September 6, 2016
“Going from academic research to industrial research, came with the realization that bulk data drives industrial research, whether its pharmaceuticals or agrochemicals. In academia you might focus in on a very specific issue and track data very carefully, on individual data points. In industry it's all about bulk data, and it's about drawing plots of multiple compounds vs multiple data points, and trying to find correlations and find the outliers and find the reason why compounds are doing what they're doing. CDD Vault is a great tool for bulk data analysis.”
Andrew Calabrese
VP of Chemistry at AgriMetis
Andrew Calabrese joined AgriMetis in 2014, upon its formation. Since 2002 he has worked in a variety of project and managerial roles for Pfizer and Celgene in a broad range of therapeutic areas. He has worked on and led projects that have delivered 9 small molecule novel chemical entities (NCE)’s into clinical development one of which in now in Phase 2. He gained a PhD in chemistry focused on natural product total synthesis from University College, London, following on from his undergraduate studies in chemistry at Oxford University. His post-doctoral work was on novel applications of the Birch reduction to heterocyclic compounds.
Interviewed by Whitney Smith, Collaborative Drug Discovery, Inc.
Today's Spotlight Interview is with Dr. Andrew Calabrese, VP of Chemistry at AgriMetis, LLC. AgriMetis is a customer of CDD’s with a somewhat different profile then many of our biotech and academic partners, in that AgriMetis is focused on the development of innovative crop protection products. With fascinating science and informatics needs that will sound familiar to those in drug discovery, we thought that an interview with Andrew would be interesting to our always growing CDD community. Andrew, thank you for joining us today.
ANDREW CALABRESE: It's a pleasure to be here.
Could you tell us about the company’s background?
AC: AgriMetis was formed about 2 1/2 years ago as a joint venture between two companies. One company is a seed innovation company based in Baltimore called Acidophil. The other company in the joint venture is Syngenta Ventures, based in Raleigh, North Carolina. AgriMetis was founded to take certain technologies that were discovered at Acidophil and apply them to crop protection – looking for new ways of discovering agents that could be used in the field by farmers to increase yields and protect their crops from the various pests that are out there.
And what's your background?
AC: I'm an organic chemist by training. I received my undergraduate degree at Oxford University in the U.K., then did a PhD on natural product total synthesis at the University of London [with Professor Karl Hale], before ending back up at Oxford University [with Tim Donohoe] doing a post-doc. During that time I was recruited into Pfizer in the UK. I worked there for about 6 years in a variety of roles, starting out as a bench chemist and moving into chemistry leadership and ultimately project leadership. Slightly prior to the site closing down in the U.K., I transitioned to Celgene out in San Diego onto their chemistry management team, where I had a variety of different roles in the 5 or so years I was there. Then about 2 1/2 years ago I took the leap into small biotech by joining Acidophil out in Baltimore Maryland.
What's the end goal of AgriMetis' work, both in the short term and in the more audacious long term?
AC: AgriMetis is working to identify a compound that it can progress into field trials. And unlike pharmaceuticals, where you might be tracking one primary endpoint in your biology, and then a number of subsidiary endpoints mostly related to pharmacokinetics, in crop protection you're looking for a broad spectrum of activity. It's a bit like working on an anti-infectives program, where you might be following several different types of fungus or pests. So each compound has multiple data points generated for it, up to…I think we’re currently running about 12 for some of the more advanced compounds. So, in the short term, by the end of the year we hope to find a compound that can progress into field trials, particularly an insecticide.
Since you've had such broad range of experience, what are the differences between small biotech, pharma and crop protection, in terms of working in and managing each type of environment? What are the requirements for conducting research in that environment, and what are you looking to accomplish at each phase in each organization?
AC: Big pharma and large biotech companies tend to be quite similar. The primary difference tends to be the volume of data that's generated for a given compound. Smaller biotech companies can't afford to generate a significant amount of auxiliary data, as they do not have the automated systems put in place. Therefore, there is a lower amount of data, but the value of the data is much higher.
Would you like to expand a bit more on the work you're doing with the partnerships and the CRO ecosystem and how you put those to use?
