Available Projects

Micro-organisms such as C. elegans (sub-millimetre worms) or drosophila (fruit fly) are interesting model organisms to biologists working on understanding conditions that affect humans (such as Parkinsons disease) and how drugs can be developed to treat such conditions. Such organisms are usually observed on a dish and their behaviour noted after the DNA is manipulated or they are influenced by drugs. This behaviour is used to learn about condition/ treatment.

Primary supervisor: Professor Christopher James - Email: c.james@warwick.ac.uk

Project detail:
Videos of worms under a microscope or fruit flies under a camera lens can be recorded and the video recordings played back, biologists then observe the recordings and note subtle behaviour changes in the organisms. These might be as simple as have agile the organism is, what distance is travelled in a unit of time or how "flexible" the organism is under certain conditions (e.g. a worm "thrashing" when placed in a liquid). More specific behaviours could also be of interest as in mating-behaviour as well as more population level activities (e.g. how many a population of worms make it to a place where food is in a given amount of time).

Whatever the organism or the paradigm being assessed, what is usually constant is that the "analysis" under taken is usually "manual" - i.e. the organism behaviour manually observed and assessed. This is not only time-consuming (and boring!) but it is also prone to inaccuracies of observation, miscounting and usually do not last long as long-term monitoring is nigh on impossible in this situation.

This project regard the automated analysis of the behaviour of micro-organisms (such as C. elegans and drosophila) from recorded videos. Whatever the analysis the same preparatory steps needed, these include: 1) image/frame acquisition and clean-up, 2) image segmentation and organism recognition, 3) tracking of organism(s) frame-by-frame/ over time, 4) presentation of organism behaviours as time-series and 5) extraction of sub-behaviours/ mannerisms for time-series extracted in 4).

The project includes a mix of basic image processing techniques, with the use of machine learning based methods to quickly and reliably identify micro-organism(s) and to track them over time. Once extracted, sub-behaviour need to be extracted based on an understanding of the particulate paradigm being assessed (this might be e.g. looking for number of "reversals" in C. elegans, to amount of time spent "roaming" versus "dwelling" in the same, or extracting seizure-like behaviour in fruit fly).

This project will use both C. elegans and drosophila videos to represent examples of automatically tracking both sub-millimetre and larger organisms. The project includes the collaboration of biologists work with both micro-organisms.