Company Archetypes ✨ w/ Vik Bajaj - Managing Director @ Foresite Capital | BIOS

0:00 absolutely we have started several companies i think three of which are publicly announced a fourth is about to be announced tomorrow and i can tell you about it these companies are really in three areas so perhaps i can go over the three areas and just give you an example in each one is precision medicine the idea that we can do in oncolo what we've done in oncology for a broader set of diseases that actually many many many more people

0:29 suffer from so in oncology precision medicine is comparably easier not to say that it was easy just easier because the genetic rearrangements which are responsible for tumor genesis or any of the hallmarks of cancer the so-called drivers their role as causal drivers of disease has been unambiguously established now through two decades of genomic and other studies also those same genetic drivers are the targets for the drugs

1:00 and those same genetic drivers are the stratifiers that we use in clinical trials and eventually through companion diagnostics the stratifiers that we use to choose patients who would respond to those drugs so it's a very linear set of reasoning that has resulted in short impactful trials and drugs that reach benefit patients who will actually benefit from them and that's been quite a transformation so to do that in precision medicine outside of oncology

1:31 is harder because we are dealing with rare genetic variation that rare genetic variation is used to nominate targets which have a causal role in disease but the variation is so rare that that is not the way in which we stratify patients into responding and non-responding families other methods are required to do that those methods depend on the kind of large data sets that forsyth labs works with so if you think about the areas in which this is important we've created several

2:02 precision medicine companies but autoimmune disease is a prime example there's a lot of information available about the genetic architecture of autoimmune disease there's also a lot of biological information available about responders and non-responders that has not been quite put together yet but we have the ability to generate large data sets in autoimmune disease as we have an oncology that would resolve the heterogeneity so that kind of analysis internally led us to create a company now called

2:34 illumis which is led by martin babler former ceo of principia and its goal is to be really the world's first autoimmune disease precision autoimmune disease company it started with a detailed analysis of pleiotropic genetic nodes that lie at the heart of multiple autoimmune diseases but for each of these the genetic architecture points to new indications new biomarker-driven approaches and new stratifiers so that company which recently closed a

3:07 large round with matrix and other partners is actually now in the clinical stage with a tick to asset that is an example but not the only one of this kind of family obviously that was a well-known target before we started it it also has a data generation program through its observational studies and clinical trials which will really seek to define the biological architecture of autoimmune disease at the right scale and with the right measurement tools to really teach us about the biological

3:40 response and the patient experience and lay the foundations of a precision medicine strategy in that space so as you can imagine autoimmune disease is not the only thing that fits that paradigm there are others and you know we have unannounced companies that you'll hear about in time that are pursuing them one example you'll hear about tomorrow is a company that is called 1016 which is going after a whole new area of biology that has to do with

4:11 somatic mosaicity or how the genome of progenitor cells changes over time and causes the emergence of genetic families or clonality and what are the implications of that it's not just that it leads to cancer that's well known but also a whole host of inflammatory associated conditions and other conditions including in the cardiovascular disease space that are very important unmet medical needs

4:43 and so this company will seek to unlock that biology of somatic mosaicity and clonal hematopoiesis in particular to really understand it through large data generation exercises but also to fundamentally intercept it in a new way and in doing so it really ushers in for us a important precision medicine concept which is that if you can understand the earliest stages of disease you can actually intervene early before the harm is done that doesn't preclude you from

5:14 using the same mechanisms to intervene later but the compelling and tantalizing possibility is to actually intervene early in the course of a disease so that's another precision medicine company that we've created sorry for the long explanation i'm very passionate about these ideas and the teams involved that company is led by mark chao who was the founder of 47. the other area which you can imagine that we're quite interested in are broader platform

5:44 companies that do not really operate in a single therapeutic area a really great example of that is a company called interline which is led by zac sweeney who was the cso of denali interline was funded by our colleagues and friends at arch after we launched it from forsyth labs along with foresight capital and interline seeks to create really the first systematic toolkit that will understand and modulate dysfunctional protein communities with

6:15 the right level of finesse so it's not producing cumbersome bidentate molecules like protacs but small molecule drugs that are capable of really changing the interactions among members of a protein complex to achieve a therapeutic effect and that platform company has multiple components but they include as you might imagine a combination of genetics and proteomics to understand from primary human evidence what are the lesions at the protein protein interface which

6:46 are causal drivers of disease and that's paired with a computational chemistry and machine learning platform that is allowing us to access in silico elements of protein protein associations and their interactions with small molecule in a way that was methodologically and just in terms of computational capacity impossible even five years ago and that's also being combined with you know a variety of new cellular phenotyping and other assays to create a really robust target discovery

7:18 platform across multiple indications another platform company that we've created is very interesting it's called sistina it's in a little bit of a different area an area of synthetic biology this company is led by bill colston who is the grandfather of the single cell revolution he developed in his academic group in the national lab systems a lot of approaches to droplet encapsulation of cells and microfluidic

7:48 handling of reactions in droplets that led to his formation of quanta life the lineage of quantalife those people involved also went on to found 10x and in scripta and sistina as a synthetic biology company starts with the following realization we are now in the perhaps third generation second generation of synthetic biology efforts and they've been very successful but they fall far short of the promise

8:19 of really engineering new biology or creating generic approaches to manufacturing substances through bioreactors where otherwise there would be environmentally unfriendly expensive inaccessible either through chemical synthesis or natural product isolation and the reason why that is happening is because even in the second generation of these companies that have room scale robotic systems to automate their experiments

8:50 the combinatorial dimensionality of those experiments is just not commensurate with the underlying biological complexity those experiments probe sometimes less than one percent of the genetic variation that gives rise to a phenotype so we can expect that in combination with directed evolution and other approaches to be transformational but narrowly so so we asked the question how could you increase that by orders and orders of magnitude and the answer

9:20 of course is to harness all the tools of since single cell biology to perform edits and read out phenotypes at the single cell level that is the atomistic unit of biology and it allows systena to do experiments whose dimensionality is 10 000 to a million times higher than any experiments that have been done before and that means that the space of real biological complexity can be explored efficiently it's also then as you might

9:51 imagine because it's a foresight labs company those data sets are a basis real machine learning experiments that are only possible when you have large-scale data sets whose biases are relatively weak and characterized but who are expansive with respect to the biology that you're trying to perturb we've seen this time and time again in industries or subfields that have been transformed by machine learning including natural language processing

10:23 including things like image analysis that it's only when data sets of sufficiently low bias and sufficiently large scale emerge machine learning approaches allow you to navigate that space much more efficiently whether the task is classification or to generate in you a sell state in this case so that's the basis of systena to combine all of those things together to really create a practical route to manufacturable synthetic biology products that is much

10:55 less limited than what's been there before so those are two examples of platform companies a third area which i won't talk about much that we're starting to explore is just the consequences of the philosophies that i'm talking to you about in healthcare services and in healthcare services often the datasets that we would use to deliver insights yes there's a role for genetics yes there's a role for complex molecular biology even in the individual to stratify to

11:26 predict risks but there's a great deal of low hanging fruit that comes from the pervasive availability of clinical records and clinical information for the first time so our goal is to really launch a few companies that exploit those concepts that also are built upon the regulatory changes and market changes that have come in virtual care in the last two years due to covet to create healthcare service delivery models that are much more

11:57 efficient from a systems perspective and also much more responsive to the needs of the consumer