Whitechapel - Bacterial neural network - as a student project

Bacterial neural network - as a student project

Luke — Tue, 04 Dec 2007 10:58:34 -0800

A team from Toronto are working on how to build a neural network out of bacteria, which collides about five different cool fields and warning signs in one (work visible here).

Even more interesting is the people doing it - this isn't some shadowy research group working on the ultimate biocomputer, this is a team of students who were trying to win the iGEM prize for synthetic biology. They also contributed a bunch of pieces to the BioBlock program - open source, biology, one of the coolest things I've ever heard. The idea is that there should be an open-source library of DNA sequences (following the software model) rather than letting businesses patent the most fundamental tools we have.

The concept of patented basepairs is terrifying - what if someone owns the rights to the genetic 'transistor' that could cure cancers but doesn't want to share? On the other hand, it's a fantastic amount of work to develop these things, and if big business is going to bother there has to be some reward. It's a tricky one.

Bacterial neural network - as a student project

Willow Bl00 — Tue, 04 Dec 2007 15:41:49 -0800

I say we make baconpunk open source.

..Luke, you make my brain hurt. I like it. It needed the stretch. Hopefully will learn enough after my final papers to get back to you on a lot of this.

Bacterial neural network - as a student project

robb — Tue, 04 Dec 2007 16:29:39 -0800

"concept of patented basepairs is terrifying..."
Kebrew McLeod's book talks more about commodifying genetic bits and the current cost of biological intellectual process stifling a lot of current research. and, more importantly, it talks about biological intellectual property in ways i understand.

Bacterial neural network - as a student project

warrenellis — Tue, 04 Dec 2007 19:21:59 -0800

"build a neural network out of bacteria" -- isn't that just a lovely phrase?

Bacterial neural network - as a student project

Admiral Neck — Tue, 04 Dec 2007 19:31:58 -0800

Like a macro-scale version of Greg Bear's Blood Music, and as such is a scary dreadful idea that puts the wind up me.

Here's a BBC interview with genome hoarder Craig Venter. Why? Just cuz.

Bacterial neural network - as a student project

Luke — Tue, 04 Dec 2007 19:50:25 -0800

warren: yep. It was the very instant I realised I'd be able to say that that I knew that I'd be pitching it to my editor - and he liked it too.

Bacterial neural network - as a student project

Cat Vincent — Tue, 04 Dec 2007 23:20:53 -0800

And nobody's done any gags yet about how easy it is for students to find and grow bacteria...

Bacterial neural network - as a student project

nexus — Thu, 06 Dec 2007 17:34:33 -0800

this stuff is amazing! i love biology<3
i agree that the concept of patenting genes is a slippery slope, though i view finding an effective anti-cancer gene as being almost identical to finding an anti-cancer drug, which we already have regulations for.(we know of several anti-cancer genes, we just can't do anything with them yet. see: stem cell research and viral delivery systems). the problem is, the pharmaceutical industry in the States is a little..... fucking evil and corrupt. i will probably sell my soul to one for a short time after college.

Bacterial neural network - as a student project

Harvey — Fri, 07 Dec 2007 09:55:38 -0800

Interestingly - patent law fans! - the new US Patent Reform Act, designed to protect IT companies from infringing patents accidentally (Microsoft had a $1.52billion judgement against them in Feb) has quite a big effect on all this. It's main thing is to strip all the cost of 'prior art' (ie stuff we already know) out of any damages. For Biotech, which depends on much fewer patents (often one or two) for their products, it's a bit of a kick in the nuts. If you've designed a modified protein or a basepair sequence, and it only differs by a few percent from something that's already out there, you'll only get pence (or cents, if you prefer). In the hypothetical situation where x has a cancer-curing sequence - just infinge the patent. You'll still make a wagonload of cash, and the patent owner won't. Until someone changes the law again, obviously.