Disclaimer: This interview is for informational purposes only and should not be relied upon as a basis for investment decisions. In Practise is an independent publisher and all opinions expressed by guests are solely their own opinions and do not reflect the opinion of In Practise.

{audio:0:04:13} I have a friend who does clinical research and he was very helpful getting the basic lexicon under my belt to browse through this space. I found Abcam interesting because it's one of those companies where when you speak to a generalist, they're like, the science is really challenging. But when you speak to a specialist, they're like, these guys talk about their catalog and their SEO too much. I view them a little like Amazon insofar as they started by taking something analogue, making it digital and then verticalizing by getting the actual capability. That was kind of like my basic entry into the space and I'm still trying to make sure I truly like and understand the situation as much as I should.

{audio:0:05:15} I followed Dr. Rothberg back in the day because Ion Torrent was incredibly fascinating to me. Early in our history as an investment company, we had a position in Life Technologies. I followed Ion Torrent through to the acquisition in Thermo. I understand you all are working on some very interesting stuff nowadays as well. I’d love to hear about your background, your history personally, your history with Dr. Rothberg and then go from there.

{audio:0:05:49} I got into biotech right out of my post-doc and the driver for me was realizing that I enjoyed the interplay between biology and the technology itself, so putting biology into boxes was what I joked about. I actually started with Jonathan at 454 in 2000. That was just really getting my feet wet and designing some of the key components of that system. It was a fantastic group and the perfect time to join; before anyone knew what next gen sequencing was.

{audio:0:06:42} Then we sold that to Roche. I went over to RainDance Technologies which was droplet based, one of these kind of microfluidic companies. From there, I went to Ion and rode that through the sale to Life Tech and then stepped out right before the Thermo acquisition. I went to a company called CyVek which was up in Wallingford, Connecticut and had a little desktop biomarker discovery unit that was antibody based. This was really my first foray into the dark side of proteins instead of DNA.

{audio:0:07:26} That’s where much of my antibody experience that I can talk about directly comes from. There we were creating an instrument that also had a consumable chip and that chip allowed for highly multiplexed, low sample volume analysis for antigens, whereas with companies like Luminex we have a whole mess of beads and each one of these beads captures a different analyte. So you can have a mixed population of beads and some are detecting Interleukin-8 and some nine and some six and some TNF-α, and you throw them all into a pot and they're all individually barcoded but there's also, cross-reactivity that occurs between these antibodies. There's always some level of non-specific binding.

{audio:0:08:28} That causes a loss of sensitivity in general, but obviously not so badly that the technology doesn't work. People use it by the bucket. The CyVek system had these parallel paths and you would put one sample in and then it would get routed through these parallel fluidic channels. In each channel there were specific antibodies that you would be testing for an antigen, so you could test each antigen independently but you used a very small amount of sample which was important for things that were biologically relevant. That's where I first came into contact with how do you screen for antibodies? How do you select the antibodies that you're going to use in these systems? Who do you get the antibodies from; all of this kind of stuff and the scalability of that aspect?

{audio:0:09:22} We then sold to R&D Systems. I was part R&D for a little while but really it didn't make any sense for me to be trying to teach R&D Systems how to do antibodies. They had people who had far more experience than I. Then Qiagen hired me to lead the GeneReader launch program there. That was the next three years which was running the system and then taking over the site administration and being the site lead, then came back to work with Jonathan again because it's a fascinating experience.

{audio:0:10:06} So at this point, he was no longer doing just single companies. He now had the incubator, the 4C incubator and so I think at last count he's got seven or eight companies that are under direct management there. I came in to work for Quantum-Si. COVID hit and I left Quantum-Si to work for Detect because it seemed timely and we needed to build up a group. We went from no people to a company of 150 people with an at-home COVID detect kit that is currently under EUA testing right now. That was a fun project but then IsoPlexis called and really gave me an offer that just was too good to pass up, so jumped over there.

