Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Levi]

I am curious about my genomic makeup as well (Lots of NASTY mutations in my ancestry) but dont feel comfortable with that information being in the hands of a 3rd party I dont control.

Yeah, I understand that.  I kind of think its going to be impossible to keep that information private in the long term though as it gets easier and easier to do genetic testing.  Especially as medicine moves towards genome-centric treatments and using your genome information to treat diseases becomes more common(We already do this a bit at my HIV lab).

Its weird, I'm more paranoid about keeping my digital files secure than my genome.   

[Arx]

Continuing to interject links with no relevance to the current topic...

Study shows direct correlation between CO₂ levels and thermal radiation levels.

Over 10 years of near-daily observations, the team found that a rise in CO2 concentrations of 22 parts per million boosted the amount of incoming thermal radiation by 0.2 watts per square meter, an increase of about 10 percent. The researchers say their results agree with the theoretical predictions of CO2-driven warming used in simulations of future climate.

[wierd]

Musical topics!

The connection between CO2 levels, changes in thermal equilibrium of gasses, and serious scientific concern over CO2 levels altering climate was made way back in the late 1800s.

I even have a PDF of a vintage scientific paper on that very subject at hand!
http://www.rsc.org/images/Arrhenius1896_tcm18-173546.pdf

That's right, this subject has been "Debated" for over 100 years, and scientists have known and been trying to educate about it that entire time!

Oh, the joys of commercially interested misinformation!

[hops]

Things tend to get a bit sticky when the wide mass debate.

[Levi]

I'm not sure this falls into the science topic, but it sort of does as its a result of our state of technology we have these days.

A story about magic.  I really enjoyed this story, because I think this is a great example of a service company enabled by technology(or lazy people enabled by technology).

Basically, Magic is the service to end all services.  You send a text message saying what you want (anything legal & possible) and when and they will give you a quote on what it will cost to get it to you, and you can say yes and get charged or no and your done.  In the article, the person needs a fish tank and a live fish that day, and they just take care of it.  It sounds pretty cool, and it even accepts bitcoin.  Its only in the US at the moment though, so I can't actually give it a shot. 

I kind of wonder if this is the future of the service industry.  A few simple front-end services that will contract out to get whatever you need done.

[hops]

I really hope a random super-rich american contract them to buy North Korea. That'd be amazing.

[Sheb]

Well, some things could be done efficiently. PCR machine for example: mostly it's because we need to get DNA to denaturate through heat. But you've got stuff like NASBA that don't need this step. So you could get rid of the PCR machine.

[i2amroy]

Really the hardest part of sequencing a genome is the part where you try to understand the results your analyzer machine outputs. Most biologists don't exactly like it when you dump a 3 GB solid block text file onto their desktop and say "here, analyze this". (My university actually offers a BIO/CS class that specifically focuses on writing and using programs to analyze the huge text files of genome sequences that the analyze machines spit out).

[Levi]

Really the hardest part of sequencing a genome is the part where you try to understand the results your analyzer machine outputs. Most biologists don't exactly like it when you dump a 3 GB solid block text file onto their desktop and say "here, analyze this". (My university actually offers a BIO/CS class that specifically focuses on writing and using programs to analyze the huge text files of genome sequences that the analyze machines spit out).

I do this sort of thing at work all the time.    We have one of these:  http://www.illumina.com/systems/miseq.html

I doubt 23andme actually does sequencing though.  I vaguely remember there was a technique that did something like tossing primers at a region and seeing which ones bind, which would tell you which variants the genes are.  It doesn't give the full on nucleotide sequence, but it'll identify things if it matches something in your primer library.

Edit:  Yeah, looks like that is it.

[alway]

So, remember that thing last page about modifying thoughts? Yeah, well, here's a company claiming to have a similar (early prototype of a) consumer-grade device: http://www.popsci.com/will-be-year-our-smartphones-link-our-brains

It's a milder version of the stuff I had mentioned, with the aim of letting you essentially push your brain towards certain moods (calm/relaxed or focused/alert are mentioned). Interestingly, it also mentions that they had been experimenting with ultrasound initially, but came to the conclusion it was pretty much useless.

Who knows how legit it is until they have solid independent results, but its methods are very much in line with what I had been saying.

[hops]

Creepy implications notwithstanding, that would be useful with depression.

[alway]

There actually has been some study of that: http://well.blogs.nytimes.com/2013/07/01/new-approach-to-depression/
Though those methods are almost certainly quite a bit more powerful, due to being intended for clinical use by clients screened for any potential issues, rather than consumer-grade use-at-home devices. Based on the description of the sound it makes, it sounds like the same sort of device they were using to override part of the brain for the first experiment I linked.

