Every single cancer is unique. The myCancerProject is a data science effort to make it possible to understand the unique features that drive every cancer and to accelerate efforts to identify treatment solutions that are personalized to each patient.
Advancing our understanding of cancer biology and developing better treatments for the disease requires the ability to organize, analyze and share complex data sets. A central goal of myCancerProject.org is to make it easier for stakeholders (scientists, physicians and patients) to transform raw data into actionable insights that improve healthcare outcomes, accelerate the pace of research and positively impact patients’ lives.
Cancer R&D depends on a diversity of data sources, many of which generate large quantities of raw data. myCancerProject.org is an effort by the Redmond Lab (EACRI, Providence Cancer Center, Portland OR) to make it easier to effectively utilize this data. We are developing software solutions that allow us to harvest the data from our research program (immune monitoring, pathology, ‘omics, flow cytometry, outcomes, EMRs, etc), integrate these disparate data streams and identify meaningful relationships. In particular, we are working on strategies to structure, analyze and visualize these data sets in ways that are accessible to users that are not computationally savvy (eg. clinicians, immunologists, etc).
Our in-house analytics utilize a cloud-based architecture, but we are also creating standalone versions of some of our visualization and analysis tools. These standalone tools are designed to be user-friendly, run in an internet web browser (Google Chrome) and accept structured CSV files as inputs. We intend to release these standalone tools as free, open-source resources in the hope that they will be used to create interactive living figures that foster data-sharing and accelerate the progress of cancer research.
Individual Projects and Visualizations
We are building new systems for visualizing, analyzing and sharing the many varieties of data that are generated during studies of new cancer immunotherapies. The ability to analyze how individual patients are responding to a cancer therapy is essential for accelerating drug development, personalizing treatment and improving outcomes. In particular, we have been working to develop tools that facilitate a detailed analysis of how a patient’s immune system responds to different kinds of cancer treatments. Specifically, we are focusing on immune monitoring data streams (flow cytometry, cytokine arrays, etc) and conventional clinical data sources (CBCs, blood chemistry, etc). We have developed interfaces to harvest data from EMR/EHR systems, immune profiling data, patient experiential data and other sources. For more information about specific projects and to explore interactive demonstrations, refer to the Visualizations page.
The Providence Cancer Center partners with a wide range of sponsors, from large pharmaceutical companies to non-commercial entities, to conduct drug trials that leverage our clinical expertise and comprehensive portfolio of cutting-edge analytical technologies. While large pharmaceutical companies have proprietary systems for capturing, integrating and analyzing the data generated from their clinical trials, many smaller companies and non-commercial trial sponsors lack these capabilities. To address this need, our team has developed solutions that allow us to integrate virtually all of the data streams generated by our clinical trial program and those generated by external sources. We have also developed advanced data-mining capabilities that allow us to deliver unique insights about predictive and prognostic biomarkers. To learn more, visit our Clinical Trial Analytics page.
The goal of this project is to facilitate the comparison of as many relationships as possible within a data set and to identify the statistically significant relationships that may warrant further investigation. The BOSS tools are open-source, standalone tools that run in the Google Chrome web browser and accept user-provided data in CSV format.
Identifying predictive relationships between potential biomarkers and outcomes is an essential component of personalizing cancer treatments to individual patients. Generally speaking however, complex biological systems are usually poorly represented by binary relationships. Therefore, we are working on developing new data mining algorithms and visualization tools to facilitate multi-dimensional analysis of biomarker-outcomes relationships.