Postdoctoral Research Assistant in Statistical Machine Learning for Genomics

You will develop scalable methods, implement reproducible research software and apply them to biobank-scale datasets, rigorously evaluating their statistical properties and predictive performance. You will prepare publications, present findings, collaborate with project researchers and provide guidance to junior researchers. You should hold, or be close to completing, a PhD/DPhil in machine learning, statistics, computer science, computational biology, statistical genetics or another relevant quantitative discipline. You should have a strong foundation in statistical inference and machine learning, with experience developing novel methods and evaluating their statistical and empirical performance. Strong research software skills and experience with large-scale or complex quantitative data are also essential. Previous experience in genomics or evolutionary biology is desirable but not essential.

The research programme brings together statistical inference, machine learning and AI to study the genetic basis of complex traits and how evolution shapes their genetic architectures. Possible directions include genomic language models, multimodal learning integrating genomic and health data with functional, genealogical and evolutionary information, federated learning, and multi-trait and multi-ancestry methods. Applications include association testing, fine-mapping and polygenic prediction.