Imad Yassin recently joined Causaly as Vice president of Sales and Commercial Strategy following an impressive career in disruptive healthtech. We caught up with him to discuss why he joined Causaly and why he thinks the company is one of the biggest game-changers for translational research he’s seen to date.
Imad, tell us a bit more about your background
So I'm a geneticist by education. And I'm completely obsessed with technology. I was very fortunate to start my career working for the first ever commercial bioinformatics company called GCG, where I got to experience many different areas of technology like chem informatics modelling, simulation, data management and knowledge management. That was my springboard.
I went on to work for some really exciting companies, from startups to enterprise, including Inpharmatica (now Galapagos pharma) - which was the one of the first companys in our field to get featured on the front page of the Financial Times for its groundbreaking work in druggability assessment. We were doing some really interesting things there, like applying algorithms and computational methods to predict what makes a good drug.
Having had all this experience in the pre-clinical area, I also got to experience the clinical area at Medidata. I then went on to set up commercial operations for a number of companies including Integromics, which was acquired by PerkinElmer in 2017.
Why did you join Causaly? What excited you?
In short, it’s the fact that Causaly solves one of the biggest challenges I’ve seen in the industry to date: extracting information from unstructured data sources.
The work I'd been doing previously was in translational medicine and organising structured data - our clients were in pharma, biotechs and research labs. They would tell us we were doing a great job at making structured data sources accessible - but what about unstructured data? There's still such a wealth of information in medical texts, publications, journal reviews. They wanted to access it.
When Causaly initially approached me, my first reaction wasn’t a positive one. I said look, this area is very saturated. A lot of people are doing really cool things in text mining.
But what really impressed me wasn’t Causaly’s ability to look for co-occurrence, and word frequencies, but the fact that the technology is acutally mimicking how humans read cognitively.
Causaly is looking to understand the context of that data in the sentence itself, extracting evidence and causality. It’s one of the only companies that can truly claim to be able to do that.
On top of that, the team has invested in building an incredible user interface.
Up to 25% of a bench scientist’s time is spent reading and trying to find out what’s publicly known about their drugs or compounds. Text mining and AI tools can help narrow their search but they’re usually only accessible to a very small number of specialists.
Causaly has come in with an incredible niche by making complex research questions simple. Causaly hides the complexity of the AI, the text mining and machine reading capabilities with an interface that anyone can use to extract their evidence quickly to get the answers they need.
How does Causaly help solve the challenges you’ve seen in translational research?
In translational medicine your success is contingent on getting information from multiple data sources. And before you can delve into the insights, you have the silos and formatting challenges to merge them.
If you're looking for genes associated with Lupus (lupus erythematosus) for example, searching PubMed will yield approx. 9,000 articles. On Causaly, you immediately get 977 genes at a glance, without having to read or analyze one document.
Causaly machine-reads and extracts that knowledge from all published biomedical papers to date. Those 977 genes represent the entire body of human knowledge on that topic.
On top of that, as a next step, you can further explore which genes are implicated in multiple immune disorders, such as lupus, psoriasis, arthritis to find possible treatment options.
Unlike every other biomedical search tool, we don’t just find articles for researchers to read. We can answer the research question itself.
What excites you most about the future of AI in translational medicine?
For years, people have been saying that the lengthy process for getting drugs to market means blockbusters are becoming a thing of the past.
But what if you could expedite that process?
For example, with COVID-19, people are racing to find out which existing drugs or potential compounds could be repurposed and launched. AI like Causaly means scientists can quickly grasp all the latest knowledge, without spending months reading it.
AI also allows companies to play a more socially responsible role too. At Causaly we machine-read and mapped all the public data and knowledge about COVID-19 and were able to make it available free for non-commercial research, as well as our existing clients. The attention and number of clients requesting access is a great indicator of just how much one AI tool can speed up research.
Ultimately that's the race that pharma companies are in. After spending billions of dollars, the last thing you want is to delay a launch because you didn't spot a certain published side effect.
In translational research it’s vital to have the full picture of what’s known about your compound.The rate of scientific advancement and volume of papers published every day means that AI has fast become the only way to do that.
What 3 tips would you give to someone responsible for sourcing software and data sources to support their research teams?
I always recommend people start by defining the scientific problem that they're trying to solve or the question they're trying to answer.
Define the challenge associated with using your current tools and capabilities. And then investigate which AI tools actually solve this problem.
Compare tools in a really measurable way. Ask yourself - do they really improve the way I work? And once you have established a yes, find out by what percentage impact in terms of time, effort and accuracy.
Finally, ask yourself - what is my change management approach and strategy? In large organisations, bringing in new technology, without having a really good change management approach will likely end up with people using it for a while and then just going back to their old habits.
If a company’s keen to try Causaly, what’s the process? How do they get started?
Book a demo to start a trial and start exploring Causaly.
Initially we demonstrate how Causaly solves a small subset of a larger problem a client is facing. Once we show it works well, we work on a pilot project together on a larger scale, typically over 3-6 months. This stage moves quite quickly since Causaly is completely cloud based, so onboarding various teams and departments is as simple as issuing them a user account. Our in-house scientific liaisons work with users, answering any questions they might have and ensuring that they can find the exact type of data they need. Once the pilot has proved successful, we work with our clients to integrate and roll out Causaly across departments and company-wide workflows.
Causaly stands apart as having one of the most advanced semantic search capabilities worldwide, and combines it with a strong solution delivery strategy powered by our in-house experts. That’s our magic formula.