Cambridge Cancer Genomics (CCG) is a leading precision medicine start-up, creating intelligent software to enable oncologists to monitor cancer patient response to therapy throughout treatment. We have a strong focus on time-series datasets, with projects ranging from investigating clonal dynamics, to reconstructing pseudotime, and machine learning predictive analytics. With this in mind, we propose a hackathon challenge centred around longitudinal response to targeted therapy:
Vemurafenib is a targeted therapy that has shown great promise in melanoma patients with the BRAF V600E mutation. The main caveat to which is commonly acquired resistance. For this challenge we provide data from 10 BRAF-mutated melanoma cell lines, treated with Vemurafenib and RNA-Sequenced across time. From this dataset, one cell line (M381) displays innate resistance to therapy, and three others (M397, M229 and M263) display proliferation inhibition at first and acquired resistance later. Your challenge is to develop a tool or pipeline that can interrogate these data for clinically actionable insights - translating raw genomic/transcriptomic data to clinical information.
Further metadata is supplied, along with the data, in an AWS s3 bucket (instructions for download from command line below). The CEO and CTO of CCG will be delighted to meet hackathon attendees who participated in this project and discuss with them the approaches adopted to address this challenge.