Data Scientist, Multimodal Data & Analytics in Clinical Development

• Perform hands-on modeling of integrated clinical, outcomes and high-dimensional, patient-level biomarker data from clinical trials (genomics, transcriptomics, proteomics, flow cytometry, medical imaging, etc.) to generate fit-for-purpose evidence that is applied for decision making in drug development programs.
• Contribute to planning, execution, interpretation, validation and communication of innovative exploratory biomarker and AI analyses, to facilitate internal decision making, and support submissions of candidate drug and associated companion diagnostics packages.
• In collaboration with cross-functional partners, provide quantitative, scientific and strategic input to as well as hands-on execution of the integrated data science and AI strategies for one or more clinical programs in drug development.
• Collaborate with other line functions. Explain statistical concepts in an easily understandable way to non-statisticians and provide adequate statistical justifications and interpretation of analysis results for actions/decisions/statements, when required.
• Contribute to departmental initiatives (e.g., infrastructure/ process-related or scientific activities)
• Contribute to the scientific content of materials for internal decision boards/regulatory/submission documents: Briefing Books, decision criteria, trial design(s), responses to Health Authority questions

• Advanced degree in computational biology, statistics, data science or other quantitative discipline with extensive exposure to clinical trials and drug development. MS with 5+ years relevant work experience or PhD with 2+ years of relevant work experience in the pharma/biotech industry; level will commensurate with experience.
• Strong knowledge and understanding of (multivariate implementations of) statistical methods such as time to event analysis, machine learning, meta-analysis, mixed effect modeling, longitudinal modeling, Bayesian methods, variable selection methods (e.g., lasso, elastic net, random forest), design of clinical trials. Experience applying machine learning or deep learning in a clinical trial context is useful.
• Hands-on familiarity with statistical and analytical methods for genetics and -omics data analysis and working knowledge of high dimensional biomarker platforms (e.g., next generation sequencing, transcriptomics, proteomics, flow cytometry, etc.).
• Good understanding of clinical study design principles and familiarity working with patient-level clinical data in a clinical trial (GxP) setting

• Strong programming skills in R and Python. Demonstrated knowledge of data visualization, data engineering, exploratory analysis and predictive modeling.
• Good interpersonal and communication skills (verbal and writing)
• Ability to leverage technical and scientific information from a variety of sources in order to effectively enable data analysis and the scientific interpretation of analysis results.
• Ability to develop and deliver clear and concise presentations for both internal and external meetings in key decision-making situations.

• Experience with modern data science methods and designing machine learning or other AI architectures for applications in healthcare
• Additional expertise in big data, web-based applications, good coding practice, source code management (Git, bitbucket)
• Exposure to and hands-on experience with various forms of RWE is a plus.
• Experience analyzing and using models built based on PET-CT/MRI scans, H&E digital pathology, IHC, OCR, physiological sequence data (ECGs) or other non-canonical data formats a plus.

Novartis