AC: AgriMetis is essentially a virtual company that generates intellectual property. The folks who work there generate the groundbreaking ideas that take the company into novel crop protection agents. The work is almost entirely outsourced. We do have a small lab in Gaithersburg, Maryland with a couple of bench scientists doing a little bit of work, but the vast majority of our science is done externally. We have various other smaller contracts that we've placed with bespoke groups in the US and in the Southern Hemisphere, because as we move into field trials and as we move into testing compounds on things like bees, for example, we can take advantage of the different seasons in the different hemispheres to try to accelerate our programs.
I’d like to follow up, by talking a bit about your journey in finding an informatics system for AgriMetis. Were you using any other system to manage all this external and internal data at AgriMetis, were you looking for a startup system?
AC: When AgriMetis got started, we were essentially using Excel spreadsheets, which was fine at the time. When we only had a handful of compounds, it was fairly straightforward to manage data and manage shipping using Excel spreadsheets. AgriMetis is now up to 600-700 compounds. Excel spreadsheets would be impossible in that scenario.
I have to admit, we didn't look around too much. At a prior company I had done an assessment of the tools that were out there, and CDD Vault was my recommendation. So we went ahead and purchased the product.
What are the primary goals for bringing in CDD Vault at AgriMetis?
AC: So what we're hoping to achieve with CDD Vault, in this form of usage, is three things. The first thing is that it will be the repository for all our data, and it will be the way we can communicate with the CROs and collect data from them. Secondly, it will be used, and is being used, in the way you would use it in research – data mining, plumbing out data, etc. The third use we can envision at some point, using it as a resource for due diligence and collaborations.
You mentioned before the security between your different collaborators or your ability to control what was being seen by those collaborators. Was that an important component of your selection of the product?
AC: Absolutely. That was an essential component. We had four essential components. First that they had to be web-based in consideration of collaborators - particularly collaborators located overseas. It's easier if the product is web-based. Secondly, there had to be adequate security trimming. Thirdly, it had to be a product we could use as a repository but as a secure repository, which is a little bit different from the security trimming. A secure repository, is more along the lines of "we're going to put all of our information in here, we want it to be safe and secure". The fourth criteria was that it had to be hosted by a reliable third party - we didn't want to put a server in house, because we are not set up to do that. CDD Vault met all those criteria.
Now that you have CDD Vault what are the top two or three things that you like about it beyond those requirements that you've already listed?
AC: Sure, as I’ve alluded to our handling with data generation is a little bit out of the ordinary, in terms of CDD Vault typically being used for pharmaceutical companies. The amount of data per compound is slightly higher, but also the way the data is generated is in parts per million activity rather than a concentration gradient. And Charlie Weatherall and his team have been so very responsive helping us work out a way to put that data into CDD Vault in a way that it is displayed correctly and usefully. I've been very impressed with the customer service element of it - you guys have been very helpful in that regard.
In terms of actual usage, if you look at the log-in statistics AgriMetis would probably qualify as a light user. We didn't use it much in the first year. That’s not a reflection on the product, it's more of a reflection on the fact that I took the decision that I wanted to set it up earlier rather than later. I didn't want us to be scrambling around. So this year we will be using it more, particularly for due diligence, that sort of thing. But so far usability has been fine. Customer service has been excellent. We are all up to date.
And do you feel like you've avoided some of the common pitfalls you've seen or heard about in other organizations because of adopting this technology early?
AC: Yes, exactly.
What are the top two or three things that CDD could do better? What are the things you'd like us to put on our roadmap
AC: That's a difficult one. I don't think I have anything because I like the product. It works for me. I don’t really have anything that I can think of.
Well if you think of anything feel free to send it our way. We're always working to improve the product.
Is there anything else you want to communicate out in this interview?
AC: The main point that I hope would come across is that while we're talking CDD Vault, we’re talking about an IT system, and we’re talking about folks sitting at a computer logging in stuff, that actually the interactions that we've had with the people at CDD have been excellent. Very helpful, very responsive, not just saying “oh, do it this way”, but being very educational. So the people side of things at CDD I rank very high.
That's fantastic feed-back. Thank you very much
This blog is authored by members of the CDD Vault community. CDD Vault is a hosted drug discovery informatics platform that securely manages both private and external biological and chemical data. It provides core functionality including chemical registration, structure activity relationship, chemical inventory, and electronic lab notebook capabilities!
CDD Vault: Drug Discovery Informatics your whole project team will embrace!