CellPort Connect - CC 102: Standardization by Design

So you mentioned that CellPort came out of audits with the FDA. What was that experience like? Experience was like with any audits, is it was kind of painful in the early days of absorption.

When our work with cells was being submitted for supporting regulatory submissions. We were using cell lines that had high passage numbers, so they were in culture for quite a while as a result. The auditors came in and needed to understand that the cells that we used at a particular place and time were the same cells that we used to validate our system and those audits really highlighted a a number of challenges. One was the traceability back to the origin of the cells and when I'm in passage 30 or 40 passages, it means a lot of different people are touching these cells over time and they wanted to know. Well, listen, did the cells change on you? So what you started with and what you finished with, are you sure it was the same cells? Well, we had the notebooks, email communications, the Excel spreadsheets, the QA team all bringing their components into the discussion, but it was a lot of variability in the formats in which we presented the data and that sort of highlighted the obvious issue is can I tie it all together in one system and enable an auditor to quickly go back in time and determine what reagent that I used? What was the composition of it? What were the vendors who supplied those reagents? Who was a scientist who did that? All of those things were questions we were asked, let alone all of the analytical testing confirmed the cells were performing as we were working with them over time. This really highlighted the need to sort of say, listen, can we tie all this together and in doing so, it raised other issues is how do I kind of format or standardize what I'm capturing with that? We put a lot of effort in to understand how we would design a system that would allow us to do that and I think standardization is one of the challenges especially in a a research environment because many of the researchers that we've interacted with that is kind of a low priority of the standardization of capturing data and I think it actually goes to the concept of FAIR with your background in chemistry and expertise in the FAIR world. Maybe you could comment on that a little bit and your learnings from the chemistry world and as you're seeing them that need to be applied in the biology world.

It's interesting because I was thinking about how chemistry world, small molecule world that I came from before coming to CellPort, so vastly different from the cell-based world in a small molecule world. You have the ability to look at something that is maybe a batch of the same thing that you've made before, but there's not that time-based aspect that you mentioned with paging of cells with cells. These are growing and living things that go from one generation to the next and that's not something that the small molecule world really needs to deal with. From the FAIR perspective. FAIR is an acronym for findable, accessible interoperable and reusable. Those standards help to make sure that the data that are in the system are not only readable by humans, but also more importantly in the future machine readable. That's really what FAIR is all about is ensuring that that content that is put in there is done in a standardized way that there are globally unique and persistent identifiers that the metadata is well defined. It's using, using the appropriate standards for either the ontologies or the communication standards. That there's some basis that makes this information much more reusable and that's more to the understanding of of cells as it was to small molecules and it's definitely underpinning the entire system of CellPort in in your time.

Now, it's been years in this space and learning all about cell-based assays and all different types of cells. Do you see the challenges within the FAIR? Can you comment on ontologies they're, they're definitely ontologies that come to bear.

Some of the ones that are relevant to us in particular, the Atro Foundation which was founded by a number of pharmaceutical companies and related companies for software and instrumentation. Definitely is a foundation that brings to bear a lot of standardization for those ontologies, but there are biological assay onto that exist that, that are also relevant. It all helps again, the information that's in the system become much more standardized and readable, not just by humans but by machines, which is the ultimate end game for all of this to create a large system of data, big data that can be used in AI applications and deep learning applications before you know, as, as we look at what the future is getting the ontology, right is, is a big part of the FAIR components.

There was a paper that was written and published in 2015, 2016 by Leonard Friedman group. They talked about the challenges of reproducibility. Can you kinda comment on that and as you see it again, coming from the outside into this world, sort of the impact the paper had especially coming from a chemistry. That's right, that was a dramatic paper to read the fact that Leonard Friedman of course, is the Chief Science Officer now at the Frederick National Cancer Lab at the National Cancer Institute.

That paper demonstrated that for preclinical research, the preclinical research spend fully 50% of it is irreproducible research and that means that half the dollars that are spent are essentially wasted. The bulk of that information. The bulk of of that research spend comes from problems with biological reagents and other materials and that for me, as a newcomer to the world of cell-based activity, was just absolutely stunning that there could be so much irreproducibility. Small molecules are easy. Cells are not and to combat that irreproducibility, you really need to get your act together. You need to know what you're doing. Who's doing what. what you, what materials you're using and track and trace all of that to ensure that, that you can bring that reproducibility down and that's precisely the problem that you face in absorption systems was focusing on cells as the key point for problems with, with reproducibility.