{audio:0:11:00} Very interesting. You’ve been involved from the nascency of working with DNA and then down into the proteomics level and into next gen, so everything.

{audio:0:11:14} I've been so lucky; what an amazing time. I was able to get into next gen when it was first starting and now, I really feel that I'm in proteomics at almost exactly the same stage. With quantum, coming up with a protein sequencing, that was just an amazing kind of application that really dropped. It wasn't the intended application for the company at all initially. We were looking at long read DNA sequencing and then the ability to do protein sequencing came up and that was a total greenfield. Nobody can do that and so that was a very reminiscent pivot. It's been so much fun. I don't stay long. Usually it's about three years or so and then it's time to go find some other challenge, but the opportunities have been so entertaining. It's just been an amazing run.

{audio:0:12:21} Could we dive into first your experience at CyVek where you were actually purchasing the antibodies and involved with specific antibody tests. Given you sold to R&D Systems, they may have been your primary vendor?

{audio:0:12:42} It's interesting. There was a perception of a hierarchy. In R&D you were going to pay top dollar, but you were going to get quality. That was the top tier antibody as far as reproducibility and quality, but you were going to pay through the nose for it; frankly, you pay through the nose for all antibodies. The bottle that you get the antibody in is infinitely more expensive than the antibody itself. The margin on this thing is insane. It's biblical. It’s a great market to invest in. Having learned what I've learned, it's just kind of eye-popping. I used to think that polymerases and stuff were a crazy markup, but antibodies are just insane.

{audio:0:13:56} For R&D Systems, we used a lot of them, but Abcam was a very strong runner for cell signaling and CST; they were also quality. What would drive it would be trying to find who made antibodies for your specific antigen of interest; it was a very strange system of testing where you never knew what antibody was actually going to be the best for the antigen that you were using. You would order antigens from everybody, whoever had it, and you'd bring them in and you'd do a variety of either ELISA testing or whatever, look at the sensitivity, look at how well it worked in whatever buffer you were using which might shift.

{audio:0:14:54} If you were multiplexing, it suddenly became an added layer that not only did that antibody have to work well in a given buffer, but it had to work well enough in conjunction with these other antibodies because you might have to find a buffer that they all worked well in, but none of them were at their 100%. There was very much a kind of horse trading. Then you would want to change a panel. I’ve got a panel that's got five antibodies, six antibodies in it and somebody wants to replace one for another. You might have to reoptimize everything and you might need to move out, replace some of your antibodies with different vendors’ antibodies because that new antibody that you've added is really particular about what buffer it needs.

{audio:0:15:45} I expected it to be a lot more predictable coming from the DNA world. In the DNA world, there are rules, dammit. You can make TCR primers or whatever and they just work. But there's really this empirical settling on what will work best and it's very fluid.

{audio:0:16:20} We used a lot of R&D, but also we’d use Abcam. You’d order a lot of them; you’d test them and sometimes you would be forced to only go to one specific vendor because they were the only people who made that antibody for that specific antigen that you're looking for. So you might have to go to some of the Chinese vendors or Santa Cruz Biosystems where everybody winces.

{audio:0:16:50} I’ve heard their alternate name.

{audio:0:16:55} Yes, so you're familiar. Let’s just be kind and say that there's a wide range of expected quality that you may get. Then there's the second wrestling match that you need to do which is when you are looking at a commercial product, it's not enough to simply get the antibody that works. If you're writing a research paper or whatever, can get a golden tube of antibody that works and that's great. That's all you need. But we were looking at being able to lock in long-term supplies and, in that case, you've got this interesting dynamic between the monoclonal antibodies which are the same antibody tube after tube after tube. They're either recombinants which are the ideal ones, so you get the recombinant monoclonals, so a different lot should perform exactly like the previous lots. Once you've got this one dialed in, it comes in.

