Tag Archives: genetics
In the last session of the day, we had a few experimental talks on noticing how food changes physical condition. It was also an interesting series of talks that shows the importance of collecting our own subjective data to back up or refute the other technological data that we might also have access to.
I kicked off the session with my talk “Quantifying My Genetics: Why I have been banned from caffeine”. My colleagues and friends helped me quantify my behavior after one, two, or three cups of coffee by giving my agitation a number from 0-10.
I found out that I’m a slow caffeine metabolizer from my genetic results and it seems like there is a correlation between how caffeine affects me and my genes. My genes are not deterministic, I couldn’t have known how caffeine affects me without making my own independent observations.
On a fun note, the crowd guessed that I had one cup of caffeine today, they were right, I had a cup of tea earlier down in the restaurant, away from the conference.
Next we had Martha Rotter who talked about how she experimented with her diet to solve her skin problems after doctors told her there was not much she could do. She did one allergy test where the results said she was allergic to chicken and soy- but after cutting out both of those foods, she did not see any changes but it gave her the idea to test different food groups.
After her experiment with a chicken and soy-less diet, she tried a few other food groups, eventually hitting on cutting out dairy. Her skin cleared up within two weeks of stopping drinking milk, eating cheese.
I think the take away message from our two sessions this afternoon, don’t be afraid to do your own testing, trust in your results.
I led a very interesting discussion at Quantified Self Europe this morning with about 10 attendees with a variety of backgounds. There were entrepreneurs who wanted to start genetic information based companies, a designer, a think tank analyst, and people who are just interested in where the field is and where it was going.
The first thing we did was to create an impromptu community, putting the chairs into a circle and starting the discussion with what brought each of us to the topic of Creating Genetic Communities. This is where the conversation started but the topics ranged from where the industry is going to how to use design to help a non-technical audience understand their genetic data.
There was agreement in the room that the price of DNA sequencing is decreasing exponentially, the discussion then moved to what an individual can do with their own data vs. getting aggregate genetic data. There was an intense debate about open data vs. transparency of who has access vs. private databases. There was also a challenge thrown out by two group members to make genetic information more actionable.
It’s an amazing group of people who have come together in Amsterdam to discuss where we are going as we get more data on ourselves.
Here’s your chance to learn how to live longer, and save money too.
Christine Peterson is hosting the first Personalized Life Extension Conference, October 9-10 at the San Francisco Airport Marriott. She is offering a $100 discount on the $275 registration price to all Quantified Self members who register with the discount code “QS”.
6 months ago, I got my 23andMe genetic test results. They showed mostly what I expected: 30% chance of diabetes, 24% chance of atrial fibrillation, 40x greater risk of Celiac disease than the general population. All of these things are found in my extended family to some degree.
But there was one thing I didn’t expect.
I have double the normal risk of Tourette’s syndrome. Yes, Tourette’s is commonly associated with people walking down the street swearing and thrashing around uncontrollably. That’s not me, but it did get me thinking.
My dad, brother, and I all have tics. They had always bothered me, but I learned to internalize most of them so they aren’t noticeable unless you’re looking very carefully or I let down my guard. They are part of the reason that I can only see people 3 days a week, because I need the other 4 days to recover from the effort. I began to wonder, what if I could address the tics and live life with more ease? So, armed with the new 23andMe information, I started to investigate this part of my health that I hadn’t really looked at in detail before.
The first thing I did was to write down all the different tics I do, which are almost all on my left side. My theory as to why I don’t have many of the vocal tics characteristic of Tourette’s is that perhaps only my right brain is affected – so the right side of my body and my language center, both controlled by my left brain, are mercifully free.
Here are all the things I do, mostly involuntarily, every day (after the jump):
The cheapest commercial genome testing right now is from 23andMe for $400. Prices in this area will continue to drop, while the number of genes sequenced rise. However nothing beats free. You can now get your genome sequenced (partially) for free by participating in a large-scale research program to try to correlate genes with disease. The Coriell Personalized Medicine Collaborative (CPMC) is being funded by charitable foundations, and they have money at present to sequence 10,000 volunteers. To get your genes sequenced for free there are several caveats.
1) You need to be over 18
2) You need to attend an educational session. At the moment these are only offered in Camden, New Jersey (near Philadelphia). They claim to be working on a mail-in version later.
3) You won’t get your gene code back. Instead you will only receive data that is “medically actionable.” In other words you will only get reports about genes that their board of doctors feel you can do something about.
The key phrase here is “board of doctors.” Unlike commercial services which return your full test results and let you do what you want with this data, this survey is run by doctors who feel ethically obligated to offer responsible medical counsel, and so they will not tell you about genes that have no medical value, or about which the science is not certain in their opinion.
For some people this is the doctor priesthood exerting their control over your health options (they would like companies like 23andMe shut down unless they let doctors take control). For others, this is a good deal. Free testing, plus free doctor advice about what is worth paying attention to and what is just fluff.
