A gut bacteria for caffeine metabolism?

 23andme says I’m a slow caffeine metabolizer because I have the AC genotype at SNP rs762551. You’d think that means I’m extra sensitive to caffeine before bedtime, but that’s not the case: I sleep just fine even if I have a cup of high-octane coffee after dinner. My 99-year-old grandmother drinks coffee by the potful, crediting its warmth as a calming effect to make her drowsy.

This week's Economist points to a new study in Nature by Javier A. Ceja-Navarro et al that names Pseudomonas Fulva as a bacterium active in the guts of a coffee bean pest. Caffeine is normally toxic to insects, but P. Fulva neutralizes the caffeine, apparently using the demethylase ndmA gene. 

I wish I knew enough genetics to test a theory that occurs to me: maybe my own caffeine metabolism is also affecting by a similar gut bacteria that I have in abundance but which is missing in other people. I already checked for Pseudomonas Fulva — I don’t have it in any of my uBiome samples.  I do have abundant levels of the genus Pseudomonas, but that is not the same thing.

It should be possible to screen every one of my gut bacteria demethylizing ndMA gene, but I’m not sure how to do that. If I did find the gene in one of the gut bacteria I harbor, then that would be pretty cool: a microbe that helps me drink coffee.

 

 

Overselling the microbiome

Here’s the cover of a magazine I saw yesterday in the grocery store checkout lane:

 

I thought about submitting it to Jonathan Eisen's list of Overselling the Microbiome, but sadly, I don’t think it would make the cut. These types of articles are becoming too common to deserve an award.

There is surprisingly little that is known about the microbiome, and for every study that shows some connection between this or that microbe and this or that condition, there are counter-examples galore. Some of it is a measurement issue: even if you do sample a person’s microbiome, are you sure it’s a representative sample? Studies of gut microbes are nearly always made on the organisms that exit the body, so by definition any measurement is of something that is no longer bioactive.

I’m interested in the microbiome because I think the whole subject is incredibly fascinating, and I believe that eventually science will find some deep insights that will radically alter the way we think about what it means to be human. I hold out hope that we may find a few tricks to manipulate the microbiome in some specific cases, and certainly in general it is good to learn more about how our bodies work. 

But c’mon people, tone it down a little.

I’m rich!

The uBiome blog posts the results of a large (n>500) study of their users that shows a fascinating relationship between household income and a large change in a particular bacterium, the non-pathogenic Genus 713, actinobacillus. Their number crunching shows that wealthy people are far less likely to have this organism in their guts. For the statistically-minded, it’s a p-value of 0.00001 – as near to scientific certainty as you can get. If you have it, you’re poor. If not, you’re wealthy.

Obviously upon hearing this my first step was to check my own uBiome results to see if I have any actinobacillus (which, incidentally, was first isolated from the respiratory tracts of pigs).

Answer: no!  My latest sample shows no Genus 713!  I must be rich!

But wait, I did have a smidge (under 0.0008%) in the sample I submitted last May. Fortunately it cleared up by the following month and I’ve been Genus 713-free ever since! Obviously my microbiome is responding in advance to the large jackpot it knows I’ll be receiving any day now.

Now that science has demonstrated my wealth with such certainty, I am feeling extra generous today. Instead of money, I’ll share this comic (from the always excellent XKCD) which explains why they found this unassailable correlation between wealth and the microbiome:

What potato starch really did to my gut

As suspected, there was a glitch in my latest uBiome results. Apparently a server hiccupped, so the scientists there recomputed my sample and sure enough, the new data is much more believable. Here is the single chart summary of all four of the uBiome tests results I’ve received so far:

You can see a number of changes, but what’s exciting about this one is that it’s the first sample where I was deliberately trying to test something: the effect on my gut of taking potato starch to hack my sleep.

A few details about this sample: 

1. Taken on Jan 19th, almost exactly 3 months after my Oct 17th sample.
2. During the 94 days between samples, I had 31 days where I took a dose of potato starch, a total of about 80 tablespoons.
3. I had been taking 1 T daily, an hour before bedtime, for a week before this sample.

To analyze my results, I first popped into my publicly-available uBiome utilities, using the data I had already downloaded from the site. If you want to follow along at home, here are some of the commands I typed.

