Digitization: The Lab of the Future
Digitization: The Lab of the Future
Complete the form below and we will email you a PDF version of "Digitization: The Lab of the Future"
Dr. Samantha Kanza is a senior enterprise fellow at the University of Southampton, where she works on several research projects, including the digitization of scientific research. At the Future Labs Live event 2022, Kanza presented a talk titled “The effects of COVID-19 on the digitization of scientific research”, where she discussed attitudes to electronic implementation in the laboratory, how this has been affected by the global pandemic and her visions for the laboratory of the future.
Technology Networks connected with Kanza at the event to further elaborate on her findings, discuss the challenges of digitizing a laboratory and why electronic implementation requires more than a just a change of technology.
Molly Campbell (MC) For our readers that may be unfamiliar with the concept, what is meant by electronic implementation in the laboratory?
Samantha Kanza (SK): This can mean several things, but in this context, I am referring to the use of technology in the laboratory, including both hardware and software to electronically capture data and information in the lab. This could be instruments, software packages, electronic lab notebooks and other digital platforms, or the hardware used for this capture, such as an iPad or computer.
MC: You have researched attitudes and behaviors relating to electronic devices used in the laboratory, such as ELNs. What key barriers to electronic implementation in the lab have you identified?
SK: There are a lot of different barriers, and unfortunately despite improvements in technology many still apply today. Some barriers appear more in academia, such as cost, both with respect to paying for ELN subscriptions, or new equipment/software, and the hidden time costs to implement any new system in a laboratory.
Accessibility is also a huge barrier. Many academic laboratories aren’t set up for the digital capture of laboratory information, both from a hardware and software perspective. Equipment is outdated (requiring the use of legacy machines and software), software packages do not play well together and there is lack of physical space to even sit a laptop or iPad to record scientific research digitally.
The lab can often be a hostile place for technology. Many researchers will be concerned at the idea of taking their laptops into the lab in case chemicals get spilt on them. Further, scientists find the act of taking their gloves off to type notes into a computer very distracting and intrusive, substantially more so than just making notes in a lab book, meaning that integrating the use of a system that requires typing into the laboratory workflow is very challenging.
Software compatibility is also an issue. During my ethnographic studies I observed scientists (in this instance, crystallographers) needing to use multiple different software packages during their laboratory work and manually transfer data between them because the software packages they required were not compatible with one another. This makes it even more challenging to encourage the adoption of digital technologies and devices in the lab.
MC: In your Future Labs Live talk, you discussed that “failed” experiments typically do not get digitized. Can you explain why this is an issue?
SK: This is a real issue because it means that we lose out on a lot of valuable data. I would argue it’s just as valuable to know that something hasn’t worked (and understand why!) as it is to read about things that did work. If we can’t get past the stigma of digitizing “failed” experiments, then scientists are going to keep repeating the same mistakes because they won’t have access to any material to tell them that these methods aren’t viable. By removing this stigma and embracing it, making all scientific research digital and available, then we can save valuable time and effort and provide a much more pleasant supportive culture across the scientific community.
MC: You note that, electronic implementation in the laboratory requires a change of attitude and organization, just as much as it does technology. Can you elaborate?
SK: Attitude is an ever-perpetuating barrier. Many academics do not want to give up using paper or paper lab notebooks in the laboratory., Unfortunately that attitude is reflected in some of the students both as they go through their degrees in the habit of using paper-based methods to record their work, and then don’t see the value of using digital methods as they progress to postgraduate study. It isn’t a bad thing that the academic environment is more freeform than industry, but it is an issue that students aren’t being taught the value of digitizing their research and that a great deal of research gets lost in the pages of lab books.
Further, in general, people can be afraid of change. It can be very challenging to persuade an entire research group to adopt a new piece of software or even more challenging to persuade them to adopt an entire new set of digital processes. This is a socio-technical task that cannot rely purely on technology, we need the support and adoption of users if we are to make improvements in these areas.
We need to start early and educate our young researchers on not just the importance of digitizing scientific research, but also provide them with the best methods and tools to realize this endeavor.
MC: Overall, what impact do you believe the COVID-19 pandemic is having on digitization in the laboratory?
SK: I think that COVID-19 has increased digitization somewhat and has increased it more significantly in certain areas. There is certainly a much higher use of software to conduct research activities, although with the participants of our study (predominantly PhD students, early career researchers and established researchers) we found that this was mostly centred around organizational software and communication software. Unfortunately, a lot of the barriers to digitization remain the same, and while COVID has paved the way for some improvements, we still have a long way to go. Additionally, some systems were implemented quickly as a necessity through COVID, so it is important to ensure that they remain fit for purpose in the longer term.
