Representation Learning of Neuronal Morphologies

The Morphology of Neurons

The morphology of neurons is highly complex with widely varying shapes that are tightly linked to their function. Traditionally, neurons have been classified based on their morphology into diverse types using manual inspection. Recent advances in recording technologies have greatly accelerated data collection and therefore the amount of data available. These developments provide optimal conditions for applying unsupervised machine learning methods to characterize neuronal morphologies and identify cell types.


We use an implicit surface representation approach to learn the 3D morphologies of the neurons and assign each neuron with a bar code. Clustering those bar codes gives us information about similar and distinct neurons and can be compared to functional clusterings in a second stage. We hope to optimize these models to gain a better understanding of the neurons' morphologies and, in the future, potentially link it to their function.

implicit model

Figure 1: Implicit surface representation approach - model and clustering.


  • Good mathematical understanding (in particular statistics and linear algebra)
  • Interest in 3D data and good spatial awareness
  • Python programming
  • Experience with deep learning (PyTorch or Tensorflow recommended)


Laura Pede

Neural Data Science Group
Institute of Computer Science
University of Goettingen