{audio:0:18:06} That's not the case for your polyclonals. If you get a polyclonal and you inject something into a goat or a sheep or a rat or whatever and then do a bleed and pull out the stuff, rather than one specific antibody that binds to one specific spot on the antigen, you get a whole variety of antibodies that the goats produce. In many ways it's a much better solution because you're going to cover that antibody eight ways from Sunday. But once that batches out, they have to bleed the goat again. They have to stick the antigen back into the goat and, in all likelihood, you're going to get a wildly different bunch of polyclonal antibodies and so you're going to have to redo this whole titration again because things have changed dramatically. Those are the kinds of questions that you would go through as you were looking at different vendors’ offerings.

{audio:0:19:07} The other driver was the commercial agreements that came out of the different companies. Abcam were very flexible, good to work with. For R&D Systems, I love the guys, they're great but they are rapacious. You talk to them, you say I want to scale up. I want to get a hundred times the normal amount, so what are we talking about for prices? They say well, find the price for one and multiply it by a hundred. You're like, how does that work? They typically add if they know it's going commercial; they'll add a percentage on because they're very cognizant of the fact that there's money being made in diagnostic kits and they want their piece of it. To my knowledge, from the producer side, Abcam has a much more reasonable relationship with people who are buying product.

{audio:0:20:25} That's pretty interesting because one of the critiques I'd heard of Abcam is that, when you break down the revenues, 90% would be what are called research use only. The other 10% would be clinical, for diagnostic applications, and the critique I'd heard generally is they're not quite as prepared with their salesforce in pursuing the FDA processes for a diagnostics effort.

{audio:0:20:56} That’s a great point. I'll tell you what I've heard and it comes with a huge grain of salt so you'll have to definitely do your diligence and check to make sure that this is actually true. I've heard that the vast majority of the sales of any of these companies is to the NIH; these independent researchers who buy a tube, they don't really care that it's $300 for 100 micrograms because they're going to get a tube of this stuff, they've got to use it and then they're never going to use it again. The whole concept of bulk, the whole concept of lot-to-lot variance, is not really a problem for the vast majority of the research team that's using it and hence this concentration on research use only.

{audio:0:21:41} Considering the amount of money that has to be invested for generating your antibodies as diagnostics, I would be curious to see if it makes sense. To the expression systems, to the money that's involved, to the scales that you have to do. That's brutal. I'm sure the margins are still pretty tasty, but just the sheer headache that you have to go through for that is substantial.

{audio:0:22:26} I don't know the answer as to what is the relative markup that they can expect to ask for to cover the headache that's induced in all that. But for us, we were always research use only with an eye on going diagnostics. We didn't have a diagnostic and so our thought was, that's a bridge that we’ll cross when it's good and on fire. I can't help you on that one but I do know that it's wildly expensive, like a couple million bucks to get your antibody through the GMP processing.

{audio:0:23:24} Effectively, if I understand right, you said R&D basically acknowledge the fact that you were inevitably going down the diagnostics route and wanted to price accordingly and ask for their piece of it; but Abcam’s approach was you could just buy quite simply from the catalog, didn't need extra handholding or anything like that.

{audio:0:23:45} That’s my recollection. I think R&D's jump was they didn't even care that we weren't diagnostic. We were just a commercial, we were reselling. Just the fact it was going into a commercial product, you would assume they would say, this is awesome. If you guys are going into a commercial product, if you're successful, we're successful. But I remember they were hitting us with a pretty substantial royalty on using any of the antibodies going through there, which is even more interesting when you consider, to my understanding, the relative whack of IP protection on any of those antibiotics. There’s no way that they could patent every antibody in their catalogue; it’s just going to be prohibitive.

{audio:0:24:48} Then I think now, legally, it's probably a little questionable whether that would even stand; here's an antibody to this. You could probably patent the use case as a diagnostic. Really, they're in business generating stuff that you could probably generate yourself except that the vast majority of the market has zero interest in doing it and so it's just easier just to buy it. I guess I feel that Abcam’s flexible response to commercial use in our experience really opens up a kind of a preferred vendor status relative to R&D and we never really saw differences in quality between Abcam or R&D. You were never concerned that you were buying stuff from Abcam.

