4 ms·

SynBio in the 2020s smells a bit like green tech in the 2000s. I see a ton of founders and funders with zero biology background thinking that they have the visi

by tbenst 6y ago

SynBio in the 2020s smells a bit like green tech in the 2000s. I see a ton of founders and funders with zero biology background thinking that they have the vision and can just hire biology expertise. Biology is ruthlessly unforgiving and complicated, and often times very low efficiency. For example, it’s a bit amusing to see the ratio of CRISPR mentions in startups vs the techniques actually being used by scientists. You would never use CRISPR if you had the option of using Cre. I’ve made several stable transgenic animal lines using Tol2 transgenisis, which is much higher efficiency, too. If you’re excited by the space, but don’t have a biology background, consider joining an organization led by a biologist.

alextheparrot 6y ago
This was my experience working in SynBio -- the iteration cycles are so much longer, the failures so much more complex, that someone who comes from a software-first background has a hard time because it just doesn't scale as nicely. If we look at the space, I think it is clear that biologist first companies have been the clear winners up to this point -- Gingko Bioworks being one clear example
ros86 6y ago
> I see a ton of founders and funders with zero biology background thinking that they have the vision and can just hire biology expertise. Genuinely curious about your data here. The two first companies mentioned in the article (Zymergen and Gingko) have founders with biology background. Many other companies I know as well. > You would never use CRISPR if you had the option of using Cre. I’ve made several stable transgenic animal lines using Tol2 transgenisis, which is much higher efficiency, too. That probably depends on the organism you're working with. E.g. with baker's yeast, my preference would definitely be CRISPR/Cas9 instead of cre recombinase (see e.g. https://academic.oup.com/femsyr/article/15/2/fou004/534426 https://academic.oup.com/femsyr/article/15/2/fou004/534426 ).
- tbenst 6y ago
  > Genuinely curious about your data here. I invest a little in the space so have seen some pitches at IndieBio, YCombinator, and talked with a bunch of companies. There are indeed some really great founders with biology background as you point out. But have been seeing more and more founders from software background. The enthusiasm is great, and people with software backgrounds have a ton to contribute! Just maybe shouldn’t be the CEO. > baker's yeast, my preference would definitely be CRISPR/Cas9 I’ve never worked with yeast but was under the impression that you could just give them plasmids and call it a day?
vikramkr 6y ago
You've also got a huge number of DIY biohackers with no business background or vision for a commercial enterprise hacking away in diy hackerspace around the country. I think those hackers will evolve into the truly successful next generation of synthetic bio companies, like what happened to tech companies.
- tbenst 6y ago
  Or, they’ll follow the path of entrepreneurship from makerspaces in 3D printing and found 2000 nearly identical companies all making an FDM printer. Synthetic biology is fundamentally different from software and much harder to build a business than even hardware. That said, the defensible moat is even stronger and there are indeed huge opportunities!
  - vikramkr 6y ago
    I'm well aware of how much harder it is in this space haha, I've spent the past couple years building a company in this space and have run into to all the roadblocks. The advantage of the environment now is that large numbers of people can spend years hacking away at some bio experiment or problem they're interested in for an amount of money that's not out of reach for the upper middle class. There's already the equivalent of a flood of 3d printer companies - so many people working on DIY thermocyclers/gelboxes etc, there's a running joke that a lot if people spend more time making tools than using them to actually do biology, but with time you might see people using those tools to build interesting things. It's not as easy as doing electronics in your garage in the 1970s, but the inflation adjusted price point is probably equivalent (for the equiptment that is - the house the garage is attached to probably costs quite a bit more...)

by tbenst 6y ago

alextheparrot 6y ago

This was my experience working in SynBio -- the iteration cycles are so much longer, the failures so much more complex, that someone who comes from a software-first background has a hard time because it just doesn't scale as nicely. If we look at the space, I think it is clear that biologist first companies have been the clear winners up to this point -- Gingko Bioworks being one clear example

ros86 6y ago

> I see a ton of founders and funders with zero biology background thinking that they have the vision and can just hire biology expertise. Genuinely curious about your data here. The two first companies mentioned in the article (Zymergen and Gingko) have founders with biology background. Many other companies I know as well. > You would never use CRISPR if you had the option of using Cre. I’ve made several stable transgenic animal lines using Tol2 transgenisis, which is much higher efficiency, too. That probably depends on the organism you're working with. E.g. with baker's yeast, my preference would definitely be CRISPR/Cas9 instead of cre recombinase (see e.g. https://academic.oup.com/femsyr/article/15/2/fou004/534426 https://academic.oup.com/femsyr/article/15/2/fou004/534426 ).

tbenst 6y ago
> Genuinely curious about your data here. I invest a little in the space so have seen some pitches at IndieBio, YCombinator, and talked with a bunch of companies. There are indeed some really great founders with biology background as you point out. But have been seeing more and more founders from software background. The enthusiasm is great, and people with software backgrounds have a ton to contribute! Just maybe shouldn’t be the CEO. > baker's yeast, my preference would definitely be CRISPR/Cas9 I’ve never worked with yeast but was under the impression that you could just give them plasmids and call it a day?

vikramkr 6y ago

You've also got a huge number of DIY biohackers with no business background or vision for a commercial enterprise hacking away in diy hackerspace around the country. I think those hackers will evolve into the truly successful next generation of synthetic bio companies, like what happened to tech companies.

tbenst 6y ago
Or, they’ll follow the path of entrepreneurship from makerspaces in 3D printing and found 2000 nearly identical companies all making an FDM printer. Synthetic biology is fundamentally different from software and much harder to build a business than even hardware. That said, the defensible moat is even stronger and there are indeed huge opportunities!
- vikramkr 6y ago
  I'm well aware of how much harder it is in this space haha, I've spent the past couple years building a company in this space and have run into to all the roadblocks. The advantage of the environment now is that large numbers of people can spend years hacking away at some bio experiment or problem they're interested in for an amount of money that's not out of reach for the upper middle class. There's already the equivalent of a flood of 3d printer companies - so many people working on DIY thermocyclers/gelboxes etc, there's a running joke that a lot if people spend more time making tools than using them to actually do biology, but with time you might see people using those tools to build interesting things. It's not as easy as doing electronics in your garage in the 1970s, but the inflation adjusted price point is probably equivalent (for the equiptment that is - the house the garage is attached to probably costs quite a bit more...)