Believe in Bioinformatics
When an industry grows rapidly, innovative approaches are required to ensure sustainable growth. Here, we speak to leaders at Ireland’s National Institute for Bioprocess Research and Training – NIBRT – to find out how a focus on talent, training and technology is changing the face of the field.
What is your role at NIBRT?
I lead the bioinformatics and data analytics research group at NIBRT. My background is in bioinformatics, with a focus on the application of multivariate statistics and machine learning for the analysis of high dimensional datasets. My primary research interest lies in the utilization of these techniques to identify the fundamental biological processes that drive the production of recombinant therapeutic proteins in mammalian cell factories. I’m trying to understand how CHO cells grow to high density and synthesize large quantities of therapeutic proteins, or indeed why some CHO cells don’t. Enabled by a Science Foundation Ireland (SFI) grant, we work with scientists here at NIBRT, the cell line-engineering group at the National Institute for Cellular Biotechnology, and international collaborators in the area to translate this understanding into increases in production efficiency. Industrial relevance is crucial to our research – and we have on-going partnerships with biopharma companies in Ireland, the UK and US.
Why did you go into this area?
I’ve always had an interest in computers and biology, and at the time the data explosion in biology was just beginning so it seemed like the natural route to take in my studies. Bioinformatics is now a critical component of biological science. The field has evolved at a remarkable pace, even in the short time since the human genome project. The development of next generation sequencing technologies in recent years has signaled the birth of a new era and pushed the demand for bioinformatics to new levels. It’s a great time to be working in this area.
How does bioinformatics fit into manufacturing?
One of the big challenges for the biopharma manufacturing industry is prediction of outcome – and the more you know about the machinery of the cell factory, the closer you can get to being able to predict if a particular cell, or indeed population of cells, is going to perform well in large-scale bioreactors. At the moment, there is a lot of cell screening to find the best cells that will give the best performance, but there’s a community of us that believe we can engineer a cell using gene-editing technology to build better cell factories. A better understanding of the biological system can also be used to better understand the potential impact of new production modes, such as continuous culture on cell performance. We know the smallest alterations in a process can impact the cells and there are many unanswered questions around how CHO cells will behave during extended culture. What happens when you start running processes continuously for longer periods of time? For instance, fed-batch culture processes running over a period of around 14 days are well established. If the culture runs for 50 or 100 days, can we maintain product gene expression?
What are you working on right now in CHO biology?
The CHO cell biology field is a relatively small community and there are few bioinformaticians working in this area. So we’ve been looking at the computational side and developing graphical user interfaces so that people who aren’t experts in the field can analyze CHO cell next generation RNA sequencing data. In addition, we have a number of projects ongoing with biopharmaceutical companies examining model cell lines displaying desirable/undesirable phenotypes, as well as investigating the origin of specific production issues. We are also developing algorithms for mass spectrometry in collaboration with Jonathan Bones’ group (see Know Your Process, Know Your Product).
Are you applying your expertise in data analysis to other areas?
Yes, we have also started to look at the utility of “big data” technologies for biopharmaceutical manufacturing. For example, if you had the computational infrastructure to combine and analyze all the data generated in a manufacturing plant in one database, what questions could you ask? What could you understand about the process that you didn’t know before? And is it possible to use predictive analytics to achieve optimal performance?
What’s the future of bioinformatics in biopharma?
I think that data analytics is one area where the biopharma industry has a lot of catching up to do compared with other industries. If we look at car manufacturers, they’re already onboard with big data technologies. Biopharma could really benefit from big data because there’s so much variability in processes.
Train and Retain
By Killian O’Driscoll, Projects Director at NIBRT
Finger on the Biopharma Pulse
With John Milne, Bioprocessing Training Director at NIBRT
Know Your Process, Know Your Product
With Jonathan Bones, Principal Investigator at NIBRT