{audio:0:25:46} Could you maybe quantify the price gap between Abcam and R&D?

{audio:0:25:54} I can't. I think probably for the actual antibodies the list price was very similar. We can jump on the website and take a look and I think it's spitting distance. But Abcam were the people who'd be like, while you're evaluating, we'll knock 25%, 50% off for the stuff that you're bringing in, so they were really open to work with us in a way that you weren't getting any deals from R&D.

{audio:0:26:30} That's very interesting. It makes me wonder if there's more actual diagnostic work going on that appears as research only with Abcam and accelerates their path into the diagnostic space. They've spoken a lot about a big opportunity emerging in proteomics and that's clearly their ambition, but I just can't wrap my head around how they get there.

{audio:0:26:58} Everybody wants to talk about it because Wall Street expects you to talk about it, especially with all this froth about proteomics at the moment. My opinion – with no data to back this up – is we were all excited about the DNA world diagnostics from sequencing and you look at how many NGS tests are actually diagnostic and approved. There's some smattering of them but, really, all the heavy lifting is being done by qPCR, because it's cost-effective.

{audio:0:27:48} It may be that there is a bigger drive to say that you're working on the diagnostics avenue for proteomics because nobody really knows what that's going to look like. For example, quantum with protein sequencing. I'm kind of challenged to think of the diagnostic tests that requires you to sequence somebody's protein and then even if they did, what would you do about it, unless you're going to edit it? That seems pretty academic. Very cool, awesome technology but still I don't see a quick path to the clinic.

{audio:0:28:33} Other areas, definitely you would want to track various protein build-ups and concentrations, unless we're just doing straight up old ELISAs, and ELISAs have been around forever. Whatever method they're using for diagnostic ELISA tests obviously is working just fine and maybe you don't need to have your full pipeline GMP approved for one of these tests, I don’t know. This is clearly indicating that I'm a technical boy rather than marketing and such, but it might just be that they're waiting to see. I suspect you don't just start converting your entire catalog over to GMP level antibodies without knowing in advance which ones are going to be the ones that will get approval.

{audio:0:29:47} My perspective from what I've gathered of the companies themselves, is a lot of what you referenced where, in research use, you're basically selling one bottle and that's it. In the diagnostic, you have this kind of royalty revenue stream where you guarantee yourself aftermarket recurring revenues and that's why everyone loves subscription businesses nowadays?

{audio:0:30:17} Yes. I would be interested to know what percentage of R&D’s antibodies are diagnostic grade. I certainly couldn't answer that offhand but I would think it's small.

{audio:0:30:41} I spoke to someone from R&D recently and they claim that their antibodies are almost all geared for that and are far higher quality than Abcam’s; though only someone in R&D has told me that they are much higher quality than Abcam’s so far.

{audio:0:31:02} That might be true. It just might be that that higher level of quality is something that's invisible to the user. It might be that doesn't really matter that much. The differences between two antibodies aren’t going to come down to the quality. It's going to come down to the larger issues like, firstly, do they have it or not and then, secondly, does it perform or not? You can bring in your diagnostic kit, your FDA submission; you can bring in things that aren't GMP but you test them as they come into the facility. I think there is an avenue through which you can use your RUO antibodies in a diagnostic device; as long as you have sufficient tests, that should satisfy the FDA.

{audio:0:32:10} That makes sense. I think that's consistent with what I've heard as well. I'm wondering, just generally, what percent of your budget was spent on ABs, as you were researching?

