Slashdot Mirror
Writing Genetic Code
An anonymous reader writes "The Globe and Mail is reporting on another group of researchers delving into the field of 'synthetic biology.' The project stemming from the efforts of two biology labs in British Columbia and Maryland is attempting to create the first synthetic life form. From the article: 'The project is being spearheaded by U.S. scientist Craig Venter, who gained fame in his former job as head of Celera Genomics, which completed a privately-owned map of the human genome in 2000. Dr. Venter, 59, has since shifted his focus from determining the chemical sequences that encode life to trying to design and build it: "We're going from reading to writing the genetic code," he said in an interview.'" This is certainly not the first group to venture into this territory.
... welcome our new two-codon overlords.
"Win treats sysadmins better than users. Mac treats users better than sysadmins. Linux treats everyone like sysadmins."
There you go, no more bitching about which is the best coding language, now you can code in the most natural one.
I hope there is a gcc backend for this. I hate using Visual Studio to write my code.
Just wait until someone writes a piece of code that cures a genetic disease, but must be 'fed' with a certain medication. If not fed with said medication, it will do something real bad.
Job? I don't have time to get a job! Who will sit around and bitch about being broke and unemployed then?
Bacteria are already used to synthesize organic materials by reprogramming their DNA. For example, some antivirals and antibiotics are manufactured this way; the desired pattern is injected into the bacteria's genome and it will then produce that pattern. Venter's project is really just an extension of that approach.
I have doubts as to the likelihood of success using present science; in twenty years, perhaps it will be possible, but today it's really casting about in the dark. Even something as elemental as a bacteria is an incredibly complex thing, with a sophisticated genome and complex organelles working in biochemical harmony to reproduce, to "mate" by conjoining with other bacteria, and to adapt and thrive in a very wide variety of conditions.
Bacteria have been around for billions of years and, as Stephen Jay Gould put it, we are living in the Age of Bacteria. In a few short years it seems unlikely that even brilliant scientists can recreate these things. Modify some, yes, but completely create from scratch something that is going to be viable--well, that's going to be interesting to see.
That said, if they can pull it off the possibilities of its use, for good or evil, are endless. They can be encoded to synthesize all sorts of compounds, eat nasty pollutants, generate fossil fuels, attack disease microbes, or be diseases themselves. Luckily, the human body has a pretty comprehensive immune system that will adapt to just about anything except retroviruses like AIDS that reprogram the immune system itself.
it's = "it is"; its = possessive. E.g., it's flapping its wings.
I've seen WAY too many sci-fi movies to consider this a good scientific endeavor. If you need me, I'll be in my concrete bunker. :)
Genetics code YOU!
Oh wait... they do...
Carry on.
http://www.TheGamerNation.com/Forums
Seriously, are we going to have both propriatory as well as free genetic code organisms? What's the legal status of a living being that's a result/offspring of a crossing/mating between a propriatory and a GPL organism?
If enithin kan gow rong it whil. (Murfey)
The long and short of it is. These "life forms" are significantly different from their naturally occuring cousins. They are this way by nature of human engineering. This argument can also be extended to bacteria that have been highly modified. There are laboratory strains of almost every bacteria that we know of that are significantly different from wild type bacteria. I am curious as to where they will draw the line. From the article is appears that they are paring down mycoplasma to the barest bones.
The other question is, once you have the DNA how do you kickstart the process. They appear to be inserting it into and E. coli with the nucleus removed. This means that the cellular machinery of the E. coli will be used to translate the DNA into protein and eventually a new synthetic cell. Does this mean that it is human created if we use naturally occuring cellular machinery?
I don't mean to detract from the research in any way because it is highly interesting and will tell us a lot about how life works on the most basic level, BUT there are a lot of questions out there and I hope that people keep them in mind as we see this field develop over the next several years.
I think one of the biggest challenges isn't in synthesizing strings of DNA, per se - it's in knowing what DNA to synthesize. The real holy grail of synthetic biology is to engineer genetic functions to accomplish a particular goal - design to spec. From the average /. POV, this means "programming" genes in some high-level language (C++ DNA lib, anyone?). Take a look at The Registry of Standard Biological Parts for a first library of genetic "functions".
As I understand it, the current state-of-the-art in terms of programming DNA is basic logic gates that still tend to lose coherence when connected together. Once this is accomplished (best guess, 3-4 years from now to work out the basic science), all of the sophisticated tools and techniques developed by the IT community over the last decade(s) can be rapidly applied, and that goal of design/build to spec will become possible.
They're going to double freak when it's discovered that human genetics are made up of spagetti code!
One line blog. I hear that they're called Twitters now.
Well, as a C++ programmer, I am quite used to hearing how C++ is evil because of all the things it can do. Therefore, I am as justified as they are in saying that C++ is better than genetic code for the following reasons:
Genetic code is too low-level. While C++ comes with a standard library defining containers, iterators, and common algorithms, in genetic code you have to do everything from scratch. In quaternary. With 3D objects. Talk about a learning curve!
Genetic code has no garbage collector, and not even a simple malloc. In fact, you have to write self-modifying code to avoid memory leaks or dangling references. This makes it very difficult for the beginning programmer to write good code, and encourages bad practices.
Genetic code is not object oriented. You have to do horrible hacks to encapsulate private information or define interfaces to it. Most programmers just use a "signals-and-slots" method to pass messages, resulting in spaghetti code rivaling the worst abuses of goto.
Genetic code is too flexible. If you thought bad C++ code was hard to understand due to operator overloading, wait 'till you see the things a bad programmer can do with genetics! And, while in C++ the worst that can happen is a crash, bad genetic code could eat you.
Genetic code takes longer to develop for. You have to write lots and lots of code to duplicate even the simplest C++ line. Furthermore, compilation times totally suck, approaching twenty years for complex programs!
Genetic code has an arcane syntax, leftover from the early days of evolution. Just imagine, we're still using constructs nearly three billion years old! If you thought having some C in C++ was bad, wait 'till you see the archean leftovers you are forced to use in your eukaryotic cells!
Genetic code is dynamically typed and favors the "duck typing" philosophy. This creates an enormous amount of security holes, where special ducks ("poisons") with appropriate appearance but malicious behaviour could be introduced into the system.
Genetic code is hard to debug. Having no debugger, one has to rely on contrived printf-like trace statements. Unlike printf, the genetic equivalents are limited in number and expressiveness, sometimes making it impossible to figure out what is wrong.
Genetic code is a bloated pig. Just imagine, you need trillions of bits to define a simple organism, while in C++ I can code NPCs in under a hundred lines of code.
Genetic code VM is slow. Perhaps not as slow as Java, but it still takes milliseconds to do even simple operations. We could all think so much faster if we were written in C++.