[wierd]

Really the hardest part of sequencing a genome is the part where you try to understand the results your analyzer machine outputs. Most biologists don't exactly like it when you dump a 3 GB solid block text file onto their desktop and say "here, analyze this". (My university actually offers a BIO/CS class that specifically focuses on writing and using programs to analyze the huge text files of genome sequences that the analyze machines spit out).

I do this sort of thing at work all the time.    We have one of these:  http://www.illumina.com/systems/miseq.html

I doubt 23andme actually does sequencing though.  I vaguely remember there was a technique that did something like tossing primers at a region and seeing which ones bind, which would tell you which variants the genes are.  It doesn't give the full on nucleotide sequence, but it'll identify things if it matches something in your primer library.

Edit:  Yeah, looks like that is it.

Exactly, but I personally wonder why they don't leverage DNA's natural semiconductor properties, instead of tagging the SNPs with photopigments.

Oh, Look-- Somebody is working on that!
http://www.technologyreview.com/news/424768/a-semiconductor-dna-sequencer/

To me, it seems like a no-brainer to use a simple aggregation technique with a bunch of redundant SNP sensitive regions per SNP of interest, and use the binding mechanic itself to turn the bead from insulator to organic hybrid semiconductor. That way all you need to do to read the chip is literally plug it into a computer, collect the data, and do a statistical analysis.  Something like 50 to 100 instances per SNP of interest randomly located on the chip, so that good signal isolation is possible per chip. (that way even if there are slight defects that would inhibit good substrate interaction with the ligated sequences, [say, a fingerprint or something] reasonable confidence can still be obtained from the run.) 

A special USB dongle, a DVD with the analysis software, a wetware kit with single use plastic lab goodies and refined enzymes for processing, a semiconductor based SNP assay chip, (and lots of time to let the computer grunt while you go to work for the day) and you could theoretically have a fully at-home "For your personal amusement" DNA testing kit.

[Sheb]

Ion torrents is not really "harvesting the DNA's semiconductor property" though. It's fairly similar to all the other High Throughput Sequencing machine, except it uses the pH change associated to the addition of a new base pair instead of light to read the addition of new base pair.

Also, you seems confused about what SNP means. It stands for Single Nucleotide Polymorphism, aka a 1 base pair change between two genomes (often a subject's genome and some "reference" genome for medical applications).

Hey, a shout to all the other molecular biologist/genetic engineers/bioengineers/biochemists/synthetic biologist out there, what would be your pet project if you had lab equipment in your garage (let say you manage to grab a bunch of second-hand stuff from your workplace)?

[wierd]

No, I knew what an SNP was.  As stated, it's a single nucleotide polymorphism-- A change of a single base pair in a gene coding region that is not 100% conserved between humans, and is part of the information in a person's genome that makes them unique. (The unique constellation of these polymorphisms is what sets an individual human apart from other humans, genetically.) I think it more likely that I just didnt explain well enough, and caused confusion. (That's a problem I often have.)

I was meaning, that for each unique sequence, you have around 50 to 100  beads that become semiconductive after the unique DNA sequence docks.  (So if you have 100 possible variants on a coding region you are assaying against, and 100 beads for each, that's 10,000 beads. It is unlikely that you will have 100% successful docking of the DNA with your assay chip, which is why you need redundant beads per unique conformation. You can overcome uncertainty that your "hit" or "miss" is random this way.) One way to accomplish this is to use the DNA molecule as the gate layer of a hybrid organic diode, where only a specifically coded DNA sequence can attach, and form this diode. 

It has been known since at least 1999 that DNA is semiconducting.
http://physicsworld.com/cws/article/news/1999/apr/01/dna-is-a-semiconductor

Assembling a purely electronic detector is well within the realm of possibility.

--------------------

Not a biotech engineer-- just an old IT guy--  but if I were, and had access to such toys, I would adapt upper atmospheric microbes into high temperature extremophiles tailored for life in a very high pH, nitrogen depleted, and chemically arid (eg, surrounded by strongly hygroscopic, nearly anhydrous acids in both vapor and droplet forms) conditions, that is able to synthesize appreciable quantities of aramid plastic, using a sulphur respiration cycle.

It's the kind of thing that I feel is needed to free Venus from its runaway greenhouse effect.

Many useful mechanisms could be lifted from existing terrestrial microbes.  Producing a viable biochemical pathway to synthesize this heavy (and energy dense) polymer would be a feat in and of itself, and getting the many different mechanisms from very disparate places to work nicely in the resulting custom microbe would be even harder.

Probably would not succeed, but it would still be fun.