{audio:0:32:27} I’m just thinking back on what we were doing. It’s a nuanced question because, for the development effort, we're making an instrument and we're making a chip. We're developing all of these things and we're also developing the assays on the chip. There's a development process, so let's set aside the instrument. Let's not consider the instrument and let’s consider a finished chip so that we're not talking about where we went through all these different prototyping and molding tools and all this kind of stuff which are just massively expensive. But now we've got a chip that works and someone calls up and says hey, I want a brand-new panel.

{audio:0:33:39} Let's say that you're buying at least two of each of these, so you're talking about a five-panel assay, just for the sake of argument. Let's say that it's $300 a tube and then you've got your detect antibody and let's say, that it's a five antibody but you're going to get at least three to five antibody choices from every additional vendor. Now suddenly, it's beginning to add up. At the end of the day, when you’ve finally made your selection, you buy a ton of antibody but very little goes on that individual chip.

{audio:0:34:37} You could imagine that the antibodies on a chip could be like 50% of the cost of that chip, for example. Then other things are going to drive that such as, do you then have to modify the detect antibodies? How are you going to visualize? How are you going to find out? Maybe these are fluorescently labeled; you conjugate your detect antibodies. That's going to add a lot of cost to the antibodies that's reflected as the antibody portion, but you might be doing that in-house or you might be buying the conjugated antibodies from the supplier.

{audio:0:35:15} It’s a tough question to answer. I would say it's typically a large driver of the price. Whenever I'm involved in companies that are using antibodies, finding ways to reduce the cost of the antibodies in the product is usually a high priority.

{audio:0:35:37} That was going to be my next question. When you spoke about the price concessions Abcam would make on getting you started with certain ABs, what was more impactful there, the warmness of a sales relationship and doing that and ingratiating themselves with you or was it very budget conscious motivation?

{audio:0:36:04} It was strictly budget really because to be honest, mostly you don’t even want to talk to the sales people and we know what the catalog is. You set up the deal that you're going to operate under and then you’ve just got folks who are ordering on the development teams who are just constantly submitting orders. Really, I would say that the sales outreach program is probably much more important in the academic setting where, maybe, I don't know what I need. For us, we knew what we wanted. We knew how to get it, so it was strictly driven by sales and quality. No matter how cheap Santa Cruz was, no one wanted to end up there.

{audio:0:36:59} That makes sense. You said said ELISA has been around a long time. With some of these proteomics platforms, do you have an opinion on antibodies versus aptamer-based approaches?

{audio:0:37:20} I do. It’s a predictable response. You come into the proteomics world and you're like, antibodies, they're really kind of cool. And then afterwards, you find out about these things called aptamers and everybody who's been around is, oh God, here goes another one. Aptamers are so cool. They should work. They should be awesome. You can IDT, you could print out buckets of these things cheaply and they're storable and they work time after time and you produce them. The whole monoclonal-polyclonal thing goes away with aptamers because you’re just printing out these oligos. We've never found them to ever give the performance that we're looking for.

{audio:0:38:17} SomaLogic has had what, 20 years? The patents are expiring, for God's sake. They've had 21 years to make these things as good as antibiotics and they never have, at least not in my experience. There are so many cool things about them but, in my experience, they were relegated to very specific niche applications due to the odd binding characteristics or the fact that you can couple them to a PCR primer because it's all DNA. I don't see antibodies going anywhere. I see aptamers being a niche player and I would have loved if they worked. I would be so happy to get rid of antibodies just because of all of these things; they're so stable, they're mass producible, all these things. They just don't work.

{audio:0:39:14} Do you think it's perhaps something where, at some point, you could get there or is there something structurally flawed in the idea?

{audio:0:39:33} With all of the systems at play, Cellex and all these selection methods and the ability to make a jillion of different oligos, they should have gotten them working. It’s the million monkeys and a million typewriters approach where, to have this random coil of DNA be able to give you the specificity to latch on and detect a specific protein, requires this confirmational structure. That’s just me thinking out loud and that confirmational structure is so highly dependent on the concentration of everything around us; salts, magnesium, the temperature, God knows what else. Whereas antibodies are just designed to do exactly this. Their sole purpose is to find that one pocket and the proteins to attach to it.

