Turning Life Science into Data Science | Mark Fischer-Colbrie from Strateos

DISCLAIMER: Below is an AI generated transcript. There could be a few typos but it should be at least 90% accurate. Watch video or listen to the podcast for the full experience!

Mark Fischer-Colbrie 0:00
They can decide what to synthesize, they can decide what tests they want to run. And importantly, they get highly reproducible data sets very, very rapidly and low costs. It’s really along the theme of thinking about turning life science into an Information Science.

Alexander Ferguson 0:25
Mark, I’m excited to be able to chat today to begin, can you very briefly five seconds share what is Strateos.

Mark Fischer-Colbrie 0:34
Strateos is incredibly exciting cloud laboratory platform that is going to change the face of life science, discovery, with the ability to leverage multiple automation modules and democratize science to allow broad scale access for new discoveries

Alexander Ferguson 0:56
democratize scientific discoveries through this automated discovery causa, can we break this down? How did even this concept begin? And how has it evolved over the years when when did it start?

Mark Fischer-Colbrie 1:10
The founders of the company in 2012 came up with the concept of the cloud with the view being that the modalities that people were using for life science discovery of traditional what are known as wet labs were vastly inadequate, and then provide a good solution set for advancing discovery. In addition, it was then a requirement where a lot of folks had to buy equipment, find a suitable building, facilitate that building, hire the right staff. And with the ability to create a Cloud Lab, it makes it much easier for people to get their work done without having all those restrictions, and lack of access to the proper tooling.

Alexander Ferguson 2:05
This concept of a Cloud Lab, can you give me a good use case of what are your clients using it? How does it work?

Mark Fischer-Colbrie 2:13
Yeah, that’s a great question. That’s a situation where there’s a wide range of uses everywhere from Amgen all the way down to academics at the incubator, company types. And that’s a condition where one can access our automation platform over the web, have the ability to synthesize chemistry, automatically, which currently is done 99.9% manually have those compounds that are synthesized run through basic biological testing, all in one facility that is directed by the user such that they can decide what to synthesize, they can decide what tests they want to run. And importantly, they get highly reproducible datasets very, very rapidly, and it will cost it’s really along the theme of thinking about turning life science into an Information Science, and duplicate the kinds of workflows that are common in all other industries, but are not yet there in the life science world where a lot of the work is 1850 hours and style of operation that needs to be brought up to today’s standards. So we’ve offered that capability. And as a direct consequence, we think we can cut drug discovery time in half, we think that elements around biology where engineering biology can have a major impact on generation of new materials will have a huge impact on carbon footprint, can be accelerated. And so we’re super excited about bringing this platform to broad scale years.

Alexander Ferguson 4:04
So it sounds like this. I see two big wins here. One is the ability to start a new discovery process without needing a lab needing to do anything, you just like, I’m going to spin it up like you’re spinning up a AWS server, just like you’re spinning up a new server, as well as repeatability to make sure if someone’s going around with a pipette and dabbing and things doing there’s a lot of potential for errors. So this reduces that to hopefully it’s a point percentage versus person. Am I getting those two main wins? Correct?

Mark Fischer-Colbrie 4:37
You nailed it. Well, I like to talk about the fact that we have the opportunity to become the Amazon Web Services for Life Sciences period. And what we’re essentially doing is reimagining lab as a data center. And it’s no different than a condition of setting up a separate server farm and a different geography in order to expand utilization. And that ties back to then data streams, and collection of data that are highly reproducible, which is shockingly not there today. And to be in a situation where, because you’ve got these controlled datasets, you can now feed them into machine learning and artificial intelligence modeling programs, and get a whole other order of magnitude of leverage off those datasets. So this is a dramatic change. From what what people are doing today.

Alexander Ferguson 5:31
Are you providing that machine learning within your platform as well just built into that, or is it something they can extrapolate and do it then on their own,

Mark Fischer-Colbrie 5:40
folks can either do it on their own, or we are in synthetic biology, for example, doing a project with DARPA, where there’s machine learning algorithms that are created for the creation of new biosensors that require gene editing of bacteria that are pushed onto our platform, we perform those operations, push the data back into the machine learning models, it’s a continuous cycle of improvement in understanding for the capability to creating these new types of sensor tools. So that can be applied also in the drug discovery side, in the context that, right now we’re starting to work with companies who have designed drugs on a computer, and they need a validation platform to understand, do their models work. And as a consequence, we’ll be able to do the same kind of cycle of continuous improvement around those models to accelerate discovery there.

Alexander Ferguson 6:47
These automated lab environments, there’s two forms you run your own labs that someone can access if they don’t have their own. But you also can set up a lab within another companies to use your type of automated system, is that correct?

Mark Fischer-Colbrie 7:02
That’s correct. And that’s a situation where there’s even another version in the context that folks like gingko Bioworks, and synthetic biology world, and Eli Lilly, both engaged with Australia’s to deploy our software across their existing facilities. So as a consequence, we are able to offer a broad range of capacity and capability depending on what people need for their utilization

Alexander Ferguson 7:35
are, obviously the software platform was one of the main elements then G are you offer also creating and deploying the hardware capacity as well as far as the robotic elements?

