Data Science Shared Resource

点击翻译按钮获取中文摘要

The development of modern high-throughput biotechnologies and the rapid generation of high-complexity biological data has revolutionized the way cancer is studied. Over the last six years, novel artificial intelligence (AI) algorithm developments together with ever-growing big data generation provide unprecedented opportunities for cancer research, but also major challenges for handling, analyzing, sharing, integrating, and interpreting big data. Recognizing the importance of these data, the Data Science Shared Resource (DSSR) of Simmons Comprehensive Cancer Center (SCCC) was established in 2010 under the leadership of Yang Xie, PhD. The goal of the DSSR is to provide comprehensive informatics, data analytics, data integration, and data management support for SCCC investigators. Specifically, the DSSR provides (1) access to high-performance computing systems, (2) support for bioinformatics and data analyses, including high-throughput molecular data pre-processing, quality assessment and analysis, and cancer image data analysis, (3) support for data integration and risk prediction modeling, including integrative analysis, biomarker discovery, and development of prediction models for clinical outcomes, (4) support for data management and data-sharing, including developing comprehensive databases and facilitating investigators’ use of publicly available datasets and bioinformatics tools, (5) data science support for grant applications, and 6) analytical tools/software distribution and education. In addition, the DSSR has developed a series of cancer data commons and web portals for lung, kidney, and liver cancers, which are high-priority research areas in the SCCC catchment area. The DSSR also developed computational tools to curate and integrate data from electronic health records (EHR) with data from SCCC-driven genomic, imaging, and tissue analysis research. The DSSR will continue to develop, maintain, and apply these integrated platforms to support SCCC projects. In the current project period, DSSR services were utilized by 136 investigators across all five SCCC research programs and provided key contributions to support the success of more than 15 NCI-funded research project grants and over 200 peer-reviewed publications, including work published in high-impact journals such as Nature, Science, Cell, JAMA Oncology, Cancer Discovery, Lancet Oncology, and Nature Genetics. DSSR services are made possible by the accumulated experience of the staff and by innovative, unique, and customized approaches to solving data analysis challenges. With strong support from SCCC and highly cost-effective operations, the DSSR provides extensive services for many SCCC members and has provided critical contributions to the scientific needs and objectives of SCCC.