PROJECT SUMMARY Genomic biomarkers enable precision medicine by matching patients to their optimal treatments. Mutational biomarkers are used as standard practice for oncology patients to assign them to treatment. However, mutations don’t have the predictive power for several therapies, including immunotherapy, which results in complete and durable responses for a subset of cancer patients. RNA sequencing (RNAseq) has been shown to improve accuracy over mutations in preclinical studies, including being able to predict response to immunotherapy, and even mutational status. The global RNAseq market is currently estimated at 9 billion dollars and is expected to grow at an annual rate of 10% from 2023 to 2030. The number of RNAseq biomarker trials has already doubled in 2020-2023 compared to 2010-2020, based on data in clinicaltrials.gov. RNAseq biomarkers currently face three major challenges: (1) bioinformatics analysis to translate signatures, cutoffs, etc. for application in the clinic, (2) regulatory issues with reproducibility and clinical validation, and (3) lack of software platforms to provide clinical trial support for biomarkers. There is currently no commercial software to support patient selection based on their RNAseq profiles at baseline, to the best of our knowledge. We have developed a prototype software suite, BIOCLIN, for an ongoing multi-site phase II RNAseq biomarker trial in kidney cancer (NCT05361720). It includes analysis of RNAseq data for each patient on the fly, run machine learning models and display gene signature scores per patient in an easy to use, intuitive user interface. Our goal is to develop our prototype into a commercial product with additional functionality that can be used to rapidly deploy platforms to support biomarker-based trials. In this Phase I proposal, we will (1) conduct extensive user research to understand the needs of investigators for running RNAseq based biomarker trials and develop a roadmap for our BIOCLIN app (2) develop methods to translate preclinical signatures for clinical use and implement a variety of machine learning modules to be used for biomarker development, and (3) develop a robust cloud-based backend infrastructure to enable rapid deployment of custom biomarker platforms at a price point to make such trials economically feasible for investigators. In Phase II we plan to implement methods to tackle regulatory challenges, a robust data security framework, an improved user interface and a cloud agnostic backend. Our proposal, if successful in obtaining funding, will address the gap in precision medicine by enabling investigators to test RNAseq biomarkers in clinical trials, which will undoubtedly have an impact on patients.