To answer the question: Which new species appeared in January?

head(uBiome_sample_unique(jan,oct),topN)

##    missing.count_norm                   missing.tax_name
## 1                1257                 Arthrobacter albus
## 2                 573         Bacillus amyloliquefaciens
## 3                 530                Enorma massiliensis
## 4                 376              Ruminococcus lactaris
## 5                 188          Subdoligranulum variabile
## 6                 163        Adlercreutzia equolifaciens
## 7                 145                Oligella urethralis
## 8                 137         Clostridium sp. NML 04A032
## 9                 137 Desulfovibrio sp. oral clone BB161
## 10                111              Streptococcus rubneri

and which went extinct (are no longer in January sample?)

head(uBiome_sample_unique(oct,jan),topN)

##    missing.count_norm                                missing.tax_name
## 1                5929                          bacterium NLAE-zl-H436
## 2                 114                        Dialister micraerophilus
## 3                 101                      Peptoclostridium difficile
## 4                  88            Dehalogenimonas lykanthroporepellens
## 5                  76                     Bifidobacterium catenulatum
## 6                  51                      unidentified bacterium ZF5
## 7                  51 Veillonellaceae bacterium canine oral taxon 211
## 8                  51                           Ruminococcus sp. 25F8
## 9                  51                              Clostridium leptum
## 10                 38                      Peptoniphilus sp. gpac018A

Here’s the overall picture of what changed between Oct and Jan:

Positive numbers indicate something that is more plentiful in January than October:

tail(octVsJan,topN)

##                               tax_name count_change
## 24              Bifidobacterium longum         7713
## 39         Clostridium clostridioforme         8958
## 78                 Ruminococcus bromii        10663
## 20               Bacteroides uniformis        11036
## 23            Bifidobacterium animalis        13578
## 69 Peptostreptococcaceae bacterium TM5        14151
## 47            Coprococcus sp. DJF_CR49        19000
## 36                 Clostridium baratii        24955
## 76                Roseburia sp. 11SE38        35712
## 57        Faecalibacterium prausnitzii        89592

head(octVsJan,topN)

##                              tax_name count_change
## 17               Bacteroides plebeius       -90998
## 30 butyrate-producing bacterium A1-86       -88414
## 38          Clostridium chartatabidum       -41972
## 11             bacterium NLAE-zl-P430       -23659
## 22       Bifidobacterium adolescentis       -18334
## 10              bacterium NLAE-zl-H54        -9970
## 62              Lactobacillus rogosae        -8575
## 27                     Blautia faecis        -7693
## 56                Eubacterium siraeum        -6040
## 5               Alistipes onderdonkii        -5878

See the difference?

Let’s look at the genus level:

(again, positive numbers are more plentiful in January)

tail(octVsJan,topN)

##              tax_name count_change
## 27              Dorea         3396
## 42     Parasutterella         3473
## 48 Pseudobutyrivibrio         5523
## 34        Lachnospira         5895
## 14            Blautia         6612
## 12    Bifidobacterium         7769
## 7        Anaerostipes         8565
## 23        Coprococcus        18317
## 52          Roseburia        34957
## 29   Faecalibacterium       135557

head(octVsJan,topN)

##           tax_name count_change
## 10     Bacteroides       -87903
## 3        Alistipes       -12141
## 41 Parabacteroides        -9820
## 53    Ruminococcus        -9685
## 35   Lactobacillus        -8601
## 28     Eubacterium        -5785
## 21     Clostridium        -5212
## 11     Barnesiella        -3809
## 17   Butyricimonas        -2702
## 26       Dialister        -2410

Summary

Organism	May	Jun	Oct	Jan	Rank
Faecalibacterium prausnitzii	99571	62316	5790	95382	species
Roseburia	13554	11157	7825	42782	genus
Christensenellaceae	82585	39713	40290	40977	family
Christensenella	269	NA	38	17	genus
Akkermansia	30960	19654	7648	6269	genus
Bifidobacteria Longum	32	NA	1858	9571	species
Bifidobacterium	8473	6532	58747	66516	genus
B. Longum as % of total Bifido	0.38%	NA	3.16%	14.39%
Clostridium	35012	41679	71326	66114	genus
C. botulinum	NA	25	x	128	species
C. clostridioforme	28902	35372	15170	24128	species
C. baratii	1223	NA	5588	30543	species

The units are all uBiome’s “count_norm” field, which you can think of as, roughly, a percentage (a fraction of one million). Items in italics are “good”.

