FDA approvals or clinical guidelines. Actionability increases to 18,117 samples (56%, 0 - 85% by cancer type) when considering an expanded set of evidence to include relationships captured from clinical trials, clinical, preclinical, and inferential evidence; consequently providing at least one therapeutic hypothesis to otherwise variant-negative patients. Conclusions: Clinical actionability of molecular tumor data was increased in individual patients by expanding the set of evidence considered. Source code and a web portal for this project are available at moalmanac.org .