Open Source: A Question of Evolution

I met Matt Ridley once, when he was at The Economist, and I wrote a piece for him (I didn't repeat the experience because their fees at the time were extraordinarily ungenerous). He was certainly a pleasant chap in person, but I have rather mixed feelings about his work.

His early book "Genome" is brilliant - a clever promenade through our chromosomes, using the DNA and its features as a framework on which to hang various fascinating facts and figures. His latest work, alas, seems to have gone completely off the rails, as this take-down by George Monbiot indicates.

Despite that, Ridley is still capable of some valuable insights. Here's a section from a recent essay in the Wall Street Journal, called "Humans: Why They Triumphed":

the sophistication of the modern world lies not in individual intelligence or imagination. It is a collective enterprise. Nobody—literally nobody—knows how to make the pencil on my desk (as the economist Leonard Read once pointed out), let alone the computer on which I am writing. The knowledge of how to design, mine, fell, extract, synthesize, combine, manufacture and market these things is fragmented among thousands, sometimes millions of heads. Once human progress started, it was no longer limited by the size of human brains. Intelligence became collective and cumulative.

In the modern world, innovation is a collective enterprise that relies on exchange. As Brian Arthur argues in his book "The Nature of Technology," nearly all technologies are combinations of other technologies and new ideas come from swapping things and thoughts.

This is, of course, a perfect description of the open source methodology: re-using and building on what has gone before, combining the collective intelligence of thousands of hackers around the world through a culture of sharing. Ridley's comment is another indication of why anything else just hasn't made the evolutionary jump.

Follow me @glynmoody on Twitter or identi.ca.

IT Lessons from the Thylacine's Genome

The thylacine is a near-mythic animal. A marsupial related to the kangaroo, it was wiped out early in the last century, surviving just long enough for a few specimens to be pickled in jars. As usual, mankind was responsible, hunting the animal to extinction. But not entirely:

Scientists have detailed a significant proportion of the genes found in the extinct Tasmanian "tiger".

The international team extracted the hereditary information from the hair of preserved animal remains held in Swedish and US museums.

The information has allowed scientists to confirm the tiger's evolutionary relationship to other marsupials.

The study, reported in the journal Genome Research, may also give pointers as to why some animals die out.

The two tigers examined had near-identical DNA, suggesting there was very little genetic diversity in the species when it went over the edge.

Although it was hunting that finally drove the Australian animal out of existence, its longevity as a species may already have been fatally compromised, the researchers believe.

So if *you* want to avoid the tragic fate of the thylacine, remember: avoid those Microsoft monocultures, wallow in the genetic diversity of the free software ecosystem.

Really Googling the Genome

When I wrote a piece for the Guardian four years ago called "Googling the Genome", it was more of a metaphor than a specific warning about Google rummaging through your DNA. But it's a metaphor no more:

A Harvard University scientist backed by Google Inc. and OrbiMed Advisors LLC plans to unlock the secrets of common diseases by decoding the DNA of 100,000 people in the world's biggest gene sequencing project.

The *first* 100,000 people, I think they mean....

Genomics Goes Read-Write

One of Larry Lessig's favourite tropes is that we live in a read-write world these days, where creation is just as important as consumption. Well, hitherto, genomics has been pretty much read only: you could sequence the DNA of an organism, but creating entire genomes of complex organisms (such as bacteria) has been too tricky. Now that nice Dr Venter says he's gone and done it:

A team of 17 researchers at the J. Craig Venter Institute (JCVI) has created the largest man-made DNA structure by synthesizing and assembling the 582,970 base pair genome of a bacterium, Mycoplasma genitalium JCVI-1.0. This work, published online today in the journal Science by Dan Gibson, Ph.D., et al, is the second of three key steps toward the team’s goal of creating a fully synthetic organism. In the next step, which is ongoing at the JCVI, the team will attempt to create a living bacterial cell based entirely on the synthetically made genome.

The team achieved this technical feat by chemically making DNA fragments in the lab and developing new methods for the assembly and reproduction of the DNA segments. After several years of work perfecting chemical assembly, the team found they could use homologous recombination (a process that cells use to repair damage to their chromosomes) in the yeast Saccharomyces cerevisiae to rapidly build the entire bacterial chromosome from large subassemblies.