I'll have much more to say as I analyze this for a future post, but so far I'm thinking that no, potato starch didn't wreck my gut. The benefits in better sleep appear to come at little or no major cost to the rest of my gut flora. What do you think?

A mistake in my fourth uBiome sample?

[Update: uBiome recomputed my results, which are now much closer to what I expected.  I'll update with a more detailed post soon.]

After all that analysis and discussion with experts about my uBiome results, I had high expectations for the brand new set of answers that arrived today.

Here’s a comparison chart showing all four of my uBiome submissions:

In a word: argh!

If the January 19th sample had been my first and only uBiome test, I’d be tempted to read a lot into this. After all, it appears that my levels of proteobacteria are way outside the norm. That’s not all: look at some other oddities about this one:

• That bifido bloom I saw after sleep-hacking with potato starch: it’s all gone. Not a single bifidobacterium was found in this sample. Hmmm.
• Lots of prevotella (almost 3% of the sample), a species that didn’t appear in any of my previous samples, and a bit worrisome for a meat-eater like me.
• No more Clostridum, either. Commonly thought of as a pathogen, it may be good to get rid of this, but why did it disappear?

All of these massive changes in the span of only three months? Not impossible – the human gut can change pretty quickly under the right circumstances. But you’d expect something different about my environment, eating habits, and certainly my health.

But here’s the thing: I don’t notice a single difference in my health or well-being over this time period. Same sleep, same weight, same general mood. Diet, bowel movements, skin – like everyone, I see minor day-to-day variations, but absolutely nothing about me is different enough to be noteworthy.

On the other hand, there are a few oddities in the sample itself. First, uBiome warned that their first run had too low levels of bacteria; the ones you see above came after they ran the sample again under more amplified settings. Second, I used an older kit, one that had been lying around the house for about a year. Finally, I also ran into trouble with the mail, so it sat around at the post office for several more weeks than normal. Shouldn’t really matter, but still…

Soooo, my bottom line is that I’m just not going to read much into this sample. I’m waiting on my next submission, one that was sent a few weeks after this one, and hopefully that will give me a much better picture.

The takeaway for you? Don’t read much into a single uBiome test. The science is too new, and there are so many other factors that go into the results. My advice: send in multiple kits, spread over several weeks or months, before jumping to conclusions.

My uBiome third sample

Eventually it will be possible, perhaps routine, to get regular real-time updates to what’s happening among the bacteria crawling all over our insides, but for now the best a normal person like me can do is continue sending samples to the uBiome testing service.Here’s an overdue look at my latest report:

 (I discussed my previous uBiome results a few months ago)

Unfortunately, I don’t see much of interest. That last sample was taken a few months after the first two. I can’t point to anything unusual that might have happened between the tests. Nothing notable about my diet (I’m a fairly normal omnivore). No antibiotics, no bouts of sickness, no unusual travel or other changes.

Three samples is not enough to draw any conclusions, especially since they are taken in informal, non-scientific circumstances (i.e. my house, at whatever random time I feel like). You can read these however you like, but it’s interesting that the first and third samples are most similar on the two strains of microbes that seem most common in modern America. The level of firmicutes, in particular, is consistent with somebody like me who is of normal weight and health. There is speculation that firmicutes (pronounced fir MIK kyoo teez, by the way) play a role in fat absorption; too much or too little may play a role in obesity).

At 6+%, the actinobacteria in the latest sample is much higher than normal. Since those strains are more common in other parts of the body (e.g. nose, genitals), I’m wondering if there’s a little contamination going on.

I’ve got another uBiome kit waiting to be used, so stay tuned for more updates.

 

[book] Missing Microbes

Another fantastic book, by Martin Blaser, professor of microbiology at NYU: Missing Microbes: How the Overuse of Antibiotics Is Fueling Our Modern Plagues.