{audio:0:40:50} My opinion is they're inherently not going to be able to do it. Maybe on an antigen-by-antigen case, you can say hey, wow, look at this. This antigen is so damn stable that you can get a lot of thermodynamic flux going on in your oligo and still have it made that connection. When we're designing PCR primers, you talk about the TM50, the temperature at which 50% of your oligos are on or off. They're constantly moving, they're breathing, they’re bubble-forming, they're closing and you're going to be asking this weird, random structure to then have this highly specific capture despite all that. I don't see it.

{audio:0:41:47} I feel like it's a little like in my world where you talk about like 5G versus cable connection in certain contexts.

{audio:0:41:57} Yes, and it's funny because there are these two things that I kind of joke about as the things that I'm always trying to see if they will work yet and aptamers is one of them; they never do. Quantum dots is another one; never work. Quantum dots don't photo bleach and they're incredibly bright. They're these beautiful little, tiny particles that you can use to label things and they never really worked for what I need them for, but it's always this thing where you say, maybe for this application it will finally be the era of the aptamers or the era of the quantum dots.

{audio:0:42:36} Has the acceleration in sequencing technology led to an accelerating pace in biologics research and do these new processes require more antibodies? Then, more specifically, thinking in terms of the general comments I hear, is that the need for antibodies has grown linearly with the understanding of the genome, but we've got into a point where it's perhaps exponential. Instead of just studying the relationship between one antibody and one antigen, we're studying the interplay between different antigens. Is there some merit to that?

{audio:0:43:25} At the heart of it, what drives my sense is that somehow, we first studied DNA because it was studyable and then we started getting to where we could start sequencing RNA. So now we can do expression profiling and we thought this is terrific. There we go, we've got it all. Now we know exactly what's going on. What we find, generally, is that you get about a 40% correlation between the MRNA expression that a cell is doing and the actual amount of proteins and types of proteins that are out there; the correlation sucks. We know what they're making but we don't know what's actually happening in the cell. There's lots of reasons for it.

{audio:0:44:20} Some RNAs hang around a lot longer. Some are actually downregulated although they’re out there. Some proteins are really stable. Some proteins are really active. I think what we're finding is that DNA sequencing isn't going to tell you what's going on at the RNA expression level and the RNA expression level doesn't really tell you what's going on about the protein levels and activities. That doesn't mean that any of those are wrong, but you really need this kind of composite picture. We were hoping to get this token, one true ring to rule them all, be able to be wow, we don't need to do this. We just can study that.

{audio:0:45:11} But I really think that the more we start looking at expression profiling, the more we realize wow, that's only part of the picture and we need that high throughput proteomics at least to find out what's going on at the heart of all of those proteomics. Again under the assumption that the protein sequencing methods are far, far away from being able to take a cell, grind it up, throw it on one of these systems and have it tell you exactly what proteins are there in what concentration. We’ve got a good distance before that's ever even a reality.

{audio:0:46:01} The only way to look at things is with antibodies. Forbes just had an article on proteomics a week ago about how this was the next super thing; this is going to go nuts. If I were investing, I wouldn't want to put my nickel on a given technology. I'd want to put my nickel on that core infrastructure that enables all of those damn technologies. In sequencing, if you'd invested in IDT oligos, they made oligos for everybody. Who cares whether you invested in Celexa or Illumina or Ion or whatever; the company that’s the arms’ dealer for all of those people was IDT. I think antibodies are going to be the same way. They are going to drive all of these proteomics because there really isn't an option at this time other than that. You win regardless of what technology wins.

{audio:0:47:33} You’ve followed Dr. Rothberg around to a couple of places. I'm curious, what makes him special? What makes him unique?

