How Open Insulin works to open-source science and medicine

Open Insulin is one of many projects at Counter Culture Labs, a biohacker space in Oakland, CA. Counter Culture Labs was founded around 2011-2012 by a group of hackers with diverse backgrounds and interests. Members joined us from Sudo Room, another hacker space in Oakland, and Biocurious, a biohacker space in Sunnyvale. Many were also involved in Occupy Oakland, and wanted to establish a more permanent organization with the same community spirit and values. Eventually, the community we built in the hacker space reached a critical mass of knowledge and interest around the idea of starting to producing insulin with a manual protocol, but one designed to be simpler and less expensive than existing methods. We named the project “Open Insulin” to reflect a commitment to make the results freely available to any interested party and publish our methods openly. The name was a deliberate reference to open source software.

Open source, as many readers may know from the software world, is the practice of making all information necessary to produce and modify a product publicly available along with the product itself. It started in software as an alternative to the practice of providing only machine-readable copies of programs, which can’t be understood or practically changed by users of the software. Without access to human-readable code, users and other stakeholders were shut out of their own tools. Open source methods of production are relevant not just to aligning incentives and improving the economics of software development, but also to scientific reproducibility and transparency, and in both software and science, open source can enable more participation and progress than trying to hold secrets close. In medicine in general, and diabetes treatments in particular, I think it holds one of the keys to breaking through the barrier between promising research and a stagnant market of treatments available to patients, just as it made software much more efficient to produce and use and enabled a great deal more innovation than was otherwise possible.

Open source software and the hacker culture that makes it was a major inspiration for biohackers to organize as such. “Biohacker” derives from the term “hacker” in the sense used in the communities of early pioneers of computing, where it’s a term used to refer to people who seek to understand how things work on the inside, instead of just using products as a consumer. Hackers are people who seek to modify things to serve their own purposes, instead of just accepting them as being limited to their originally intended purposes. It’s an approach that emphasizes the philosophical concepts of phronesis and techne, which describe an embodied, contextualized, practical approach to things, applied to science and technology. Biohackers are people who take a practical approach to understanding and engineering biological systems, and look beyond appearances and inside the black boxes of commercial products to understand the substance and true implications of things.

If we can achieve the creation of open source insulin, it could contribute to at least three important goals - first, by making insulin production more economical at a smaller scale, and opening up manufacturing to much more competition, it could improve cost and access for patients. Second, we hope the protocol will serve as a basis for future research into improvements to insulin - variants that are longer acting, shorter acting, more temperature stable, and so on - that address different concerns that arise in treatment. Third, we hope it might serve as a basis for research and production of other proteins by small groups, and open up participation in research and development to accelerate progress in other aspects of diabetes treatment besides insulin and other areas of science and medicine besides diabetes treatment.

Currently we’re working on a novel method to produce human insulin, which is not patented, and as far as I know is not patentable. There are variations on normal human insulin to make them longer or shorter acting, which involve very small changes to the sequence that codes for human insulin.

We’ve been working in the lab for about 10 months, and we’re just seeing the first signs of success on our first major milestone, the production of proinsulin, which is the precursor to insulin. Once we’ve verified our success there, we’ll move on to making the final form of insulin.

We face two main challenges. First, while we are a group of talented and curious folks, most of us are learning challenging lab protocols from scratch, and second, we’re working with limited amounts of time and money, fitting the work into gaps in our schedules left by work or school, and mainly relying on surplus equipment and reagents that add delays and uncertainties to our work. So progress can be slow and involve a lot of detours on top of those implied by the already uncertain nature of scientific investigation, and we have to dig deep to figure out what to do next when something goes wrong. We do our best to learn fast, but it’s difficult to follow up on everything we should with our limited time and resources and background knowledge. There’s a lot of practical wisdom around making insulin that doesn’t show up directly in the papers published in scientific journals, and we’re learning these nuances of making things work as we go. Much of the value we hope to provide to the community is documenting as much of this practical wisdom as we can, and perhaps eventually automating the kind of work we’re doing by hand right now.