He even gives some details (don't try this at home):

The process to synthesize and assemble the synthetic version of the M. genitalium chromosome began first by resequencing the native M. genitalium genome to ensure that the team was starting with an error free sequence. After obtaining this correct version of the native genome, the team specially designed fragments of chemically synthesized DNA to build 101 “cassettes” of 5,000 to 7,000 base pairs of genetic code. As a measure to differentiate the synthetic genome versus the native genome, the team created “watermarks” in the synthetic genome. These are short inserted or substituted sequences that encode information not typically found in nature. Other changes the team made to the synthetic genome included disrupting a gene to block infectivity. To obtain the cassettes the JCVI team worked primarily with the DNA synthesis company Blue Heron Technology, as well as DNA 2.0 and GENEART.

From here, the team devised a five stage assembly process where the cassettes were joined together in subassemblies to make larger and larger pieces that would eventually be combined to build the whole synthetic M. genitalium genome. In the first step, sets of four cassettes were joined to create 25 subassemblies, each about 24,000 base pairs (24kb). These 24kb fragments were cloned into the bacterium Escherichia coli to produce sufficient DNA for the next steps, and for DNA sequence validation.

The next step involved combining three 24kb fragments together to create 8 assembled blocks, each about 72,000 base pairs. These 1/8th fragments of the whole genome were again cloned into E. coli for DNA production and DNA sequencing. Step three involved combining two 1/8th fragments together to produce large fragments approximately 144,000 base pairs or 1/4th of the whole genome.

At this stage the team could not obtain half genome clones in E. coli, so the team experimented with yeast and found that it tolerated the large foreign DNA molecules well, and that they were able to assemble the fragments together by homologous recombination. This process was used to assemble the last cassettes, from 1/4 genome fragments to the final genome of more than 580,000 base pairs. The final chromosome was again sequenced in order to validate the complete accurate chemical structure.

But the real kicker was this comment:

“This is an exciting advance for our team and the field. However, we continue to work toward the ultimate goal of inserting the synthetic chromosome into a cell and booting it up to create the first synthetic organism,” said Dan Gibson, lead author.

Yup, you read that correctly: we're talking about porting and then *booting-up* an artificial genome, aka digital code of life.

Open Politics

One sphere where openness is generally acknowledged as indispensable is politics: true democracy can never be opaque. In the past, providing that transparency has been hard, but with the advent of Web access and powerful search technologies, it has become markedly easier. Despite that, there are still very limited resources for searching through the raw stuff of politics.

A new pilot project, called Hansard Prototype, may help to change that:

This site is generated from a sample of information from Hansard, the Official Report of Parliament. It is not a complete nor an official record. Material from this site should not be used as a reference to or cited as Hansard. The material on this site cannot be held to be authoritative. Material on this site falls under Crown and Parliamentary Copyright. Within these copyright constraints, you are encouraged to use and to explore the information provided here. We would be especially interested in requests for functionality you have.

Even though it's still limited in its reach, playing with it is instructive. For example this search for "genome" not only throws up various hits, but also shows graphically when they occurred, and ranks the names of speakers.

It's also got the right approach to code:

What technology has been used to build and run this site? Code: Visible Red, Moving Flow. Hosting: Joyent Accelerators. Server OS: OpenSolaris. Database: MySQL. Web server: Apache. Application server: Mongrel. Code framework: Ruby on Rails. Source code control: Subversion. Search engine: Lucene, Solr. Backup: Joyent Bingodisk. Development and deployment platforms: Mac OS X, Ubuntu.

The source code for this site will be made available under an open licence.

More please. (Via James Governor's Monkchips.)

The Genome Goes Read-Write

Good Craig:

Craig Venter, the controversial DNA researcher involved in the race to decipher the human genetic code, has built a synthetic chromosome out of laboratory chemicals and is poised to announce the creation of the first new artificial life form on Earth.

...

Mr Venter said he had carried out an ethical review before completing the experiment. "We feel that this is good science," he said.

Bad Craig:

He has further heightened the controversy surrounding his potential breakthrough by applying for a patent for the synthetic bacterium.

The old dichotomy....

Maybe Genomics is Getting a Little Too Personal

So Jim Watson's genome will soon be made public. But not all of it:

the only deliberate omission from Watson's sequence is that of a gene linked to Alzheimer's disease, which Watson, who is now 79, asked not to know about because it is incurable and claimed one of his grandmothers.