{audio:0:47:47} I think he has a vision and there's always a story that goes along with the vision and whether that story is accurate or not, I don't know. I do hear, my son told me this or my daughter asked me that and I thought, we should create a company for that. Maybe that's true, maybe that's not true. But from somewhere, some news, he gets an idea of, I want to be able to do this high throughput sequencing. I think he's relentless in his pursuit of technologies that are health driven and are working only in human health. It’s not 100% true. At one point, he had some investment with one of his nephews or something in a financial software company. But by and large, it's all health oriented and it's designed to increase access globally to this thing.

{audio:0:49:09} With the high throughput sequencing it was all democratizing sequencing, trying to be able to reduce the cost. Again, it's still expensive but when we first started out, how much was the human genome project? Billions. Then I think we did a $10 million sequence and then a $1 million. I think democratizing might be overstating the realm, but what he does is he has a vision and when he's at his best, he says, this is where I want to be and I am going to hire really, really innovative people and I'm going to hire them from a bunch of different areas that aren't really related to what where I want to go. So they bring in all of these divergent worldviews and all this kind of stuff and he's just like, go.

{audio:0:50:03} On top of that, he has the greatest command of depth of information of anybody I've ever known. From the whole company, people would send him nightly emails with just bullet points of what they're working on and you're thinking these are all going to the spam folder; there's no way in hell that he's actually caring about this. But periodically, you'd get a response that was absolutely on point, absolutely questioning you what was going on. In the days of Ion, he would walk through the company and just stop people. Every day, he would show up. He'd already made his millions, but he would just come to the office every day, walk through the lab, stop somebody and ask, what are you working on and why is it important to the company?

{audio:0:51:13} If they couldn't explain what they're doing in the broader context, that person wasn't in trouble, but their manager sure as hell was for not communicating what the goals of the company were so that everybody understood them and the part they played. Where the process stumbles a bit, or can stumble, is sometimes the conventional wisdom on how to do things is right. He always says, I like to hire people who don't know what isn't possible and it's absolutely true. He’ll get great things out of people who do things that nobody thought was possible because they don't know; kind of like the bumblebee flies because it doesn't know it's aerodynamically impossible for it to fly.

{audio:0:52:06} But there are also times where that ends up in a bad place, where you get people who are hammering down paths that you're like, look, I've been in the business for 20 years and I can tell you, I don't need to do all the tests to say that a zero-gravity machine isn't going to work. I'm just telling you right now a zero-gravity machine isn’t going to work. Let's find another path. Sometimes it gets inefficient and sometimes he gets personally involved in actually directing the tasks of the individuals and, at that point, you've kind of lost that the beauty of having your bunch of disparate innovators put together. But he has built a company of the most innovative, creative, collaborative people I've ever worked with in my life.

{audio:0:53:04} Just a fantastic organization. It'll be interesting to see the impact of there now being eight different elements. For example, with this incubator, it's spread across multiple companies and so the percentage of Jonathan in each one of those companies drops and so it would be interesting to see whether the companies have developed or that isn't important anymore or whether they might struggle a bit because he's not kind of providing that very directed impetus.

{audio:0:53:41} In some ways when I hear this, it sounds a little like he's creating a Bell Labs of healthcare where it's all these multidisciplinarians working together.

{audio:0:53:53} I think that's accurate and, not only that, but within the incubator, people are fungible. So if you're working on one of the companies and you've got a problem, but there's a guy a company over who's an expert in this, they'll come over and work for a month on that project. Really, you've got this massive kind of cross-pollination that goes on and that's incredibly powerful.

{audio:0:54:24} Do you know of anyone else trying to do something like that?

{audio:0:54:29} No, and the fact it happened in Guilford is mind-blowing to me. I joked that I've been the remora to Jonathan’s shark for most of my career. He started with CuraGen which kind of was ahead of its time, so that was this high throughput gene sequencing company, but using the technologies of the day, and the idea being that they were doing gene sequencing for Philip Morris, for all these companies that would have a project that wanted to find out whatever. They were trying to find specific genes that could be patented. It was a massive organization at one point. I joked that, during my post-doc, I sent them so many CVs I finally got a cease-and-desist order.