If all goes well, we hope our techniques will be available to people with diabetes, especially to help meet the needs of people in resource-poor and lower-income countries. However, it’s hard to say at this point when we might be able to deliver a result useful for making a pharmaceutical-grade product. After the scientific work, which might take a few years, there will be the work of seeking business partnerships and going through more engineering work and the biosimilar approval process in the US. In other areas of the world, it may be simpler, but might also involve more unusual local political considerations. We’re hoping that our work will inspire others to move in parallel alongside us and increase the chances that one group will succeed quickly, and there are some preliminary signs that this is happening, which could yield results sooner. But there’s no doubt we’re at the start of a very long and winding road.

We’ve had the good fortune to have many experts in relevant fields reach out to advise us to smooth out our path, but many important questions remain, and the most important ones remain for us to find new answers. Scientists who are reading this – can we count on you to contribute your expertise in protein expression and purification? Organizers – can you share with us your vision for expanding these efforts into a movement with participation from parallel groups like ours around the world?

The production of this article was supported by Op3n Fellowships - an ongoing program for community contributors during May - November 2016.

Question to Organisers

My two cents, @dfko, as someone involved in building a movement (as one seems to be these days, as if there’s no other way [sic]):

Not to be too pessimistic although it’s in the air given recent events in the US, but it seems that no matter how good a scientist or how decent your values are, or how promising your early results: common decency is no longer enough and one has to have grit. In other words, I feel more and more that any ambitious effort to lead to greater equality, access, fairness in the world needs to become somewhat political and build consistent support behind them.

Example: look at Academia as a space for sharing expertise and see whether it is going in the right direction paywalling everything and being very slow to reform in terms of open access policies. What you have is a set of no-way-out reactions, people like A. Swartz or this lady in Kazakhstan taking action in their own hands. I would not favor those of course, but middle ways are long term battles with a lot of coordination across the net. Repeat, repeat, repeat…

Thanks for the analysis Noemi. Grit is about all we’ve got right now! It’s gotten us some results in the lab and gotten many people interested in helping out. And it has been key, in the preceding years, to founding the hacker space itself that we work out of. We’ve gone out on a limb and inspired some people and now we need to make the most of this interest. Influencing politics in favor of the values we’re trying to serve is a side interest of most of the team and we’re hoping in collaborating with Edgeryders here to be able to contribute to these broader questions even as we focus on our scientific and engineering work.

Community support

Hello @dfko! The Open Insulin project inspired my friends and me a lot, which ultimately led to starting ReaGent, a DIYbio space in Ghent, Belgium. We’ve been going for about a year now.

I’m sure there are more people with similar stories and who would love to help you - us included. Let me know if you want to skype sometime to discuss what we can contribute. Have you reached out or collaborated internationally before?

Interested!

I am more a engineer/scientist on the dead matter side of things - but I know just enough to understand that this is a pretty big thing.

If you do skype with @WinniePoncelet I’d love to be a fly on the wall and see if I can contribute in some way.

Do you know the question function of research gate? Insulin comes up with 385 researchers who follow the topic and 300 questions discussed. I often get good and detailed responses on things that you usually have a hard time finding in papers. Check here: https://www.researchgate.net/topic/insulin

Skype call

@dfko and me ended up Skyping on Monday. It was kind of last minute after long silences and trying to find overlap in each of our hectic agendas. Sorry @trythis and @Alberto ! I’ll try to give a short summary. We discussed that as I’m more active on ER, I can serve as a link here.

The project is still going well in technical terms. Only now, due to time and communication constraints, international collaboration seems possible. A local group in Sydney just launched a few weeks ago and already made good progress. Other groups are open to launch a local chapter.

We’ll be launching a chapter in Belgium to see if our community is up for joining this. Like that we will test the international collaboration and see where it leads us, perhaps others will join as well.

There’s a diversity of angles, eg. we could contribute more in terms of communication, while the others are more technically adept. It’s interesting to me that each group can have their own approach, and all contributing to the same goal.

I’m reaching out to the other labs in Belgium to see if they are up for joining. There’s already a good diversity there (arts, technical, educational, …).

