Institute
The European Bioinformatics Institute https://www.ebi.ac.uk/ is an international, innovative and interdisciplinary research organisation funded by European states. The EBI’s goal is to help scientists realise the potential of big data in biology, exploiting complex information to make discoveries that benefit humankind.
Project Supervisor: Ewan Birney / Tomas Fitzgerald
Project location: Remote
Project Requirements
- Essential: Experience of image processing and modern image analysis packages
- Desirable: Knowledge of machine learning methods and modelling techniques
- Essential: Python, bash
- Desirable: C++
Project Summary
Over the past years in close collaboration with institutes in Germany and Japan we have developed the first near isogenic (inbred) panel of vertebrates derived from a wild founder population of Medaka fish from Kiyosu Japan. This panel allows investigations into gene to environment effects by making it possible to modify environmental conditions while keeping the genetics of individuals fixed. These experiments require detailed and robust measurement extraction from a range of imaging datasets. We have developed initial image processing methods for heart rate estimation from imaging of live embryos in 96 well plate format. The datasets here comprise of high-resolution videos (multi tiff format) which are looped 3 times where the temperature of the plate is changed. These initial methods are written in Python making heavy use of the open-cv and scikit-image libraries and can be run on the command line resulting in the generation of a detailed report containing heart rate specific results.
Currently we have approximately 60 full plates from this heart rate assay with more data being generated daily. Primarily, in this project we would like the heart rate analysis package to be improved in terms of package structure and useability.
A desirable goal would be the development of a GUI interface allowing collaborators to run basic analysis at their end. There is also very large scope for research method development, the heart rate extraction is currently based on fourier transforms and power curve assessments which could potentially be improved, and we are also interested in further parameter extraction relating to morphological characteristics.
If you are interested the analysis of large / rich imaging datasets within a real-world novel genetics research project allowing both novel image analysis method development and software design this project is for you!