Interestingly, one of the main changes we observed was the significant uptake of Juypter Notebooks (a web-based interactive development environment for notebooks, code and data). However, I do not think that this was due to COVID, they just happened to align with the timescales of COVID. It does demonstrate however that more and more scientists are writing code and want to be able to store their code and data together. It also shows that with information that naturally aligns itself to electronic storage (e.g., code/datasets) that scientists are making use of digital tools. The challenge is to widen that pattern into all scientific research and data, even that which does not align as naturally with electronic capture.
Something that I am really interested in, and that I will be investigating next is talking to and surveying the student generation that started their studies during the pandemic, as it will be very interesting to see if or how their perspectives on digitization and the use of digital tools differ, given that they would not have started their studies in an environment where paper lab notebooks were the norm.
MC: Can you talk about the PSDI, what it is and its aims?
SK: PSDI stands for Physical Sciences Data Infrastructure and it was initially run as a pilot project funded by the UK Engineering and Physical Sciences Research Council. PSDI aims to accelerate research in the physical sciences and beyond by providing a data infrastructure that brings together and builds upon the existing data systems that researchers currently use. Digital transformation is happening in many areas of scientific research, but the support and progress in this transformation is still fragmented.
PSDI will develop technological components to connect data sources together and enable better data discoverability and reuse. However, it isn’t simply about building the technology; there is a significant amount of requirements gathering and stakeholder engagement to be undertaken in the community as well as training and application of the technology.
One of the areas that we researched in the pilot phase was process recording, where we built on previous research from Frey Group members (led by Professor Jeremy Frey) and looked at how things had changed over the last few years, particularly with reference to how things had changed since COVID. Process recording is central to scientific research, and this will be one of the strands of future PSDI work.
PSDI will provide an excellent opportunity for cross-domain communication and sharing of best practice to maximize our research ability.
MC: What are the required technologies for electronic implementation in the lab?
SK: We need integration software to enable existing software to work together (e.g., notebooking software and domain-based software), and specialist software for the different domain-based needs of the diverse physical sciences community.
We need data standards and the automatic capture of metadata. Alongside this, we also need data format converters as one of the big issues in scientific research currently is both the lack and saturation of data standards and the fact that researchers are all either using different standards or no standards at all. It is highly unlikely that we will ever persuade the entire community to use one or two standards, and some domains do require specific standards for specific things. However, if we can identify the most prevalent standards for data and metadata and work out how to convert between them such that then suddenly data can be used in multiple software packages (that utilize different standards) and different researchers can use data in different forms to link with their own data, that will make a major improvement to digitization.
Further, we need semantic enrichment of our data, which involves identifying the relevant ontologies and identifying where there are gaps in the available schemas.
Finally, we need hybrid technologies and a variety of methods for data capture (including paper-esque methods and voice). One of the biggest reasons that our labs aren’t entirely digital is the ease of use and many other affordances of paper over using a laptop or desktop computer. Short of coming up with an entirely new ergonomic design of the computer (which may well happen in the future) there are limitations to how well a desktop or laptop can fit into the laboratory workflow right in the middle of running an experiment. However, there are other technologies that can be used in that place; hybrid notebooks or smart notebooks where users can write on them like paper could be a way of capturing this information electronically and in the freeform method that users like imparting it. Further, the use of voice would preclude the requirement to even write things down.
MC: In your opinion, what do you envision the “Lab of the Future” will look like?
SK: While I appreciate this is very much a future vision, and does come with the pre-requisite of the improvement of various technologies (e.g., voice recognition, handwriting recognition, hybrid devices, formalization of data standards, etc.,) I honestly see the lab of the future as one that is full of technologies and opportunities.
Much like smart homes have become commonplace in today’s society, so will smart labs. Users will be able to control their laboratories by voice using smart lab assistants, all of the laboratory systems will be seamlessly linked together, and users will have multiple options to record their data via voice, tablets, phones or computers if they wish. I think that PSDI will play a significant part in helping us realize this vision.
For the readers interest, with respect to smart laboratories, we are already working on a version of this with our talk2lab project.
Dr. Samantha Kanza was speaking to Molly Campbell, Senior Science Writer for Technology Networks.