A different singularity

Ferrante and I have a lot of shared interests.  Video games, music, music games, and scifi/fantasy to name a few.  But we’re individuals and of course we have our differences.  He’s got comics; I’ve got sports.  He sports a full beard, while my facial hair is more akin to that of a middle schooler.  And he’s got computers, while I’ve got science.  I’m not saying I’m not computer literate.  I am.  But I can’t (also won’t) make my own computer.  Full control over the GHZ or whatnot is outside the realm of things I need from the machine I’m currently typing on.  Hell, the Dell laptop I’m using has a broken graphics card fan that makes a high pitched whining noise sometimes.  Most computer nerds would probably have a seizure over that.  Anyway, I fill the void left by lack of computer skills with science.

Fortunately, science and computer technology are not too terribly far apart (hence the name computer science?).  And they’re getting closer all the time.  So sometimes we have a meshing of our individual fields, like when scientists start using bacteria for problem solving. The method is pretty cool.  They inserted a plasmid carrying mixed up pieces of an antibiotic resistance gene into E. Coli.  They then inserted a Salmonella enzyme to randomly flip genetic material, waited for a bit, and exposed the E. Coli to antibiotics.  Any bacteria that survived would have had to form the entire resistance gene, thus “solving” the problem.  I’d be interested to know where they go next with this kind of technology.  Clearly the DNA computing system they’ve created can solve certain problems much faster than a normal PC.  But the obvious problem is priming the system so that it actually solves the problem.  You have to put in the plasmid/enzyme/etc that’s akin to the code for a computer program.  The more complex problems you want to solve with bacteria the more “stuff” you have to prime the system with and, as every scientists knows, there are always consequences of putting foreign materials into living things.

On the other hand, I’m looking forward to a day when I try to calculate something in Excel and my computer transfects bacteria to do it.

Hypochondriacs FTW

John Timmer over at Arstechnica has an good rundown from the World Science Festival.  Specifically, he summarizes a panel about the prospects of genetic testing.  Not like the simple genotyping currently done to see if patients are carrying mutations or markers for a select few disorders but whole genome scans, producing a vast array of information for multiple phenotypes.  The holy grail of genomics is a scenario in which your DNA can be easily and quickly sequenced, risk factors in your genetic code identified, and therapies prescribed.  Clearly that scenario does not currently exist.  All the panelists agreed on one thing: genetic testing hasn’t gotten to a point where it’s a viable basis for treatments.  And I completely agree with that.  Right now the technology does little more than provide hypochondriacs lots of stuff to worry about.

The problem I have is the negative vibe that comes across from some of the panel.  Sure, running a SNP chip on your DNA right now would accomplish practically nothing.  The chip would be laughably incomplete and you’d only be getting a test for the fraction of disease-related SNPs that we’ve identified.  That doesn’t mean the technology should be shunned or put down.  You have to crawl before you can walk, other trite phrases, etc.   Technology is an exponentially growing field.  And as the gaps in our knowledge base get filled in, we’ll be able to provide better and better diagnoses for patients.  Each individual marker or polymorphism may only be a tiny piece of information when determining disease risk.  If I have a G instead of a T somewhere, maybe I have a 0.5% higher risk of getting Disease X.  But putting together a large number of them forms a foundation for a quality prediction.  Now you’ve got the genotype for hundreds or even thousands of revelant points in the genome.  And their benefit is exponential as we learn not only how they relate to risk by themselves but in conjunction with the other polymorphisms.  We may never reach the holy grail scenario I mentioned earlier (possibly because of environmental factors) but the potential is still there and that’s a reason for optimism.  Besides…if we don’t push forward with genetic testing the hypochondriacs will be stuck in the past, using WebMD to diagnose themselves with various ailments.  Won’t somebody please think of the hypochondriacs!?