{audio:0:55:30} They then spun off 454. He’s completely unafraid, but that doesn't mean he's haphazard. He used to say he doesn't mind making mistakes as long as he keeps the receipt and that you can print money, but you can't print time. Working with him, the amazing thing was, you’d bring in people from academia and they'd be really penny conscious and you're like, no way, that's not how we operate. If you can save two weeks by spending X amount more, we're going to need the time later, get it done.

{audio:0:56:33} I really enjoy working with those teams, just great people. I don't know of anybody else who's doing a similar model. What are they doing on the West Coast with Accelerate?

{audio:0:57:04} Yes, there we go. We've had people who have gone through that. I know people who have, but that doesn't seem to be exactly the same thing. I don't know if there's that same level of singularity around one individual.

{audio:0:57:24} Got it. What do you think of Quantum-Si? Any general thoughts on that? Is that worth spending some time on?

{audio:0:57:37} Yes. Full disclosure, designed a lot of the system there and own stock in it. So this is a non-biased kind of statement but they are, to my knowledge, the only de novo sequencing company; the ability to take a protein and sequence it to the individual amino acids that make up that protein. It’s a total greenfield. Right now, you can do mass spec and look at that, but you're comparing the patterns to known peaks. I really feel that's one to watch. Like I said before, I don't look at that and say, oh my God, that's going to be in every doctor's office and whatever, because I don't know that there's a direct-to-consumer application for that as it stands.

{audio:0:59:03} I think the similarities between that and DNA sequencing are overwhelming. We don't know what this technology can do because no one's ever had this capability before and it will be really exciting to see what happens and they're just making progress. I haven't been associated with the company since COVID, but they're just making great progress from the little I still hear and I'm really keeping an eye on that.

{audio:0:59:40} What would you watch to see their progression over time to make sure they're making the right progress?

{audio:0:59:49} That's a good question. The big thing is, it’s essentially built around the concept of, if we build it, they will come. If you have this technology, people will find it and that can work, but also we've seen that the world is littered with hulks of companies that built it and nobody came. I guess I would look at a couple of metrics. Obviously, how many instruments are they placing is going to be a big one. What scientific literature is coming out? What papers are being published using that? Because again, in your first wave is going to be those key opinion leaders, the evangelists who are going to start writing and saying this is useful. You can do shit with it.

{audio:1:00:43} Then the other one is going to be numbers that tell you, not only how many instruments they sell, but what's the pull-through on the reagents? That's going to tell you how many times they are turning on the machine. You can have situations like 10x Genomics did for some time – I don’t think it’s the case any longer – but everyone bought a 10x box and they all sat there and gathered dust because it was expensive. Now 10x is cranking but if you looked at the sales numbers, you're like wow, they're selling the hell out of those things. But people aren't using them and the whole model is set up where you can buy the instrument but, frankly, you can give the instrument for free. What you want is the steady reagent pool because that's where you're making all your money.

{audio:1:01:36} Yes, the Gillette razorblade model at the end, right?

Abcam & Research-Use Antibodies: Customer Perspective II

Buying research use antibodies from Abcam vs R&D Systems 

Perspectives on the opportunity for Abcam in proteomics 

Comparison of RUA vendor pricing strategies 

Demand outlook for research use antibodies 

Abcam & Research-Use Antibodies: Customer Perspective II

Former VP, Biology at CyVek

inpractise.com/articles/abcam-and-research-use-antibodies-

Why is this interview interesting?

Executive Bio

Former VP, Biology

Former VP, Biology at CyVek

Interview Transcript

Related Content

Abcam & Research Use Antibodies: A Customer Perspective

Chief Scientific Officer and Sr. VP of R&D at Metamark Genetics

Partner Interview