Senior PKPD and Machine Learning (ML) scientist - 2-years position

The position

In this role, you will combine your proficiency in modelling and simulation within pharmacology, which requires efficient collaboration and communication with subject matter experts within drug discovery projects. You will be part of a growing team that collectively drives the implementation of PKPD methods and engage with scientists across the Research & Early Development organization. Furthermore, you will contribute to a variety of projects and therapeutic areas. Some of your main responsibilities will include: PKPD modelling on research compounds tested in different/selected animals and/or in vitro assays PKPD support in discovery drug projects including application of machine learning models for PK predictions and design guidance. Development of project specific machine learning models, in collaboration with scientists from the Digital Sciences and Innovation (DSI) function area, to allow prediction of PKPD properties from structural features together with in vitro data Translation approaches for prediction of human PKPD based on animal and in vitro data; extracting literature/competitor data on similar compounds to include in modelling Involvement in digitalization initiatives of making PKPD data ready for modelling both PKPD and machine learning Explore combinations of data-driven machine learning methods with traditional mechanistic models Qualifications You hold a Ph.D. in engineering, physics, math, computer science or other disciplines related to quantitative biology or pharmacology and ideally have 2+ years relevant experience (post PhD) in the pharmaceutical or biotech industry in pharmacometrics, mathematical modelling or machine learning. You have an excellent understanding of theory, principles, and statistical aspects of advanced mathematical modelling preferable including machine learning and have ability to translate these tools into actionable insights. To succeed in the role, you have: Excellent understanding of pharmacokinetics/pharmacodynamics (PKPD) theory and modelling, statistical methodology and data interpretation Extensive expertise and programming skills preferably also with PKPD modelling software (Phoenix WinNonlin/NLME, NONMEM, and/or Monolix) Extensive expertise and programming skills with software for data manipulation, exploration and visualization (e.g. with seaborn, plotly in Python) Experience and knowledge in common machine learning models (e.g., tree-based ensemble models, Neural Networks, Gaussian Processes) and relevant libraries (e.g., scikit-learn, statsmodels, pytorch) Experience applying machine learning tools on biological data e.g. for proteins and peptides with knowledge in molecule encoding and descriptor approaches, e.g. molecule fingerprints, rdkit molecule descriptor, sequence z-scales Knowledge and interest for biological systems and questions As a person, you are keen to learn new skills and have a large drive and desire to change the way of working by integrating various types of modelling approaches in decision making in drug discovery projects. You have excellent communication and stakeholder management skills and the ability to effectively interact with colleagues with a variety of backgrounds. About the department The Discovery PKPD and Quantitative Systems Pharmacology (QSP) department is growing to support a wide range of discovery drug projects within diseases from diabetes and obesity over hemophilia to growth disorders and a range of rare diseases. The models we develop and apply help to support drug discovery, optimize dose & schedule selection and combination therapy and much more. We work in interdisciplinary teams, with a focus to support and integrate in vitro and in vivo pharmacology data into current or new PKPD, predictive machine learning and/or QSP models to describe the dynamic interactions between drug(s) and biological systems. The models inform and support drug discovery, development and regulatory interactions.The department is anchored within Global Drug Discovery (GDD) in Research & Early Development (R&ED). The area supports the discovery and development portfolio with pharmacokinetic (PK) and pharmacodynamic (PD) evaluations using a variety of modelling and simulation tools, as well as mechanistic modelling (QSP). You will have highly skilled and dedicated colleagues supporting projects from early discovery, through non-clinical development, as well as early clinical development. In our daily work, we have close collaboration with all parts of the global Discovery and Development organization, research project teams as well as external collaborators. Working at Novo Nordisk At Novo Nordisk, we don’t wait for change. We drive it. We’re a dynamic company in an even more dynamic industry, and we know that what got us to where we are today is not necessarily what will make us successful in the future. We embrace the spirit of experimentation, striving for excellence without fixating on perfection. We never shy away from opportunities to develop, we seize them. From research and development to manufacturing, marketing and sales – we’re all working to move the needle on patient care.