Microbes, a word that refers to microscopic lifeforms like bacteria, archaea, and viruses have been around way longer than anything else, and show far more variety than any visible life. If you imagine a circular clock face representing the degrees of microbiological difference between various life forms, humans and corn plants would be separated by only a single degree — the rest of the clock is an unimaginable variety of life, so different from us and inhabiting every nook and cranny on earth, from the radioactive sludge inside a dark nuclear plant to high in the atmosphere. Life is everywhere, and almost all of it is microbial. But as ubiquitous and resilient as these microbes are, many of those that matter most to humans have been under attack for the past century because of antibiotics, and the resulting changes may be the root cause of many modern afflictions, from obesity to autism to diabetes to cancer.

Here are just a few of the facts you’ll learn in this book:

• Specialized lab mice that are raised germ-free may appear outwardly normal, but their blood contains only 52 out of the 4200 compounds found in normal mice. The implication is that microbes in the gut and skin generate many thousands of chemicals — vitamins, hormones, and more-- that are important for life.
• Gut microbes produce a native compound similar to valium, normally cleared out by the liver. End-stage cancer patients often slip into a valium-induced coma when their livers fail.
• Veridans streptococci usually live harmlessly in the mouth and serve to prevent Step A infections by simply crowding out other bacteria. But when they get into the heart, they are the major cause of heart valve infections.
• The FDA doesn’t require labeling for products (like milk or organic apples) that contain less than 50mg of tetracycline. That seems like a trivial amount, with no possible affect on your health, but the dosages add up: drink milk every day and you have ingested a noticeable amount after a week.
• Your gut bacteria produce 80% of the serotonin your brain uses to remain calm and promote good sleep. There are certainly many other examples of important hormones produced, not by the body, but by microbes.

Many, many more facts and ideas, well worth reading.  Highly recommended.

 

What's in my microbiome?

Although I was an early customer of 23andme, I found the results unsatisfying because there is so little that is actionable. I mean, you don't need a test to tell you that you have green eyes or that you're lactose intolerant. And if you have some weird disease in your future, the main advice -- eat well, get exercise, buy life insurance -- applies to everyone whether you know you're at risk or not. Although I think it's fun and interesting for its own sake, I frankly understand why normal people wouldn't bother.

That's why I'm now so intrigued with the microbiome, that collection of hundreds of species of bacteria and other microbes that live all over you, inside and out. In total, they weigh about as much as your brain; it's as if your body has a whole other organ that you can't see. Massive improvements in genetic sequencing technology over the past decade have suddenly brought the ability to view and measure this micro-world, and important new discoveries are happening every day.

The best part? you can change your microbes! It can be difficult, because once these things have found a foothold somewhere in your body, they don't want to let go, but at least it's possible, whether through eating different foods or supplements (fermented products, probiotics, fiber) or by avoiding antibiotics (including germ-killing hand sanitizers), your actions have an affect your microbiome. So if you don't like something, you can (at least in theory) change it.

To find out what's in my microbiome now, I bought a $400 kit from the San Francisco company uBiome. For that, I received swabs for my mouth, gut, genitals, nose, and skin. The most interesting one is the gut, which you can buy for $89, and is the one I'd most recommend. After you collect the samples (warning for the squeamish: it involves soiled toilet paper), send them back and they give you results like this:

The microbiome is so complicated -- we're dealing with hundreds of different species, each with its own genome -- that it's hard to summarize in a single, quick takeaway. The web site lets you dive deeper into the various strains of bacteria, looking at the results organized by phylum/class/order/family/genus, digging into more detail at each level. Since each user fills out a detailed questionnaire when returning the sample, you can compare your results to self-described vegans, paleo dieters, and several other categories to see how you stack up to people who have been gaining or losing weight recently, for example.

The tools on the site are fantastic, and they are obviously putting more effort into improving them (they've become noticeably better just in the past few weeks). Better yet, you can download all of your results in XML format to keep forever, or analyze anyway you like.

My only complaint was the amount of time it took to get results: a full three months for this sample. Their customer support staff is very responsive -- I always got a helpful response within a few hours, even on weekends -- but I do wish it didn't take so long.

But frankly that complaint rings hollow when I consider how long it will take me to understand the plethora of fascinating torrent of data I got back. You can see from the chart above that I have more of certain types of bacteria than their average customer. Is that good? bad? neutral? Who knows?

This is not uBiome's fault. Science just hasn't figured out these answers yet. In my case, I'm a healthy, fit, normal-weight omnivore, so I assume that the percentage differences with "Average" are perfectly fine, but I could imagine how somebody with weight or other problems might be concerned if they saw something significantly different from average. The good news is that, in theory, you can change your results, but the bad news is that science really has little clue exactly how to do that, or even what constitutes "good".

