Role Overview
Takeda's Clinical Data Science (CDS) organization partners across Research and Development to apply statistical thinking, quantitative methods, and data science to support evidence generation and informed decision-making. We are seeking a highly motivated PhD candidate for the 2027 Summer Internship Program, working alongside experienced statisticians on a research project advancing quantitative methodologies for clinical drug development, with an emphasis on oncology applications. The work spans methodological development, statistical analyses and simulation studies, and evaluation of proposed approaches to support data-driven decisions under uncertainty, with opportunities to present findings and contribute to a peer-reviewed publication.
Responsibilities
- Conduct focused literature reviews on advanced statistical methodologies and AI/ML applications in clinical drug development.
- Develop, implement, and evaluate innovative statistical and AI/ML approaches addressing clinical trial and drug development research questions.
- Design and conduct simulation studies assessing methodological performance across scenarios and assumptions.
- Build reproducible analytical workflows and implement statistical models using R and/or Python.
- Collaborate with statisticians to interpret findings and explore practical applications.
- Present key findings and deliver a final presentation to the broader team.
- Prepare a scientific manuscript for potential publication in a peer-reviewed statistical journal.
Requirements
- Currently enrolled in a PhD program in Statistics, Biostatistics, or a closely related quantitative discipline.
- Preferably completed at least two years of doctoral-level coursework or research.
- Strong foundation in statistical theory, statistical modeling, and computational methods.
- Advanced knowledge and demonstrated hands-on experience in AI/ML methodologies, such as random forests, gradient boosting, Bayesian Additive Regression Trees (BART), neural networks/deep learning, and other supervised learning approaches.
- Strong understanding of AI/ML model development, training, validation, performance evaluation, and interpretation.
- Solid knowledge of survival analysis, including Kaplan-Meier estimation, Cox proportional hazards models, and other time-to-event modeling approaches.
- Familiarity with oncology clinical trial endpoints, including overall survival (OS), progression-free survival (PFS), and objective response rate (ORR).
- Strong programming proficiency in R and/or Python, with experience implementing statistical and AI/ML models and conducting simulation studies.
Preferred Qualifications
- Knowledge of Bayesian statistical methods, causal inference, or advanced predictive modeling.
- Experience with survival prediction, treatment effect estimation, or analysis of clinical trial data.