My long-term prediction has been that pharmaceutical companies will eventually pay for your genome sequencing in full since they can target drugs to specific genetic cohorts and avoid those patients with genes that may produce negative side effects. But again you may not get your full sequence back. But as in the rest of life, there is no such thing as a free lunch.
To clarify what kind of results you get back, here are some excerpts from the CPMC FAQ:
This study will only report back to participants those genetic variants that are potentially “medically actionable.” Potentially medically actionable genetic variants are those for which 1) there is a scientifically valid association between the variant and a specific health condition, 2) there are actions or interventions that can be taken to reduce the risk of the health condition, and 3) the risk of adverse events from these possible interventions is likely small in relation to the risk associated with the genetic variant if no medical action is taken.
You WILL NOT receive results for all genetic variants. Genetic variants associated with medical conditions for which there is no treatment or intervention to reduce the risk of disease WILL NOT be reported back to participants. For example, variants elevating risk for incurable diseases such as Alzheimer’s disease will not be reported. If a new therapy or lifestyle intervention is reported, the ICOB may update a condition to be “potentially medically actionable.”
The technology employed by the CPMC™ is not designed to detect single-gene mutations that cause rare Mendelian disorders such as sickle cell disease, cystic fibrosis and Tay-Sachs; therefore, these are very unlikely to be detected and reported to you.
Ann Turner, co-author of the best book on DNA-based genealogy: Trace Your Roots With DNA, wrote me to say that she too has been comparing results from the two big genetic test companies, 23andMe and deCode. She wrote in response to my earlier posting comparing results between the two vendors.
The big news is that places where errors are showing up are probably not random. Here’s the argument, starting with her post on ancestry.com:
The two companies overlap on 562,532 SNPs. They agreed on 560,128 calls, or 99.6%. 23andMe didn’t make a call on 1,970 SNPs where deCODEme did, and deCODEme didn’t make a call on 399 records where 23andMe did. That leaves a mere 35 records where they actually made different calls [see the list below]. In all of those cases, one company would make a homozygous call while the other company made a heterozygous call — there were no cases where they made a completely discordant call.
Here’s the kicker from Ann’s letter to me:
Four of those (rs11149566, rs4458717, rs4660646, and rs 754499) were also found in Antonio’s list. That’s more than you would expect by chance.
Four out of 23 from Antonio’s list and four out of 35 on Turner’s list of discordant results indicates that these regions (at least) are unreliable.
This is why sharing results is so valuable and a key to great quantified self understanding.
This is a micrograph of the bead array on which these tests are conducted.
Turner’s 35 SNPs with different results, if case you also have done a comparison.
rs10435795 rs1045363 rs10743414 rs10945383 rs11149566 rs11179382 rs11707159 rs11915402 rs1209171 rs1221986 rs12907462 rs1303912 rs13422439 rs161381 rs17328647 rs1961196 rs1966357 rs2016461 rs2064034 rs2290516 rs2853981 rs3952469 rs4336661 rs4423481 rs4458717 rs4572718 rs4660646 rs6531490 rs6942478 rs7102702 rs754499 rs7812884 rs845217 rs9332128 rs9476380
I am taking a crash course in genetic literacy by having some of my genes sequenced by the two major genetic sequencing services, 23andMe and deCode. I am still in the process of comparing the two sets of results to see which vendor is better, but while coming up to speed in this new realm, David Ewing Duncan, another self-experimenter, turned me onto a very cool site: SNPedia.
SNPedia is a wiki for personal genomic raw data, which come in SNPs, or in my usage, snips.
Snips are the current desired unit in personal genomics, like pixels in digital photography. For now, more snips is better. A snip is a particular part of the gene that researchers have noticed will vary between individuals (most of the gene does not vary). These single-nucleotide polymorphism (SNP) variations are the tiny spots on your chromosome that are actually sequenced and reported back to individuals. Each snip position has a unique number, and some of these snips such as rs1815739 (good sprinter) or rs795174 (green eye color) indicate particular traits.
On SNPedia you can paste in a snip number — from the results of your DNA sequencing — and find out what is currently known about it. Or conversely you can enter a trait or disease and see if there is a snip tagged to it. As information is gleaned from medical journals, wikians add it to the SNPedia. Usually very quickly.
A typical entry will look like this:
rs7495174 is located in intron 1 of the OCA2 gene. The (A) allele (in dbSNP orientation) is associated with blue or green eye color in Caucasians. [PMID 17236130].
This SNP is 1 of 3 SNPs defining a haplotype that has been studied for association with eye color. The full details on the correspondence between the haplotype and eye color can be found on the OCA2 page.
In theory this is what the news personal genomics sites of 23andMe and deCode are supposed to be doing. Only with slick user-friendly designs. They do offer this information, but in their effort to filter this large deluge of data, they both are selecting certain snips as being more important/interesting, and hiding the rest in the page pages of their sites. To surf the ocean of data beyond these selected traits, diseases, or snips is cumbersome. And of course, it costs $1,000 to enter the door — the price of getting your DNA sequenced at either place.