Because the technology is so new, you may find that your snapshot looks different each time you take the test. Sometimes people report different results from different locations on the same sample. So it's best not to read much into of any findings yet. Still, it's the ability to change your microbiome that makes all the difference and I expect to learn much more as the science progresses.

My sleep genes

I’ve tracked and analyzed my sleep for years, and of course I also have a long-time interest in understanding my genome, but I’ve been unable to combine both pieces of information until now. The new issue of the journal Sleep identifies a gene that seems associated with an ability to function just fine on six hours of sleep per night.  (also see the New Yorker overview)

23andme doesn’t tell me the exact genes in my genome — only the SNPs that tell who my genes differ from a reference genome in key places — so this is not necessarily the final word about the BHLHE41 Variant identified in the journal Sleep, but here’s the next best thing:

rs4963955 TT
rs4963956 CC
rs1480037 CC

These are my SNPs in those locations, and they appear to correspond to the gene variant that means I can survive just fine on six hours of sleep.

Having said this, it’s now past my bedtime. Good night.

Personalized genetic analysis

The biggest problem with all the great new self-measuring tools is the torrent of data they produce. You could devote your life to analyzing your results and not come close to understanding what it all means.

You’d think this would be a nice business opportunity for a wave of consultants and advice-givers who will take your data and supply meaningful, actionable summaries, but the FDA’s recent clampdown on 23andme makes this harder than it needs to be. Although automated tools like Promethease make it a bit easier for the rest of us to do it ourselves,  I’m glad that a few brave companies are stepping into the void — they’re at the cutting edge of something that I think will be commonplace, even routine.

 XRGenemics will send you “the world’s most accurate fitness DNA” kit for £150 (about $250), promising results in about a month.

Genetrainer sells a lifetime for $80: give them your 23andme results and they’ll give you a personalized fitness summary, which they’ll keep up-to-date as more science becomes available.

There are also services that specialize in helping would-be parents figure out the likelihood of potential birth defects. Counsyl charges $1000, but with many insurance policies the cost can be closer to $300.

I haven’t tried these services yet — I fortunately don’t have any serious health issues that I want to analyze that deeply — but I’m glad these companies are out there, and I’m looking forward to more new ones in the future.

The FDA needs to be more Bayesian

John Wilbanks is one of my favorite open science geeks (I even met him once in person, when he did a presentation for Microsoft Research). In a blog post this week, he writes a well-reasoned explanation for why the “Evil FDA shuts down lone entrepreneur” narrative is wrong.

Tech companies like 23andme, he writes, think in a “Bayesian” way, where the safety or “truth” of a medical claim is a probabilistic concept that depends on the number of data points (i.e. users). This doesn’t sit well with the FDA, to whom Truth is a binary fact: something is either safe or unsafe, period:

That “traditional” submission to the FDA would be of a very specific kind of analysis based on randomized controlled trials. It is designed to keep bad things from happening to people, not to make sure good things happen to people.

He concludes, correctly, that this is a clash of cultures and that if 23andme wants to succeed (and he hopes they do), they need to accept reality. This is how the FDA works. They should have known that:

[U]ntil the FDA learns how to deal with Bayes’s rule and its discomforts - and until DTC companies figure out a business model that isn’t based on massive loss leadership - we’re going to keep coming back to this clash of culture and business models. Both sides need to make some changes if we’re going to avoid doing this over, and over, and over

But why must “both sides” make changes? I’m reminded of similar advice given to Chinese dissidents that they need to “work through the system”, rather than make public their often misleading and “socially irresponsible” opinions.

I am able to make up my own mind about 23andme’s “marketing” claims, and so can you. The FDA, regardless of how understandable their position, is wrong. Shouldn’t those of us who believe in open data just say so?

The argument for regulating information

When Nobel Peace prizewinner Liu Xiaobo was imprisoned by Chinese authorities, his captors were unafraid that he alone, a single individual without guns, an army, or even military training, threatens the Chinese government, which has plenty of those items to spare. And few in the government would argue that Liu himself, or his ideas, are themselves irresponsible. He’s a well-educated, perfectly sensible individual well within his rights to think the thoughts he was thinking. No harm if he had simply stopped there.