Does anyone have any ideas in terms of collaboration process? We’re in a new field to set this up at such a scale.

Collaboration: can we help?

@WinniePoncelet and @dfko , this is really great news! I’m sorry to have missed it, but the important thing is that it worked for you.

I was thinking. We (Edgeryders too, but OpenCare in specific) would be really interested in seeing this collaboration unfold. We would like to help, if we can. If our next batch of fellowships are in the pipeline, we could even support you. @markomanka , what do you think?

I guess the main question is this: what is the main collaboration challenge that you are facing? Is it the launch of a Belgian chapter of Open Insulin, and specifically the uptake of the ReaGhent community of the project? And: who is going to lead on this? Is it going to be you, Winnie?

Also ping @Noemi | @Ezio_Manzini | @Federico_Monaco

Main challenges

I’m also fascinated to see if/how we can make this work. Testing something like this is right up there with the things we absolutely wanted to try out when setting up our lab. Some initial thoughts, I may be off.

In my view, there’s two clear levels to this: local collaboration and international collaboration. Both are interesting to research seperately, and the combination even more so. For good measure, local research groups participating in this are (or should be) inevitably autonomous in practical aspects. Save the occasional shipping of a sample to replicate experiments or other small stuff like that. This entails that the local element will have coordination costs in the form of time, money and energy to keep the community and project going through a potentially long and tough research track.

The big advantage is that all global groups are interconnected for the information aspect. The international element will probably happen relatively smoothly if the right digital infrastructure and practises are in place, as there is a clear incentive to do so: faster learning, collective intelligence. Coordination costs should be low, there’s only information being shared and this is cheap and fast if you have people who keep track of data.

I’m optimistic about the uptake of the project by the community here. The main challenge that I see now for us is the local coordination cost (Ghent or Belgium level TBD). I’ve (tried to) set up, participated in and observed a few similar local research collaborations. They all had high coordination costs. This insight comes from scientific fields that are way less of a pain than biotech/pharma, so I can’t see it turning out easier than expected. Another challenge is gathering relevant insights in terms of process to have some insights at the end of the road. More so when you haven’t set it up right at the beginning: having to change process during the project is costly.

For Open Insulin in SF locally it seems to have turned out well and I already discussed some potential pitfalls with @dfko. This is very hopeful.

As for international collaboration, there’s already some stuff in place: filesharing and weekly Skype calls. With the Sydney group recently started there is already a clear synergy.

I think we should have a call about how we can make it happen, @Alberto and anyone else?

Guys you’re really going for this, wow.

In addition to possible support from OpenCare (for which we are just now drafting terms with Marco) you might want to look into a uni partner that would be willing to assign students to work for it and make it part of a semester assignment, that should really help kickstart.  Of course, this may be too advanced stuff for coursework, I have no idea.

What’s a good timeline? Is it reasonable to expect advances in 6 months? They can be, if not research results, community mobilization results, contributing to global documentation and similar things. Well done!

Yes!

Let’s totally have a call, @WinniePoncelet and @Noemi . Also calling @Ezio_Manzini . Best dates for me are Tuesday and Wednesday 14/15 Feb.

Only next week

I can only do Feb 23, 23 or 25. @WinniePoncelet ?

Good

@Alberto @Noemi Great! All 3 dates work for me:

23 afternoon/evening

24 evening

25 anytime

Finally found it - sorry it took me so long

You even mentioned me! I need to train my eyes better when skimming all those headers. @Matt maybe it would help if the routine messages have a clearly different header e.g.:

  • MENTION [by Winnie] on [XY] Edgeryders

  • Digest of Edgeryders activity [X-Y]

  • New comment (keep as is)

@WinniePoncelet for the local coordination I am a huge fan of audio recordings. Especially if the group is not huge, but maybe not everyone is on the same page, or arrives later - it is great to catch up, or look up things you didn’t get the first time around.

Also if you eventually want to share parts with another group it gets much easier. Either by writing a (late) summary or (if in English) just pushing part of the convo to them. One nice thing is that a written summary posted will probably have a high fraction of the key terms that are relevant for ethnographic analysis.