jmetz
Zebrafish

NeuroAnalyser

NeuroAnalyser12 is an automated analysis framework for brain-activity analysis in zebrafish. This project was instigated to investigate various mechanisms related to Epilepsy, using zebrafish as a model.

The input format can be 2D (images) and 3D (volumes) time-series data, resulting in very large 3D or 4D datasets.

As this became a longer-running project. The core was significantly refactored in recent years, to both update the modules used allowing this to run on more modern systems, as well as adding the 2D timeseries functionality as the project was initially conceived to work only with volume-timeseries.

The core modules of this project are written in Python, but it also features a Rust-based launcher, allowing it to be distributed much more easily than a standard Python project, as no knowledge of Python (or Rust!) is required.

References

Zebrafish fluorescence analysis: semi-automated workflow

Working in collaboration with Dr Jon Green and Prof Charles Tyler, I developed a semi-automated analysis workflow for detection and fluorescence analysis of Zebrafish3.

For this project I wrote an automated detection pipeline and also a manual verification web application that allowed project members and collaborators to quickly and efficiently screen the resulting detection results prior to the statistical analyses.

References

Image-stack stitching: stacknstitch

Dr Mourabit approached me for help in techinques I could recommend for stitching images together to form a 3d image stack from smaller 2D images taken on a microscopy setup with brightfield and fluorescence imaging objectives4. This ultimately led to stacknstitch5, a Python framework for automatically stitching the images together.

References

4

Mourabit, S., Fitzgerald, J. A., Ellis, R. P., Takesono, A., Porteus, C. S., Trznadel, M., Metz, J., , Winter, M.J., Kudoh, T. and Tyler, C. R. (2019). New insights into organ-specific oxidative stress mechanisms using a novel biosensor zebrafish. Environment international, 133, 105138.](https://doi.org/10.1016/j.envint.2019.105138)