Mark Fischer-Colbrie 7:46
Yes, and that’s one where we’ve got deep automation expertise. The person who heads up our San Diego operation, just at Novartis alone has 20 years of automation of biology. The company founders started off with automation of biology. We allied with Eli Lilly to leverage 10 years of their experimentation of automation of chemistry, which is a unique workflow. And then in conjunction with Lilly, have automated both chemistry and biology together to make a complete design, make test analyze cycle of extreme rapidity and high reproducibility. It’s actually quite shocking outpour reproducibility data points are in the life sciences. Bear did a study on 67 programs that they had kicked off based on promising data from academia. They could only replicate a quarter of the datasets that had been provided them in the course of doing just a quarter. And then there was another major study done by Amgen, where they looked at 53 landmark, cancer papers, landmark breakthrough, they can only replicate data for 653. So that that’s indicative of the phenomenon of different recipes, different people doing manual operations, and a long list of activities that have been totally avoided in almost every other industry. So we think we’re on the right path there.

Alexander Ferguson 9:33
The bar is set pretty low so you’re and you’re just like shooting it up to a whole new new level can help me understand the the best person that is implementing these these systems in the in the different companies, what’s their title? What’s their role?

Mark Fischer-Colbrie 9:48
That’s a great question. Often they’re across functional areas. And that ranges the gamut from people who are more involved in say High Throughput Screening operations that are Looking to use our platform for rapidity and lower costs than what might be done internally, for it would involve folks in what’s known as the admin area where you’re looking at, in effect, toxicology and other distribution of compound through to get to the patient, if you will. And so there’s a variety of folks at the types of institutions that have interest in the platform, both from utilization over the lab, as well as for potential adoption in their own labs, as well as for replacing contract research organizations. And the fairly manual efforts that occur within those CROs.

Alexander Ferguson 10:46
I heard you correctly, you could potentially replace a whole industry?

Mark Fischer-Colbrie 10:50
Well, I think we’re in a situation where just in drug discovery alone, this is gathering a ton of interest, because everyone inherently has learned from other industries. You need to automate, eat, industrialize, you need flexible automation, you need reproducibility, and you want to leverage on AMI lane, and in that success, exactly what we’re doing. So so the conceptual underpinnings are fairly straightforward. And this is a situation where we’re, by and large, how people are going to do this going forward. I, you know, there’s, there’s there’ll be no particular reason, with, obviously, some exceptions in larger organizations? Why would you set up your own lab? To do these things? There’s no particular need? If you look back, not that long ago, companies had their own server rooms, large IT staff. And that’s all gone. Right? That’s all on Amazon Web Services and distributed server farms around the world. So you know, that’s, that’s the parallel model here.

Alexander Ferguson 11:59
Life Sciences now gone cloud, where everybody can access and and do it. So tell me more about then the business model? Is this something that people pay for is like, a yearly contract? If they want to use one of your labs? Is it based on the per project base? How’s it work?

Mark Fischer-Colbrie 12:14
Yeah, it’s more on a per utilization basis. So as people are looking at synthesizing chemistry and running a variety of tests, that’s how often how much they use the platform is. On the services side. If there’s a software deployment in their lab, it’s on a subscription basis, if you will. And then we have a little different variation for those who want us to build and set up the automation for them, in addition to the software deployment, what can you

Alexander Ferguson 12:45
share about your roadmap next year? So where you guys are headed, what are you excited about and long term?

Mark Fischer-Colbrie 12:52
Yeah, well, first of all, we’re incredibly excited about the fact that we’ve got five automation modules up and running in Menlo Park. And we have now 10 automation modules set up and running in San Diego, our chemistry modules are coming online. And we have a number of those online already. And we expect the balance of the chemistry modules to be up and running by first quarter. So inherently, we’re building out that platform that involves a ton of software in terms of communication directly to the instrumentation, the automation module, and extremely sophisticated scheduling across the whole platform. So we can run many, many projects simultaneously, to keep that lab humming. And so on top of that is the communication over the cloud, and along with proper information security, to be able to manage those workflows. So that’s our near term focus. The future ability is to replicate these data centers, if you will. And there are a number of people that that have, are speaking with us about doing that, and adding additional technical capabilities to the biology workflows, which we which we fully intend to do. So. Then, as the questions go, lay everything out and expansion.

Alexander Ferguson 14:18
Where can people go to learn more and what’s a good first step for them to take if

Mark Fischer-Colbrie 14:21
they go to strangers.com There’s a wealth of information there. There’s also great YouTube videos including the secret video that ties in our artificial intelligence activities. And there are blog posts at Australia’s where you can get further sense of customer utilization and the extreme benefits of the platform whether somebody is from a large pharma or whether they’re an academic or from an incubator company. So we’re excited to be able to share data and information about our capabilities. Because of I, what I see is the opportunity for really advancing science and really advancing therapeutics, I think the the model of, you know, $2.5 billion in 15 years to get a drug to market that is broken. Nobody can pay for that. And platforms such as ours can have a big improvement on those opportunities. So we look forward to spreading the word about our capabilities.

Alexander Ferguson 15:25
That concludes the audio version of this episode. To see the original and more visit our UpTech Report YouTube channel. If you know a tech company, we should interview, you can nominate them at UpTechreport.com. Or if you just prefer to listen, make sure you’re subscribed to this series on Apple podcasts, Spotify or your favorite podcasting app.