Real world computing project
While completing Computing Units 1 and 2, our class, led by Dr McIver and Mr Rajewski, had the opportunity to engage in real-world computing projects that were designed to benefit researchers from Monash University. The aim was to create software that would assist teams from either Neuroscience, Sleep or Eco Campus with visualising or sorting their data.
Our team - consisting of Zoe, Tij and George - worked collaboratively on building a program that would assist the Neuroscience researchers.
When initially determining how to undertake a project such as this, the first step is to understand what data we were receiving as inputs. The researchers were studying Rhythmic Whisking, a process in which mice use their whiskers to detect which objects are in their surroundings and from there, have the ability to discriminate between a range of different materials.
The researchers were conducting experiments on how different types of stimuli trigger neuron firings in the brain of the mice. Every time a neuron fired, a timestamp was recorded, and this could be associated to the different objects the mouse's’ whiskers touched. The neuroscientists were looking at density of neuron firing versus time for the different stimulus - the more neuron firings at a specific moment, the more likely that those reactions had to do with the Rhythmic Whisking.
Running these tests hundreds of times for multiple different stimuli left the researchers with copious amounts of data that would take them hours to sort and plot. Our team decided that we wanted to shorten the data sorting and graphing time, creating a program that would automatically be able to do these things within a matter of seconds.
Using a coding language called Python, we were able to create a program that processed the large amounts of data and generated both a heat map and a scatter plot to visually represent the data provided. To get to this point though, we faced many challenges, such as converting the data contained in Matlab files to a useable file format. This was a challenge as none of us had previously used Matlab and hence initially struggled with making sense of the data.
However, once we passed this stage, we were able to look into many modules and form our graphs using the data set. Finally, we built a user interface to provide the Neuroscience team with graphing options, and allowed them to upload a file of their choice into the software to generate their chosen plot.
After completing this software, we were able to present our finished program back to the researchers to demonstrate the capabilities of the program. Our program was well liked, and the neuroscientists have decided to use it for their research. They will now use our program for visualisation assistance and to help come to conclusions in their understanding of their research.
Our team will now further collaborate with them to provide them with any other features that they may require.
The ability to work on a real world project was a fun and amazing experience that allowed us to better our coding skills. It has given us the opportunity to participate in a task that can make a difference to scientific research, and has given us an insight into the sort of work that we can pursue in the future.
- Zoe Freihofer, Tij Sharma and George Jose