Having Trouble Reconciling Lab Automation, Science and IT?

Written by Steven van Helden

It starts with a simple idea

Driven by the science and the impact you may have on improving your R&D process, you rush into a small automation project without thinking too much about the broader consequences. And then you find yourself overwhelmed in the world of lab automation!

Imagine you finally got it in your lab: Robot Arm 2 shiny new instruments connected by a robot arm to perform your analyses in an automated manner. Of course, you thought ahead and you have requested the vendor to develop workflows so you can get off to a flying start. Everyone is enjoying the efficiency improvement in the lab, but you know there is more to come... The volume of testing will increase thanks to this investment so you would like to extend the automation even further. Wouldn’t it be cool if the result files generated by these 2 instruments were automatically transferred to the office network so that they can be processed in the company LIMS? And that conclusion data were automatically loaded into the central repository so everyone in the company has easy access to your data? That would be a nice next step!

 

Thinking about the next step(s)

IT-Network The idea sounds simple enough. But how can these systems communicate with each other? And how do I ensure there are no security breaches in these connections? And are there any other systems in the company that might be impacted? You realize the project does not end at your doorstep. There are other stakeholders in the game such as IT. So, who are these other stakeholders? There are so many flavors of them in teams like data management, research informatics, infrastructure, office IT, ….

Database Assuming these technical IT hurdles can be overcome you can focus again on the scientific data. With these automated workflows you need to make sure that annotation is properly taken care of so that users of the databases understand what kind of data they are looking at.

 

Welcome to the domain of FAIR data!

FAIR With the ever increasing diversity and content of research data, it is crucial to make sure these data are Findable, Accessible, Interoperable and Reusable (FAIR). So now you start thinking about the annotation of your data and realize that it would be great to have a dictionary, or even better an ontology, of the types of data you generate and would like to store for future use. Again a simple thought, but a quick chat with your colleagues in the lab reveals that there are different terminologies in use and everyone is adamant theirs is the right one. Maybe you need some negotiating here? Data Mining

Following FAIRification of your data, you will finally be ready for data mining and exploitation of your valuable data. That is, after selection and installation of sophisticated data mining software. Maybe this is the right time to dive into AI as well?

I need help!

This is the point where you may feel overwhelmed. What started as a simple idea for improvement turns into a serious project involving multiple disciplines you are not an expert in. This is where wega can be your partner, there is no need to re-invent the wheel. The Research Data and Knowledge Management group at wega has expertise in the process described above. We know the business processes and the software vendors that may be of interest to you. And the earlier we come into the game, the easier it is to anticipate all the hurdles that you may find on your road. So let’s discuss how we can work together to turn your (not so-)simple idea into reality!