Here's our "source code"

Nov 10, 2015

830

Here's our "source code"We make a big deal about being "Open Source". What does that actually mean? It's not as if we have a bunch of code that we can show you...
Or is it?
To demonstrate how serious we are about openness, we figured we'd show you *exactly* what protein sequence we are trying to make, and what all the bits and pieces do. That's about as close as we can get to letting you pick through our source code. 
We don't expect that a lot of you will read through this - just like there's only a select few people who will bother reading through the source code for an open source software package. But if any of you care to try do this yourselves, or want to dig through it for any bugs - here you go!
To really understand what is going here, it would be useful to have a good grasp of the structure of insulin, how insulin is being made in the body, and how insulin currently is produced commercially. So I strongly recommend you first read through Isaac Yonemoto's "Insulin is hard, but not impossible" article on Indyscy.org.
Go on -- we'll wait...
Ok - got all that? Now, the method we are currently pursuing to make insulin is what Isaac originally called a "semi-chemical Nordisk strategy" - because we're producing both insulin peptides from a single artificial proinsulin protein like Nordisk, and "semi-chemical" because we are actually planning to use some clever bit of Palladium chemistry for one of the cleavage events.
This is the protein sequence we're starting from, with all the interesting bits nicely highlighted in different colors for you:
                
(Note the version number v0.2. Version 0.1 was an "oops, forgot something" version...)
                
So there you have it. In practice, there will still be plenty of details to be worked out, to make sure we can get the best yield of correctly folded protein etc. But that's all still work in progress. And as soon as we figure those parts out, we'll tell you about those as well...
Bug reports welcome!

Denny, Maureen, and 12 others like this.

6 comments

Join the conversation!Sign In

Brad Prod

Awesome!!!

Jan 12, 20190

Keith GuerinBacker

Brilliant! As a software developer, I'm interested to know where you store & share your "code". Do you use GitHub or some genetic equivalent? I'd like to naively believe that you just a have versioned insulin.dna file in git that's filled with g,c,a,t's. :) Knowing little about bio chemistry, I'm not sure what the characters above individually represent, but would love to learn more. I'm just starting to learn about this space, as my organization at work has created a Cloud Genomics service https://cloud.google.com/genomics/

Jan 03, 20160

Maureen MuldavinResearcher

Awesome! The characters above represent the amino acids that make up our protein. That picture is in fact our code and could be used to order a plasmid identical to ours. There are a number of different ways that each amino acid can be coded for in DNA, so we do have a file somewhere with the bases that make up our construct. Optimizing the code to best fit the organism (e-coli in this case) and the protein to be expressed is called codon optimization. Here is a brief lesson: http://www.dnaftb.org/22/index.html Counter Culture Labs is teaching beginning bio-engineering classes in the next few months. If you live in the bay area, we would love to have you come. http://www.meetup.com/Counter-Culture-Labs/ I believe that we just e-mailed a text file to the company that created our plasmid. As far as online genetic databases, GenBank allows you to search known genes. It contains data from the US, Europe and Japan. http://www.ncbi.nlm.nih.gov/genbank/

Jan 03, 20160

Maureen MuldavinResearcher

Here is a chart that shows the standard abbreviations: http://www.genome.jp/kegg/catalog/codes1.html

Jan 03, 20160

Sasker

Nice idea, but I see a major problem. If you manage to prduce insulin (and it's biologically active), you will probably want to use it (inject it in humans). Then you are producing a drug/medicine, and this is tightly regulated: you don't want allergic reactions of the patients or worse, inject a product contaminated with baceria or fungi. You may shortcut this process by using the same DNA construct and purification and sterilisation procedure as currently used by pharmaceutical companies (check their patent applications for this). Otherwise, you may face massive costs for clinical trials onec you have your insulin and you need to prove that it's save to use.

Nov 22, 20150

Maureen MuldavinResearcher

We are creating a protocol for the creation of insulin. Once we have created the drug we will confirm identity and purity through laboratory testing. Our lab is not set up to make drugs for human use. After the scientific portion of the project is done, a generic company will need to take over the approval and manufacture of the drug. True homebrew insulin is a different problem then this project.

Nov 22, 20150

Tyler Alexander

I see you're cloning the gene for wild-type insulin. Why not clone one of the rapid acting analogs? The patent for Humalog (insulin lispro) has expired. I also see you're planing to use IMAC purification. Are you planning to use another purification step or are just going to wait and see how clean the IMAC gets it? I'm super excited for this. I'm type one and I'm pursuing my Ph.D. in biochemistry. I've always complained about the cost of insulin and would make jokes that I'm just going to grow my own. Now someday I might! Great work!

Nov 21, 20150

Maureen MuldavinResearcher

Thanks so much for your support! You should check out our google group. You might really enjoy the scientific discussion. According to Patrik: "The reason why we're doing wild-type insulin is that we wanted to keep things as generic as possible. Many of the existing variants like lispro only change an amino acid or two, so if we can make wild type insulin, the variants are typically just a matter of changing the sequence a little bit. (In lispro, the lysine and proline amino acids at the end of the B chain are reversed, which blocks the formation of the hexameric storage form of insulin. As a result, lispro insulin is far more fast-acting than wild type insulin.)" We are still debating the best way to purify our insulin, but we most likely will do Reverse Phase(RP) HPLC after IMAC.

Nov 23, 20151

Tyler Alexander

Thank you for your answer, Maureen. I can understand the reasoning for choosing to do wild-type, however rapid acting insulin analogs have been shown to have absorption rates that better match how the pancreas would release insulin during a meal time bolus. You're correct in that rapid acting analogs only differ by a point mutation or two from wild-type which prevents hexamerization. You may find that monomeric insulin may require a completely different purification protocol than hexameric insulin (especially if you end up having to use size exclusion chromatography, and perhaps to a lesser extent in RP-HPLC). Just something to consider. :)

Nov 24, 20150

Maureen MuldavinResearcher

I think we are pretty set on wild-type insulin at this point, but we have really enjoyed talking about your insightful questions and comments. In fact, we are hoping to recruit you! https://groups.google.com/forum/#!forum/ccl-insulin If you have some free time, we would love for you to participate in our planning and research!

Nov 30, 20150

Valentino MegaleBacker

Congrats guys, really exciting and inspiring project!

Nov 12, 20150

Maureen MuldavinResearcher

Thank you! Really appreciate the support!

Nov 13, 20150

Andrew GrayBacker

this is awesome. Can't wait to have a crack at this ourselves! Really hope this makes it to the funding goal great work guys.

Nov 10, 20150

Maureen MuldavinResearcher

Thanks so much! Once we get the protocol working and online we'd love for you guys to use it to make insulin!! #openscience

Nov 24, 20150

About This Project

Open Insulin

Counter Culture Labs

A team of biohackers is developing the first open source protocol to produce insulin simply and economically. Our work may serve as a basis for generic production of this life-saving drug and provide a firmer foundation for continued research into improved versions of insulin.