Running intricate experiments with bespoke sensors: case study
How can you help lab teams struggling with a sensitive subject matter, a huge workload and lots of data analysis? We’ve recently been working with Professor Andrea Brand’s Group at The Gurdon Institute at the University of Cambridge, helping them improve how they manage intricate experiments on the genetic traits of fruit flies.
The Gurdon Institute Team
The Brand Group at Gurdon are working on a very intricate and sensitive experiment monitoring the genetic traits of fruit flies. Simply, the team want to see the effect of conditions they impose on the environment of the fruit flies – but achieving this is much more complex.
They run many experiments concurrently, each taking many days. This means the project has many different variations that occur within each experiment, usually in growth conditions or the environment. These variations mean that factors influencing the experiment can go untraced, and with so many repetitions in the experiment, this can be a recipe for a data disaster.
What the team needs is sensitivity in their equipment so that they can control the environment of the experiment. They only want to change genetic and biochemical factors to monitor the fruit flies, to make clear key events and what is a natural activity or a disease presentation. Any other anomalies can render their experiments invalid very quickly.
Small investment, big results
This challenge is one OpenIOLabs are primed to tackle. Our kit was installed to track the temperature of their incubators, where the fruit flies are kept during each experiment, and to bring all of the lab equipment together. As you can imagine, their type of research is fairly niche so there isn’t any purpose-built solution for them, instead, they have to make do. They have a lot of equipment that they’ve bought and tweaked to do what they need to, but of course, these run on different systems, use different code and are controlled in different ways.
The OpenIOLabs system allows them to use all their equipment together easily and to create a bespoke sensor solution for their specific situation. They can now control the sensors, monitor the conditions, set up alerts (in case of a change in conditions) and access all the data easily in one format.
Transforming control and analysis of complex experiments
OpenIOLabs offers control for the Gurdon team, and for anyone running a complex and sensitive experiment like this. Some of the main advantages include 24-hour remote access, bespoke sensor installation and much-improved data analysis.
As soon as the OpenIOLabs system was installed, the team had instant access 24 hours a day. There are multiple experiments which are extremely intricate and affected by the slightest changes, so being able to monitor and control them from anywhere (not just the lab) is incredibly valuable. As we’ve explored in previous blogs, this not only protects their work but even improves work/life balance.
Their new setup has also allowed the team to install bespoke sensors where they actually need them, which then means they can see the conditions in which the fruit flies exist. This sounds like a small matter, but the difference in temperature from the top and bottom of the incubator can vary greatly, and this would have an effect on the development of the fruit fly. The additional sensors allow for sensitive and specific monitoring of the entire environment, meaning that researchers don’t discover anomalies after the experiment is over.
Managing big data
Then we come to data analysis, these experiments produce huge amounts of data, which were previously in different formats. OpenIOLabs collects the data from each piece of equipment in one language or format and then tags it with relevant metadata ready for analysis. This metadata contextualises the results and the potential to see cause and effect over long periods of time – something which is hard to achieve in traditional lab data analysis.
The data is also stored in a central place – creating accessibility and allowing the team to employ machine learning techniques to analyse the data for patterns.
The team have now implemented OpenIOLabs but this isn’t the end of the story, we’re collaborating with them to achieve much more from our system. The team want to improve reproducibility, something which is very important to scientists. Doing this can help reduce workload and improve the accuracy of experiments. As the system becomes established in their lab they have a lot of data to work from, and it’s all meta-tagged and therefore in context.
We’re excited to continue working with the team at the Gurdon Institute, and we’ll report back on their activity in the future.