Epilepsy is a chronic disorder that affects 2.4 million of people per year, according to the World Health Organisation. It provokes recurring seizures that appear with no warning after a latent period (known as epileptogenesis) and can result in physical injuries. Epilepsy has many causes and its symptoms may first show after the exposure to known risk factors as, for example, strokes, infectious diseases or brain injuries.
Current research focuses on predicting and preventing the development of epilepsy in individuals exposed to risk factors.
The only reasonable way to describe the role of EGI resources in my daily work is that they make possible what is practically impossible to achieve using ordinary computing power. Massimo Rizzi
To do this, researchers need to identify a marker that can predict the development of the disease before clinical signs occur. This marker can then be used to study the efficacy of new anti-epilepsy treatments.
Massimo Rizzi and his colleagues at the Mario Negri Institute for Pharmacological Research studied the problem using mice, which can develop epilepsy just as humans do.
They started by looking at the behaviour of mice and what changed in their brain’s electrical activity after exposure to a risk factor. For that, they performed epidural electrocorticograms (ECoGs) to record what happened to the mice during the experiments.
Then for the analysis, Rizzi and his colleagues used the High-Throughput Compute (HTC) and storage resources made available by the Italian Grid Infrastructure (IGI) and the biomed virtual organisation.
Using HTC meant shortening the time required for the calculations. Rizzi estimates that using a single PC to complete the calculations of 25,000 jobs, they would have needed 54 days. With HTC they completed the work in under 48 hours. Given that in total the team submitted about 200,000 jobs, the time saved was in the order of years.
The analysis revealed an oscillation pattern – also called a dynamic intermittency – in the ECoGs of mice developing epilepsy. Applying an experimental anti-epileptogenic treatment successfully reduced the rate of the event. The results, published in Scientific Reports, confirmed that high rates of dynamic intermittency are a marker of epileptogenesis.
Rizzi is confident that the findings will help to develop therapeutics that can prevent the emergence of epilepsy following the exposure to potential risk-factors, thus reducing the incidence of epilepsy. To this aim, the computational power available to Rizzi and colleagues has been significantly increased thanks to the EGI Cloud Compute resources.
“The only reasonable way to describe the role of EGI resources in my daily work is that they make possible what is practically impossible to achieve using ordinary computing power. This means that EGI resources can play a fundamental role in clinical and pre-clinical investigation on epilepsy, as well as in the Neurosciences as a whole. I am confident that important benefits for patients will come soon, thanks to the exploitation of such resources.” Massimo Rizzi
Rizzi’s team submitted about 200,000 High-Throughput Compute jobs and they completed the work in under 48 hours.
Rizzi et al. 2016. Scientific Reports. doi:10.1038/srep31129 (full text PDF, open source)