Postdoktor i fysikinformerad generativ modellering av proteindynamik
Core
Developing computational methods to model proteins as dynamic conformational ensembles using physics-informed generative modeling and machine learning.
Role type
Postdoctoral researcher in computational biophysics and machine learning
Builds
Physics-informed generative models for protein dynamics and conformational ensembles
Domain
Computational biophysics / Molecular dynamics / Machine learning
Deliverable
production ML models
Required skills
Molecular dynamics simulations, Generative machine learning, Flow matching, Latent representation of free energy landscapes, Markov state models, Bayesian network modeling, Python programming
Preferred skills
Enhanced sampling, Free energy methods, GROMACS, AMBER, MDAnalysis, PyEMMA, PyTorch, TensorFlow, SLURM, Singularity
Responsibilities
Develop, implement, and evaluate machine learning-based models for protein dynamics; Generate, curate, and analyze data from molecular dynamics simulations; Integrate Markov state models and Bayesian networks to analyze metastable states and allosteric couplings; Validate generated ensembles against simulations and experimental data; Document code and data according to FAIR principles; Publish results in scientific journals and present at conferences.