Interpretable machine learning is a growing research area, with important applications in the biological sciences, such as understanding how different genes regulate each other within biological pathways. However, the current common practice is first to build complicated, “black-box” models and try to understand what they learned afterwards. This has the disadvantage that a) interpretation is still not always possible and b) it is inefficient as the model must re-discover patterns that are already well-known from scratch every time it is trained.

Recent technological advances have allowed us to study DNA and RNA not only in a “bulk” tissue, but also at a single-cell resolution, giving us unprecedented insights into how organisms form, how diseases develop, and how cells communicate with each other inside a tissue.

In this project, you will build interpretable-by-design machine learning models for biological data that offer biological insights in a direct way. Such models will be applicable to a variety of single-cell datasets and experiments in different fields of biology.

Tasks and responsibilities:

• complete a PhD thesis within the official appointment duration (four years);
  
• develop novel algorithms for analyzing single-cell (multi)-omics data and design and perform in silico experiments to test them;
• be an active and responsible member of the research group and collaborate closely with fellow group members;
  
• discuss work with group members and at departmental meetings, and incorporate feedback;
• take a leading role in writing manuscripts for publication in peer-reviewed journals and conferences;
  
• participate in the PhD training programme of the University of Amsterdam;
• assist with teaching and supervision of Bachelor’s and Master’s students.

You will have the opportunity to:

• work at the interface of machine learning and biology;
  
• collaborate closely with biologists from a variety of disciplines including immunology and cancer biology;
• present results at national and international scientific conferences;
• expand academic, professional and personal skills;
• use high-end compute infrastructure via the national ICT cooperative (SURF)

You are passionate about research and want to develop into an independent scientist. You have a background in machine learning and artificial intelligence and like to build novel methods for the analysis of biological data. You are methodical, curious and able to take initiative, while also valuing close collaboration in an interdisciplinary and international research environment.

Your experience and profile 

You

• hold a MSc in Bioinformatics, Artificial Intelligence, Data Science, Biostatistics or a closely related discipline;
• have demonstrable experience with training machine and deep learning models, preferably using Python;
• have a basic understanding of biology and/or are interested in using computational methods to understand biology;
• are fluent in English, written and spoken.