BIOSTATISTICS SHARED RESOURCE - PROJECT SUMMARY/ABSTRACT The increasing variety and complexity of data types, analytic approaches and study designs utilized in cancer research necessitate the availability of an organized and centralized biostatistics resource that can offer a wide range of statistical expertise, collaboration, and training opportunities to investigators at the Montefiore Einstein Comprehensive Cancer Center (MECCC). The Biostatistics Shared Resource (BSR) is staffed by experienced statisticians with strong track records in effective collaboration across all scientific programs, innovative research in cancer relevant statistical areas, and in training and mentoring investigators at all levels. BSR personnel have critical roles in enhancing the Center infrastructure and fostering multi-disciplinary team science given their broad knowledge of the scope of research activities within the MECCC and active involvement on various cancer center committees. The specific objectives of the BSR are to: (i) provide state-of-the-art biostatistical and bioinformatic support on all phases of cancer research, from experimental design and study conduct to data analysis and manuscript preparation; (ii) collaborate on the development of methodologically rigorous grant applications and new research initiatives; (iii) assist with the development and scientific review of clinical trial protocols; (iv) develop innovative biostatistics and bioinformatics approaches to address analytical challenges from new technologies in cancer research; (v) offer a wide range of biostatistics training opportunities to MECCC members and to mentor early stage cancer investigators; (vi) enhance the MECCC infrastructure and foster interdisciplinary collaborations via participation on scientific and administrative committees and interactions with other shared resources. In the next funding period, to be in line with the strategic research priorities of the MECCC, the BSR will focus on strengthening methodological expertise to address the anticipated needs of existing and newly recruited MECCC researchers. Specifically, we have recently established the Subcore of Clinical Trials Methodology focusing on novel clinical trials and pragmatic trials on cancer screening and therapy. In addition, we aim to increase expertise in machine learning approaches that can integrate clinical, laboratory, and different types of high dimensional omics data on cancer etiology, prognosis, and precision medicine and prevention. The overall goal in accomplishing these objectives is to provide a robust, comprehensive and cost-effective system of statistical support for MECCC investigators that contributes significantly to advancing the understanding of cancer etiologies and prognosis, as well as improving cancer prevention, detection and treatment strategies.