The trouble is, the better our bioinformatics gets, the more genes we will be able to analyse usefully, and the better our ability to make statistical predictions from them. Which means that more and more people will be snipping bits out of their public genomes in this way. And which also means that many of us will never put any of our genome online.

Open Source Bacteria

Another reason to understand openness:

When a team of geneticists unlocked the secret of the bug's rapid evolution in 2005, they found that one strain of multidrug-resistant Acinetobacter baumannii carries the largest collection of genetic upgrades ever discovered in a single organism. Out of its 52 genes dedicated to defeating antibiotics, radiation, and other weapons of mass bacterial destruction, nearly all have been bootlegged from other bad bugs like Salmonella, Pseudomonas, and Escherichia coli.

In the open source world of bacteria, everyone is working for the resistance.

Locking Down the Digital Cat

This is what happens when the genetic code of an animal matters, as here, with "the world's first scientifically-proven hypoallergenic cat":

Purchaser shall not sell or transfer any Cat purchased hereunder to anyone other than an immediate family member, and shall not offer to any person the purchase of a Cat or any genetic material from a Cat, the rights Purchaser may have under this Agreement, or any other right related hereto, without the Company’s express written authorization.

In other words, you don't really own the cat, you just have a licence to use it, like any other software. And don't even think about giving a copy of the genome to a friend.... (Via BoingBoing.)

First Catch Your Neanderthal

This stuff is getting too easy.

First, find some ancient remains - Croatian Neanderthal bones are great. Next, sequence lots - at least 20 times coverage. Don't worry if all you're getting are tiny fragments with around 100 DNA letters, and the signal is vastly swamped by bacterial noise. Just bung the results into a computer, and tell it (a) to cancel out all bacterial genome sequences (b) to join up all the rest. Result: one Neanderthal genome.

There's just one problem:

If the Neanderthal genome were fully recovered, it might in principle be possible to bring the species back from extinction by inserting the Neanderthal genome into a human egg and having volunteers bear Neanderthal infants. There would, however, be great technical and ethical barriers to any such venture.

Understatement of the Year, Number 369.

Going to the Dogs

My heart leapt last week upon seeing the latest issue of Nature magazine. The front cover showed the iconic picture of Watson and Crick, with the latter pointing at their model of DNA's double helix. A rather striking addition was the boxer dog next to Crick, also gazing up at the DNA: inside the journal was a report on the first high-quality sequencing of the dog genome (a boxer, naturally).

This is big news. Think of the genome as a set of software modules that form a cell's operating system. Every change to a genome is a hack; like most hacks, most changes cause malfunctions, and the cell crashes (= dies/grows abnormally). Some, though, work, and produce slight variants of the original organism. Over time, these variations can build up to form an entirely new species. (In other words, one way of thinking about evolution is in terms of Nature's hacking).

Mostly, the changes produced by these hacks are small, or so slow as to be practically invisible. But not for dogs. Humans have been hacking the dog genome for longer than any other piece of code - about 100,000 years - and the result can be seen in the huge variety of dog breeds (some 400 0f them).

Getting hold of the dog genome means that scientists have access to this first Great Historical Hack, which will tell us much about how genomic variation translates to different physical traits (known as phenotypes). Even better - for us, though not for the dogs - is that all this hacking/interbreeding has produced dogs that suffer from many of the same diseases as humans. Because particular breeds are susceptible to particular diseases, we know that there must be a strong genetic element to these diseases for dogs, and so, presumably, for humans (since our genomes are so similar). The different breeds have effectively separated out the genes that produce a predisposition to a particular disease, making it far easier to track them down than in the human code.

That tracking down will take place by comparing the genomes for different breeds, and by comparing dog genomes against those of humans, mice, apes and so on. Those comparisons are only possible because all this code is in the public domain. Had the great battle over open genomics - open source genomes - been lost at the time of the Human Genome Project, progress towards locating these genes that predispose towards major diseases would have been slowed immeasurably. Now it's just a matter of a Perl script or two.

Given this open source tradition, and the importance of the dog genome, it's a pity that the Nature paper discussing it is not freely available. Alas, for all its wonderful traditions and historic papers, Nature is still the Microsoft of the science world. The battle for open access - like that for open source - has still to be won.