Max Hodack Transcriptic

So we're... yes it will be. We're from the opposite side. We looked at this from the perspective.. let me spend the first few minutes at how we arrived at this problem. But. So we. There was me and the four friends. We were doing research and looked at the process of science: ideation, design, experimental execution and interpretation.

Ideation is taking everything you know about your science, and coming up with new connections about the data. It's an art to take these hypotheses and test these, and nothing else, then you complete the circle and do it again. Ideation is highly creative. As is the interpretation and the design, and on the other hand, doing the actual bench wetlab work is not very creative at all. It takes up the vast majority of the time for wetlab biologists. You really measure the career of grad students in units of 1000s of wells pipetted. We laugh nervously because it's true.

That was really kind of what, questions for you outsourcing. What can you do about 50% of the time pretending to be a robot? Outsourcing is not new to biology. There's about 1100 about CROs, and $5-$7B spent on preclinical CROs every year and rapidly growing. There's many in China. There are some in the US that are shrinking. They are growing rapidly. The deals they sign are $100k-$3 million.. the relationship is mitigated by many meetings over six months. This wasn't applicable to the day to day problems of people pretending to be robots.

So we looked a closer look. What about a highly packaged product like molecular cloning? We tried to order a couple. Let me tell you a story. The first day we sent an email with the backbone. Two days later they sent back the email about why it wno't work. Eventually it gets submitted to a lab tech, two days later the lab tech emails the sales rep, then they say they need the billing info over email. Great. So another week later it's submitted, and then they blow through their 1 week turnaround, and they send back the results. It costs about 6x-10x the underlying cost of biology.

It's really frustrating to get an order submitted at all. The cost is totally divorced from reality. The pharmaceutical companies, if you look down the mass market problem, you have 1000s of grad students and postdocs who just spend their entire day pipetting stuff for hours. This problem is unsuited to attack that.

When you price out the costs, look at how much it costs to do molecular cloning. The cost is like $300-$400, it's how much you're inserting and stuff. But the CROs charge you at least 8x or 10x that. So that cost comes in for the sales reps, the techs, it's never automated in China, that's as a rule- no automation. They are a blackbox, they take responsibility for a moment, then they go away and come back when they have a conclusion. So what's going on here?

You'll an email them a sequence, and you say you want this restriction site, and they will say yeah it cuts this insertion site.. well why not search for a place, why did this take a week? So we built a platform that also gives us custom interaction that looks like this. You do the checks on a website. THe order forms for CROs submits to email. You can do these checks on the site. You check for secnodary structures. You focus on the logistics and operations, you heavily automate it, and you treat your schedulers as a core piece of technology. We subcontract the commoditized parts, but it's really about the operations.

You think about this differently because you're no longer a counterpart to a pharmaceutical company. We become an extension of a ustomer's lab, when the grad sstudent thinks it will take 2 days at 4 hours each day, or I can configure it and be done 3 days later.. when you bring the cost down 10x, you increase the orders 20x. The business of existing CROs is fundamentally unsuited. We've ordered from many, it's hard to work with them.. the cost and turn around time, you're just dancing around in the ?? shoes.

So we built a platform called Transcripted, and we started with molecular cloning, we wanted to make it as fast and cheap as possible. You can go online, order your materials, you can mail it in or synthesize it in or the genes, or you can configure the plasmid and inserts, and then you can get back to your research. You can follow it all the way throug hthe process. As .. it's immediately posted, you can ollow it, and if there's a problem you can get an email immediately. Right now it's just molecular cloning, but what about assays whereyou need a microscope image, and we can do a microscope controller and viewer in the webpage. It's very different from traditional CROs.

We see science as four steps. Our goal as a company is to look at this critically and do everything we can to shrink this iteration cycle. Our goal is to make this culture of biotech be much more.. ?? it's much cheaper and much faster and contained in a computer. We started with blah but it's ?? . I haven't discussed this with laypeople.. we started taking orders last week, and I'd be happy to answer questions.

Do you know what your error rates or problems that occur with each of the elements of business that you perform? We picked molecular cloning because you can just sequence it and verify. Since it's so heavily automated, we can formally do it, but more complicated experiments are less easy to verify. But we can guarantee that it is run exactly the same way each time. This is a strength over CROs because they have highly variable results, and after a while their reproducibility or trustability has a very powerful attribute, it's worth probably 100x to certain audiences than any member of statistical results coming from the CROs in general. I think that's something that your presentation doesn't put across strongly, but ti turns otu to be compelling to yuor marketplace. I suggest you talk to them about pricing based on that component. Some way of ensuring that the next sample they try will be like the previous sample they did, because that builds a powerful loyalty that is ver yhard with automated businesses to establish. Once you got it, look at ATMs these days.. people trust them. When they first came on, people didn't trust them at all. Depending on the instrument and the make and model, you can get different characteristics of the data. This isn't a tecan robot or something or agile thing, or agilent thing, we will tell you the name of the platform we will use that one, but if you don't we schedule one automatically. So it will always run on the same class of equipment to try to preserve the aspects of your data.

We're a service provider, we have wetlab space and we run it ourselves. One limitation now is what you can actually execute.. we're trying to grow as fast as possible, if we get 1000 orders for plasmids, we do have operations restrictions, we'd like to move up and do other services, but if you're going to do this, you have to pick your battles one at a time.