Stage: Neural codes and computations: modeling and analysis of neuronal population activity in visual cortex
Internship aim: Analyze previously recorded large-scale two-photon calcium imaging data in awake mice.
What you can gain
- Advanced data analysis skills of large data sets (e.g., Montijn et al., 2015: https://doi.org/10.7554/eLife.10163)
- A conceptual understanding of neural circuits as multi-dimensional operators (e.g., Montijn et al., 2016: https://doi.org/10.1016/j.celrep.2016.07.065)
- The ability to program computational models of neural circuitry (e.g., Montijn et al., 2012: https://doi.org/10.3389/fncir.2012.00022),
- A co-authorship on a scientific publication, provided the internship is completed successfully
What you need
- excellent programming skills in MATLAB or Python with NumPy
- an affinity and/or interest in theoretical neuroscience; especially population coding
- an internship duration of at least 6 months, but preferably longer
- Assist with silicon probe experiments in mice
Cross-orientation suppression is the paradoxical effect that when visual stimuli with multiple – and different – orientation components are overlaid, this leads to a suppression of neuronal activity, rather than a summation of their responses to both orientations individually (Prieve & Ferster, Nature, 2006). This process can be used to study the underlying mechanisms of neural circuit formation, and the computations they perform (Busse et al., Neuron, 2009; Popovic et al., Journal of Neuroscience, 2018). Prior research has focused on single-cell level analyses of pyramidal neurons, but new methodological tools now allow us to investigate these phenomena at a population level. Several of the data sets include genetic labelling of PV cells, the largest subclass of GABAergic interneurons. In combination with the visual stimulation protocol that allows the investigation of the effect of short-term adaptation, we can now disentangle the contributions of inhibitory control and non-GABAergic adaptation. During this project, you will learn myriad analysis techniques from your daily supervisory (Jorrit Montijn), as well as how to develop computational models of neural networks. The overall aim of the project is to get a better theoretical understanding of neural circuit computations and neuronal population codes.
If you are interested, send an inquiry to Jorrit Montijn (j . montijn at nin . knaw . nl).