John: Zach, it's great to have you here to discuss Biocartography. I'd love for you to share a bit about your background and how you arrived at where you are today.

Zachary: Yeah, thank you for having me! So in 2017 I joined Roy Wollman's lab at UCLA, and at the time, the field of spatial biology was really changing. The first MERFISH and SeqFISH paper had just come out a few years before that, and so the lab was excited about that. We were actually the first team to recreate MERFISH outside of the original lab that developed it. Spending time building the systems and tweaking the protocols really got me interested in method development. We ended up making some really cool advancements to the tech and so it made sense to take what we learned and bring that to the commercial space.



John: What sparked your personal interest in spatial biology?

Zachary: So when I first joined, at the time, there had only been MERFISH done on cell culture, so at that time there actually hadn't been any tissue MERFISH. And so my original project was focusing on corneal wound healing, so working in the eyes. I really liked the combination of the signaling pathways involved in wound healing and wanted to use spatial transcriptomics as a readout to see exactly what was going on in that tissue. One thing we came across was how long it took for even small datasets and how often they failed because of that.



John: Thank you – it's always interesting to hear about a study or project that truly captures the attention of scientists. So, we've observed a variety of capabilities across different spatial platforms currently available. In your view, what are some of the unique values that Biocartography brings to the spatial field?

Zachary: There are two main limitations in my opinion to the existing solutions and that is the time it takes to collect data and the need for advanced bioinformatics to interact with it. So right now there is a pretty high barrier to entry especially for those coming from pure biology spaces who may not have the computational background needed to interact with the data.  By imaging cells instead of imaging molecules we are able to image up to 10X faster right now addressing the first limitation. We also plan to offer complete sample to solution services where you send us samples and we provide a user-friendly interface for interacting with your own data. This should reduce that technical barrier and help make the technology more accessible. 



John: It's exciting to hear about the solution you're offering in response to these challenges. I'd love to learn more about where Biocartography stands today in terms of its plans and product offerings. Could you please share some insights?

Zachary: Right now we were excited to partner with early adopters, as long as they're okay with a little bit of a longer lead time while we get up and running. By focusing on a handful of key partnerships we hope to figure out exactly what products and services are in demand. There has been so much effort towards mapping organs across multiple model organisms at single cell level, and so one thing we anticipate offering is atlas services. Given the scale at which we can produce data we want to make it a reality that these atlases can be done in replicates and across disease states. For that reason the first ATLAS dataset is over 40 million cells mapping the effects of genetic variation on mouse brain anatomy at single cell resolution. This is currently in review at Nature but the data can be accessed via our preprint. On top of atlases we are really excited about expanding into the drug discovery space to see how our scale can be applied to those problems. On top of services, we may be offering kits that way researchers can use their own microscopes and their own wet lab facilities in order to do some of those stains. These will most likely be geared towards niche cell types or disease states that currently don't have antibody stains available.



John: Could you please tell us about your automated processes and how they will impact both throughput and data quality?

Zachary: Right now, most spatial transcriptomics platforms still require a significant amount of hands-on work in order to process those samples. They might provide you with a kit, but you have to be the one to physically add the reagents and do the steps. This can add a significant amount of variability in the quality of data that is produced. What we've done is we've created liquid handlers that are able to replace almost every single one of those steps. So, in our service, we essentially take your slides, put them in a cassette, pop them into a sample prep machine that will eliminate the need for potentially inconsistent hands-on time, this allows us to parallelize to many more samples per technician. Our microscopes are built in a similar way, which allows about a million cells a day to be collected per system. By turning to automation, we're expecting a lot more reproducibility, as well as increasing the scale that we can do at a time.



John: It's great to see automation being integrated into platforms like this. Is there something unique about your sample prep process that sets Biocartography apart and adds distinctive value?

Zachary: Absolutely, as of right now, our main strategy for sample prep is pretty different from the way most people are approaching spatial transcriptomics. Instead of directly hybridizing our probes to the RNAs in the sample. We first anchor those RNAs into a hydro-gel which allows us to clear away the proteins and lipids. This will essentially harmonize all the samples together. Say you had a more difficult tissue that maybe had a stubborn protein that would have previously caused a lot of non-specific binding or prevented specific binding to your RNAs. We can get rid of that before we actually stain, which creates a much more consistent and uniform stain with lower background. On top of that the biggest innovation for us is that our stains are focused on the cell level instead of the transcript level which allows us to design really robust signals that are extremely fast to read. 



John: We noticed the comparison on your website regarding sample areas with MERSCOPE, Xenium, and CosMx. Could you elaborate on the advantages you’re offering through this comparison?

Zachary: Our current ATLAS system has the capacity for 6 slides that are each about 3x larger than competitors. This is enough for at least 3 million cells at a time. Users are welcome to use this area however they like. One way that's really convenient, is to put more conditions on each slide which reduces the cost because you need less slides and also reduces batch effects even further. Exactly the same reason why we turned to automation.  This is really only possible because we've switched from a closed chamber fluidic system to an open chamber fluidic system where we don't have the limitations of laminar flow to consider, which also allows us to scale to more slides per microscope at a time. So we're not limited to two or four. We're currently at six, but there's no reason we couldn't do many more cover slips or slides per microscope or sample prep system? 



John: Where is early data currently being generated with Biocartography technology at this stage?

Zachary: Currently, all of our initial preliminary data has been done in partnership with UCLA, with Roy Wollman's lab. We have a paper in review at Nature and available on biorxiv  that is the largest single cell data set to date, at over 40 million cells across 15 animals. We are actively still collecting data, so we're initiating projects focused on Alzheimer's disorder as well as addiction studies in mouse brains. This includes coupling our ATLAS stains with protein and single gene readouts. There's still a lot of active development in that partnership with UCLA to produce that data. One of the most exciting partnerships that we are working on is the ability to scale even further to thick 3D samples on an ultra fast light sheet system. We're actively looking for partners to expand that pilot data, especially once we set up our service facilities. We are also in the process of developing our user interface that way anyone could play with our early data even without coding experience. 



John: One of my favorite topics is the advice that new founders have to offer. As a new founder yourself, what insights would you share with someone interested in launching their own startup, whether in spatial biology or another field?

Zachary: I think the best advice that I would give is to start as soon as you think you want to start. For us, we had the idea of the company and started the patent process over two years ago. We decided to focus entirely on developing the technology. If I could have gone back, I would have split some of that time towards participating in accelerators, getting some early funding and some early traction with industry partners. That way, when you're ready to transition full time into a founder mode, you already have a significant amount of work done to increase your chances of success. 



John: As we wrap up, Zach, for anyone interested in your technology, what’s the best way for them to get in touch with Biocartography?

Zachary: We have the option for anyone to contact us through our website. On top of that, I'm always open to chatting with people that get in touch with us through LinkedIn. We're very much open to learning how our technology is a good solution for them and learn what services and products best fit their needs. The more people we can talk to, the better.



John: Zach Hemminger, thank you so much. It’s been a pleasure having this time to talk with you and learn more about Biocartography. We look forward to seeing what the future holds for you and Biocartography!

Zachary: Great, thanks for having me! Building a spatial bio community is super exciting and we are looking forward to being a part of that!

Contact Links for BioCartography:

• Zachary Hemminger, PhD (LinkedIn)

• Biocartography Website