Share this project

Done

Share this project

Done
A scientific quest to understand beer at molecular level: Send us a beer ▸▸ we study its DNA ▸▸ we trace a tree of beers ▸▸ you discover new beers.
A scientific quest to understand beer at molecular level: Send us a beer ▸▸ we study its DNA ▸▸ we trace a tree of beers ▸▸ you discover new beers.
124 backers pledged €10,773 to help bring this project to life.

Go find your beer on the tree ;-)

Posted by BeerDeCoded (Creator)
4 likes

Dear all,

Last weekend, we brought all your beers at the Open Food Hackdays. We analysed their molecular profile. We traced a molecular tree with all your beers and created a working prototype of the web-app.

It is live. You can try it. It works on google chrome. Go find your beer on the tree.

Luc voted for Monty Python's Holy Grail. It's here!
Luc voted for Monty Python's Holy Grail. It's here!

On May 27th 2015 we asked your help to support this project. Our goals were:


  1. to expand public understanding of molecular technologies and their connection to culture.

  2. to help non-scientists to study and analyse their beer samples at molecular level.

  3. to motivate citizen scientists to generate an open molecular dataset.


Did we meet our goals?

 

Goal 1. BeerDeCoded has been an excellent excuse to talk about molecular testing. We count 50+ media mentions. Thousands of social media interactions (2,000+ during the kickstarter the campaign). Hundreds of in person conversations (we breakdown the topics of the first 107 interviews). Mission accomplished.

 

Goal 2. With your support, we have been able to take part in hackathons/workshops around the world: Lausanne, EXPO15 Milan, New York, San Francisco, Paris, Basel, Ljubljana, Solothurn.  Some backers took a NY-Geneva flight to visit our bench. 

Pipette in hands, we enjoyed co-working with you all. Mission accomplished.
Pipette in hands, we enjoyed co-working with you all. Mission accomplished.
 

Goal 3. Cool but where is the data?

In the last update, we sent the DNA samples to the genomic facility. Some weeks ago, we received a QC report.

The quality control is a bit frustrating: on 96 beer DNA samples, only 39 beers (40%) have quality DNA that can be sequenced.

These results can have different interpretation: It is possible that some brewing processes degrade DNA more than others. It is also very possible that our "Do-It-Yourself" lab and our "Do-It-Together" operations were not properly controlled for the best quality standards. For instance, did all of us "biohackers" washed their hands, wear gloves, washed thoroughly the spin columns?

Meh.

We decided to sequence the 39 "good" beers. The university we are collaborating is taking its time: the results are less of a priority, especially after September, when a renowned group of scientist published the full genomes of 150+ yeast strains used for brewing. It's an exciting study, go to read it. Anyway, we look forward to receive our results as well. We will keep you posted.


We also feel a moral obligation to give back all the molecular results of your beers, also the ones in which DNA seems not present. How to do that?

An Hackathon, what else?

Last weekend we borrowed from the Criminology department a molecular scanner.

Technically, it is called Raman spectrometer. We excite the beer with a strong laser. Inside the beer, each molecule changes its rotational and vibrational status. When the laser is turned off, the molecules relax back. The scanner measures the difference in the rotational and vibrational dance and creates a molecular fingerprint.

The policemen compare these fingerprints with other fingerprints from a library of known illegal substances (drugs, explosives) and we are happy to share that your beers are clear.

Yes, we scanned 118 beers including all the ones that you have sent or voted. The documentation is available on the Open Food Hackdays page.

For some beers, the spectra are compatible with the presence of alcohol, preservatives and - at least in one - an herbicide. The molecular fingerprint contains much more information and due to a sort of molecular crowding, many parts of the fingerprints are difficult to identify.

The good news: to compare two fingerprints you do not need to identify them first. The great news: everybody now can help us with the analysis. Overall, it's open data sitting on a github repository and here is a first web-app prototype.

On Chrome, click on the Clustering button to show the tree of beers and search your beer. MISSION ACCOMPLISHED!!!
On Chrome, click on the Clustering button to show the tree of beers and search your beer. MISSION ACCOMPLISHED!!!

It has been a privilege to work these 20 months with your support. It is now time to analyse together the data. Test and Taste: sign-up on our mailing list if you want to stay engaged with our future experiments. So long, and thank you for all the beers!

With our fullest gratitude,

Gianpaolo, Jonathan & BeerDeCoded team


Special thanks to Luc Patiny (EPFL) for the invaluable help with the molecular spectra.

PS
Premium backers: expect a personalised message in the upcoming weeks. We will start soon looking at your beers in more details. It would be great if you could join us in person for the analysis. If you are around Lausanne, our lab at Hackuarium.ch is open every wednesday night. 

Comments

Only backers can post comments. Log In