Here is how I see the nascent field of personal genomic testing shaping up. The retailers are 23andMe and deCode. They don’t sequence genes. They outsource that specialized job to microarray manufacturers, while the retailers sell the website interface, and supporting information to consumers. There are only two main manufacturers of the large scale mircoarray chip which is used to provide up to 1 million SNIPS. One is Affymetrix, and the other in the Illumina. Affymetrix and Illumina will sell their microarrays to anyone, although they currently don’t sell to individuals. Affymetrix lists the cost of a 500,000 SNIP array chip at $250 today. You need their specialized machines and software to read the chip.
If a third-party vendor were to start selling the naked chip’s data for a small fee above its costs, it would be possible to do a large personal sequence using one of these tests and managing your data using open-source wiki technology like SNPedia. Hard-core recreational genomist could probably do a better job than either 23andMe or Decode are doing right now.
This is close to a DIY kit for geneboys. With some mashing of websites, you could get more info in, faster, more personalized to you. If you are already doing this, write me.
Like other early 23andme customers, I’ve been struggling to find something interesting to do with my genetic results. After quickly learning [what kind of earwax](http://www.nytimes.com/2006/01/29/science/29cnd-ear.html) I’m predisposed to have, the path fades out amidst a tangle of SNPs.
Here’s an example of a typical quest, and its results.
It is well known that particular variants of the ApoE gene, which is found on chromosome 19, are associated with an [elevated risk of Alzheimer's ](http://www.nia.nih.gov/Alzheimers/Publications/geneticsfs.htm)disease. Other variants are associated with reduced sensitivity to statins, which are commonly used to lower cholesterol in patients at risk for heart disease. [Lab Tests Online](http://www.labtestsonline.org/index.html) provides a good, brief description of how [genetic testing of the ApoE gene](http://www.labtestsonline.org/understanding/analytes/apoe/test.html) can be used in mainstream medical practice.
Reporting an upcoming Wired story on a different topic, I recently visited an M.D. outside Denver who takes an aggressive approach to longevity treatment, and found that he routinely orders ApoE tests for his patients, as part of a panel of 11 genetic markers that costs about $500. He told me that if I wanted just the ApoE results, I could get them for about $90.
But who needs to pay by the gene when you’ve got 23andme? More than 600,000 SNPs for $1000 bucks is awfully cheap compared to $90 for a single genetic test, or so it seems.
I began hunting around the Web. I quickly discovered that I was following in the footsteps of Mark Fletcher, also an [early 23andme adopter](http://www.wingedpig.com/archives/2007/12/23andme_results_266.html). When I read Fletcher’s entry, I knew I had found a fellow traveler. Like me, Fletcher immediately wrote to 23andme upon getting his results, asking if he could download them in a more convenient format for analysis. He got the same email back that I did:
> At this time 23andMe does not have a way to give customers their genetic data on a CD, flash drive or other downloadable or stored format. But we are working to make that possible, and hope to be able to distribute raw data to our customers in the near future.
> The 23andMe Team
There is of course something unconvincing about this answer. Having launched a company premised on giving consumers access to their genetic data, it is hard to believe that 23andme can’t manage to make this data available in a standard, easily downloadable format. But they are just beginning, so let’s cut them some slack. Besides, we are curious about ApoE.
Like Mark, I first consulted the great [SNPedia](http://www.snpedia.com/). Here, I learned that ApoE variants, known as alleles, are indicated by the following SNPs:
> ApoE2 allele = rs429358(T) + rs7412(T)
> ApoE3 allele = rs429358(T) + rs7412(C)
> ApoE4 allele = rs429358(C) + rs7412(C)
That’s just what I needed to know. I went to 23andme and entered rs429358 into the Genome Explorer. My results:
Then I entered the rs7412 into the Genome Explorer. My results:
According to SNPedia, another SNP, [Rs4420638,](http://www.snpedia.com/index.php?title=Rs4420638) is relevant to my quest. Says SNPedia:
> One form of this SNP is very commonly inherited together with the APOE variant ApoE4 and thus predicts its presence.
So I searched on Rs4420638. My results:
Mark went on a largely parallel journey, ending in similar [non-satisfaction](http://www.wingedpig.com/archives/2007/12/23andme_and_snpedia_267.html).
Mark Fletcher says he spoke to another 23andme user who had a no call at the Rs7412 location. Today I received an email from yet a fourth user who had a no call at this location, too. So far, 23andme is zero for four at this SNP.
Like many early adopters, my interest in sequencing my genome is driven more by intellectual curiosity than by pressing health reasons. So I’m hoping that George Church gets his [Personal Genome Project](http://www.technologyreview.com/BioTech/wtr_16169,259,p1.html) geared up quickly. I’ve registered to participate.
But for now, I’m looking for ways to make my 23andme results more relevant. I welcome suggestions.