It's not as much data as one might think, especially if you store the deltas -- there's not that much variation between individuals. That fact has actually been exploited to a great extent in enabling the $1,000 genome.
How much is "not much?" How many "values?" If stored as byte strings delimited by \n, how many lines? (I don't know if these are even answerable questions at this point, just wondering.)
The raw data (the original image files), while mostly transient today, is massive. At some of the big genome centers they are in the double digit TB's of data processing per day
So DNA is a long string, with certain identified parts that have meaning (code for something we know). What would a layman like myself need to get myself to a hello world program that could take an actual DNA string(file) as input and give a hello for the presence an actual known gene? Not code-wise, that's obvious. I'm asking where I could get a sample genome, information about a gene, fill in the gaps of my understanding, etc. Anyone know?
Honestly, I believe that people need to understand the biology, some depth of what genes are, and what SNPs are etc. It's important as molecular medicine becomes the norm (still some ways away).
You can start with your favorite language by downloading a library, there is bioPerl, bioPython, bioRuby and even bioPHP!
The libraries have all the routines to connect to public sources.
The National Center for Biotechnology Information http://www.ncbi.nlm.nih.gov/nuccore/89889045?report=fasta... has an enormous database and is all open source. The link I quoted lists the genome for a virus. I wrote a short program in JavaScript that did frequency analysis on the pairs. It took 30ms in Chrome. It is actually not all that difficult at a certain level of analysis.
In my mind the big question is, how can we understand how it works? It is like giving you a binary for windows to reverse engineer it. My own question is; does my DNA makes me? If my grandkids have it on a memory stick (it can actually fit), would they be able to 're-construct' me in some form of virtual reality?
A potential area for hackers to get involved in my opinion - is the potential of the Cloud. You can imagine say 10 million genomes stored out there at the not so distant future and distributed software analyzing it and indexing it and all the ethical and other problems that come from it.
Somebody needs to get on top of appropriate encryption and anonymization for this area. After all, there's nothing I want more private than my DNA sequence (against health insurers, if no one else), and nothing that is less anonymous than something that can identify me uniquely out of all the world.
At the same time, I want my doctors to have access for testing before they begin a course of treatment.
I would also like to participate in studies that use my genome for others' benefit and data mine it, but appropriate safeguards and protocols will be a must. I want a way to share part of my sequence, to compare one sequence to another, or to perform a test on the sequence, and be secure doing so.
I take a very different approach: When I have it, I'll gladly share my entire genome with anyone who wants it (after all, with the rapid decrease in price, and the ease of collecting a sample from me without my knowledge, any sufficiently determined person won't even have to ask my permission in a couple of decades).
But, I'll always be very strongly in favor of legislation that prevents entities like health insurance companies from using my genome against me.
I don't think your genome going public is mostly in danger from nefarious ne'er-do-wells collecting your skin.
But there will be mass data breaches from stolen laptops or social engineering scams at these sequencing/medical record companies. Past experience tells us that the data will not be secured properly. So if you care about it enough, the right solution is a more federated solution where the DNA owner controls the key, like PGP.
Insurance companies have been known to hire private investigators to check up on their customers. If it was legal and cheap enough, why wouldn't they give genetic material collectors?
In any case, I agree with you - there will be mass data breaches, so I prefer to just assume all genome data will be essentially public, and then figure out what to do from there.
> But, I'll always be very strongly in favor of legislation that prevents entities like health insurance companies from using my genome against me.
Would you also want legislation to prevent them from using it in your favor as well? In all likelihood those who can prove they are 'good risks' will find a way to benefit from this.
It's hard to keep it anonymous when you shed it everywhere you go, and it's not illegal for people to pick up those particles, analyze them, and hold it against you in a court of law.
18 comments
[ 2.7 ms ] story [ 52.5 ms ] threadThe $1,000 genome is indeed within sight.
Let us hope that software will be evolving as fast to handle the data.
http://www.snpedia.com/index.php/Promethease
23andme's software does a pretty good job.
You can also use the UCSC genome browser although it probably as a learning curve for most.
http://genome.ucsc.edu/
Honestly, I believe that people need to understand the biology, some depth of what genes are, and what SNPs are etc. It's important as molecular medicine becomes the norm (still some ways away).
The National Center for Biotechnology Information http://www.ncbi.nlm.nih.gov/nuccore/89889045?report=fasta... has an enormous database and is all open source. The link I quoted lists the genome for a virus. I wrote a short program in JavaScript that did frequency analysis on the pairs. It took 30ms in Chrome. It is actually not all that difficult at a certain level of analysis.
In my mind the big question is, how can we understand how it works? It is like giving you a binary for windows to reverse engineer it. My own question is; does my DNA makes me? If my grandkids have it on a memory stick (it can actually fit), would they be able to 're-construct' me in some form of virtual reality?
A potential area for hackers to get involved in my opinion - is the potential of the Cloud. You can imagine say 10 million genomes stored out there at the not so distant future and distributed software analyzing it and indexing it and all the ethical and other problems that come from it.
At the same time, I want my doctors to have access for testing before they begin a course of treatment.
I would also like to participate in studies that use my genome for others' benefit and data mine it, but appropriate safeguards and protocols will be a must. I want a way to share part of my sequence, to compare one sequence to another, or to perform a test on the sequence, and be secure doing so.
But, I'll always be very strongly in favor of legislation that prevents entities like health insurance companies from using my genome against me.
But there will be mass data breaches from stolen laptops or social engineering scams at these sequencing/medical record companies. Past experience tells us that the data will not be secured properly. So if you care about it enough, the right solution is a more federated solution where the DNA owner controls the key, like PGP.
In any case, I agree with you - there will be mass data breaches, so I prefer to just assume all genome data will be essentially public, and then figure out what to do from there.
Would you also want legislation to prevent them from using it in your favor as well? In all likelihood those who can prove they are 'good risks' will find a way to benefit from this.
http://hplusmagazine.com/articles/health-medicine/open-sourc...