We looked at the business model that we had as in the contract business, a contract research business. So people would outsource their work to us and we were using cell-based assays to help them characterize small molecules, drug like properties in particular in the beginning and one of the challenges in the business model with the, the CRO business is, you know, you get three things. What we tell our clients, you get, you get three things, you get to design your experiment. The experiment has to be performed and then you get the interpretation of the experiment and usually they can afford two of those three things and so what happens is, is with limited budgets and the, the interest or desire to learn as much as you can. It, it's problematic. It's problematic because most companies, these little companies don't or even the big companies don't have the budgets, especially in the research phase to do all the work they want to do to answer the questions they need answered. So you start cutting back on the experimental conditions. You want one thing, we wanted to make sure that when we worked with a company that if we completed the work and issued a report that the company could come back in six months, do the same experiment, pay us to do the same experiment again and they get the same result and as we know that can be a real challenge, especially when you're on a limited budget and you've cut corners on the the actual number of conditions and time points and all of this. So when we looked at our operation, we said, where are we most exposed in this whole performance of all these services and cell culture kind of rise to the top where we said listen, we want to know everything we can about the cells on the day of the experiment that they were performed and to do that, I had to put systems in place or the team had to put systems in place to make sure that all our cell culturing conditions in the QC was sort of a top priority. And that's what we did and that actually again, led to this, this concept that we need to be able to make sure we understand we trace we could trace what we've done and it was transparent. It was visible because in the days you cannot repeat reproduce those experimental results. It's difficult for everybody.

When, when you think about the the history that you had in absorption systems and implementing that kind of system, talking to you about implementing the system at absorption precursor to that was Excel spreadsheets from the discussions that we've had with potential customers and customers. At CellPort, one of our biggest competitors, it seems is Microsoft Excel that people are trying to store that kind of information in a place that's really not meant for it to be stored. It's not scalable, it's not trackable, it's not particularly traceable and it's definitely not 21 CFR part 11 compliant. Can you speak to that experience that you had going from paper, lab notebooks with Excel and having that as the Genesis of CellPort?

Yeah, if one word comes to mind it was painful. It was painful. The, the, the pain came into a couple of issues is we had sought out a solution to this problem and we looked in the what was commercially viable and we, we couldn't find anything that was number one. Number two is we also had a process by which everyone knew how to perform and so when we stepped back and said, all right, we're going to make this transition, we're going to try to create a system that solves these different problems. It was one the buyer build, buyer build and there was just nothing available that we saw that could do what we needed to do. So we built, in doing so, it took a while because we ran in parallel. We built a system, but we kept all of our notebooks going at the same time until we could validate and transfer over and this went on for probably a year and a half. I would say it was about an 18 month time frame, but when the transition was made, when the system that the scientists had actually designed, asked for and then it was delivered to them and there was a lot of give and take in this process. It wasn't just like we had a blueprint and it was all clear. There was a lot of start, stop, pivot, start, stop, pivot. What became readily apparent was without the scientists and the daily operations, it was going to be hard to deliver something without that knowledge. Well, we delivered it and a couple of things came from that once the system was delivered and you could sort of push the notebook aside. Efficiency was one of the most notable aspects of this and, and, and the mindset of we're not going backwards. Now, remember our people had to go through 18 months of this. If we look at clients going forward, that transition should occur in eight weeks and so it's a much shorter time frame. And that was number one. Number two is what we saw as another aspect of it was the transition or the translation of data I had already generated. I generated a notebook in a lab and I go back to my desk. I enter that data into an Excel, a word, word doc or whatever and then I share it with other people and it gets changed and then somewhere someone embeds a Macro into an act into an Excel spreadsheet, they share the spreadsheet. The macro gets changed. All of these things we saw happening and we said, how do you lock it all down? And so this ability to lock this down, at first, we the, the, the scientists thought it would be somewhat restrictive. Well, once I entered the data, once I'm not recording a notebook and then going and translating that and generating a report, it's all done. There. The second it's entered, the data is entered that was somewhat transformative time became available.

One of the things that I'm most excited about, I've seen version one of CellPort that you're talking about. I've seen version two. I've lived through version two. Now we're about to launch version three and that I think has some of the most exciting and valuable features. Not only in this software, but in any software package that I've been associated with the ability to have the full protocols that are no code version control, that administrative users can put together as precise or as, as flexible as they need to be so that their work can be done by the people who need to be done or who who need to have it done. That those people can also add notes with at notification. I make an observation. I can add notice at notify you make an @-mention for you to look at this particular problem that I've seen. You can respond to that note. There are all these capabilities that we're adding the CellPort that are really taking it to the entirely to an entirely another level.