Meet Lhasa's CEO, Dr Chris Barber
Find out more about how Chris got here, his future plans, and what advice he has for aspiring scientists.
What are you most looking forward to in your new role as CEO?
For the past 11 years, Lhasa has grown steadily under the leadership of Dave Watson and I intend to build upon that legacy. All of us at Lhasa will continue to focus on the delivery of science, through software, for the benefit of our members. I’m particularly excited by some of the long-term science that we are developing in order to address future needs. I will also be working with our senior management team to ensure we coordinate our resources to deliver software which directly meets the needs of users.
From a personal perspective, I’m really excited about the potential of Mirabilis. Our newest piece of software is designed to support purge factor calculation for impurities introduced during synthesis. Mirabilis directly addresses a need under the global ICH M7 guidelines1. It has the potential to increase efficiency, without increasing the risk to patients, by helping the user to better understand the impurity profile of drug products. The development of Mirabilis was inspired by our members and its development continues to be spurred on by an incredibly passionate and supportive user community.
What do you think are Lhasa’s most exciting recent achievements?
We have recently won the Queen’s Award for Enterprise: Innovation for Sarah Nexus. Sarah Nexus is our first purely machine-learnt model and has set a high standard for transparent statistical models, which all too often are difficult to interpret by the user.
Across all of our software, our aim is to provide sufficient information for a user to make a decision. Whilst most model builders can provide a means to define an applicability domain (effectively describing the scope of the model), this is not the right question. We believe that rather than asking the model builder ‘when can I use your model?’, the user should ask ‘how confident is the model for this particular prediction, and why?’. Understanding the confidence with which the model made the prediction is much more useful to someone making a decision. Not only does Sarah supply a confidence measure, it also provides supporting data to help the user interpret a prediction.
For more information, Dr Thierry Hanser has described our current thinking in a recent publication: Applicability domain: towards a more formal definition.
Why are Lhasa’s software and solutions important?
There has never been a more important time for Lhasa and a greater need for the solutions we work to deliver. It may seem rather grand, but I believe that we are at the dawn of much wider acceptance of in silico predictions.
This is the result of pressures to reduce testing (for many reasons, including: cost, speed, a desire to reduce animal testing, and recognised limitations of in vitro and in vivo models) alongside an ever growing repository of data and new modelling techniques that can better exploit it.
The global acceptance of in silico models in lieu of the Ames test under the ICH M7 guidelines is a powerful catalyst for this future.
However, in silico model predictions must continue to develop if they are to provide sufficiently transparent, robust and detailed predictions to support confident decisions across more complex endpoints.
Our close engagement with members gives us an almost unique understanding of those decisions and what it takes to provide enough support to make them. This spurs us all on to meet those needs – without this, it could become very easy to forget the challenges our members face and the impact of making poorly supported decisions.
When did you first know you wanted to be a scientist?
I’ve wanted to be a scientist for as long as I can remember. I think in truth, it’s just remaining curious and always wanting to understand why. I ended up focussing on chemistry and physics because of the ability to predict and explain what would happen. This contrasts to my early experiences of biology, which I found was based more on observing and classifying.
In the end, chemistry won out because practicals were more exciting (burning something will always beat measuring how high a ball will bounce!!).
How did you get here? Tell us about your background.
After a PhD in organic chemistry (and yes, a few unplanned fires), I joined Pfizer as a medicinal chemist, eventually leading teams across a range of endpoints from early lead discovery through to delivering candidates for development.
One recurring theme throughout my 20-year career at Pfizer was my desire to fail less often; so I spent considerable time trying to predict PK profiles and off-target toxicology ahead of synthesis. When Pfizer closed Sandwich in 2011, I moved to Lhasa with the hope of finally being able to predict those key causes of failure. While we’ve not finished yet, we are getting better and better!
My background in the pharma industry is something I particularly value; having been an end-user of models, including Derek, I understand some of the decisions that our models are supporting.
If you could have dinner with any scientist, past or present, who would it be and what would you like to ask him or her?
That’s a really hard question, but I suppose if I have to answer then the last person I researched was the Swiss mathematician Euler. I’d wanted to understand a clinical trial design that used a Latin Square and I ended up discovering that Euler inspired the name of this model, which is also behind Sudoku puzzles. Leonhard Euler was influential across many areas of maths including calculus, graph theory, topology, and mathematical functions, but he was also able to apply this theoretical maths to real-world problems. If that wasn’t remarkable enough, he averaged a maths paper a week in 1775, despite losing his sight, and his books on physics and maths were widely read by non-mathematicians. The latter fact would give me a sporting chance of understanding some of these mathematical concepts if I was to meet him!
Do you have any advice for students considering a career in science?
Quite simply, nobody can tell you the best route because there are so many different ways to get into science.
For learning about a subject, the explosion in online resources available from universities, e-learning companies or even articulate ‘youtubers’ makes almost any subject accessible. It is fascinating to compare my journey (involving hours sitting in lecture theatres and libraries) to today, where it is possible to take a video-based lecture course anytime, anywhere, which you can even rewind and re-watch!
Similarly, it is now much easier to research different companies, jobs and even interview techniques online, and to seek out people who could help you, using communities like LinkedIn.
So in conclusion, my advice is to get out there and explore!