When you or I, dear reader — the educated elite users of a product like the genetic testing service from 23andme, when we use the information about our genes, few at the FDA will argue that there is a danger. After all, we’re the early adopters, the people smart enough to seek this information in the first place. The trouble is not you and me, it’s them, don’t you know, the unwashed masses out there who may become — how shall we say this delicately? — overexcited, causing themselves potentially tragic — and avoidable — harm.

History shows that the ideas Liu outlined in Charter 08, might actually help China. Reasonable people, those bearing the full responsibility for the stability and long-term future of the country, have no fear of the ideas themselves. Once the country has matured a bit more, once the people are ready for this information, then yes, it may become appropriate to discuss the issues publicly. But right now, here in the real world, where leaders with actual accountability for China’s long-term stability, know that to throw Liu’s ideas out there, wily-nilly, without the proper preparation…well, think of what could happen if those ideas landed in the hands of the irresponsible masses who might be tempted to take action without understanding, as we do, the full consequences.

You see, an expert, whether at the FDA or in the Chinese Communist Party, has been carefully vetted, with years and years of education that brings a better sensitivity to the long-term benefits, as well as the potential downsides, that come with access to powerful ideas.

The government has been very patient with Liu Xiaobo, offering years of warnings, giving him plenty of time to realize the potentially destabilizing consequences of his behavior. The FDA was similarly patient with 23andme, spelling out over dozens of meetings and countless emails, precisely what the experts fear — know — can happen when important information gets into the wrong hands.

Liu Xiaobo has no gun, but many of his potential readers do. 23andme doesn’t perform mastectomies or administer drugs, but many of their potential readers may not be so limited.

You and I may be able to handle a world without sensible regulation of ideas and information. But do you really think that others can?

Is the FDA the only source of experts?

An SFGate blog, supports the FDA’s decision to prevent informed consumers   from having our genes sequenced halt the marketing of 23andme:

Silicon Valley whines any time regulators bog down their ability to “move fast and break things.” But as a society, we’re far better off when a neutral party that isn’t in the business of selling a medical product takes steps to ensure it does more good than harm.

Nobody disagrees about the importance of neutral parties ensuring that products do more good than harm. But isn’t that precisely the purpose of SFGate, the rest of the media, and for that matter, the entire education system? We are surrounded by neutral parties giving their opinions about every product and service we buy. That’s one of the many great benefits of a free society.

The FDA, like any other group of experts, is often wrong.  What makes them so special that they, and only they, should be allowed to decide what kind of information an intelligent American is allowed to have? Can’t I make up my own mind?

The inevitability of human enhancement

I watched a TED talk by Juan Enriquez (Will our kids be a different species?) that wasn’t particularly informative if you’re up to date on biology, but it got me thinking about how quickly some of these technologies are going to affect us.


Human enhancement technologies range all the way from tattoos to adding a new chromosome, and many of these are straightforward to implement to anybody with some basic knowledge of the field.



A generation ago, computer programming was a specialist skill that was mostly practiced by well-educated people employed by large institutions. Although it’s still true that most of the mainstream computer programming happens that way, there are millions of “amateurs” with no particular specialized education or access to expensive capital equipment, and those amateurs are doing as much heavy computing as the top experts thirty years ago.  You don’t see these people much because the mainstream experts are doing so much more, and the world has simply moved on to a stage where the bar for money and more is so much higher.



The same thing is becoming true for biology. An entry-level college biology lab now exposes students to the basics of recombinant DNA, and once learned, a fairly intelligent and curious person can do it on his own, without particular access to specialized equipment. There is already a large movement of DIYBIO people who find refurbished or underutilized biological instruments that they repurpose for amateur uses, with costs a fraction of what the mainstream people pay. And as much of biology moves into computers, the costs go even lower: you can design what you want and have a third party “print” manufacture it for truly citizen prices.



The amateur in America may have some interesting access, but this pales in comparison to professionals from other countries who using the state-of-the-art knowledge available to everyone on the internet, can make seriously interesting biological products with the help of a national-scale lab.



The implication is that even if the United States or Western countries try to ban or regulate something, it will be possible for motivated people in other countries to do it anyway – and the competitive pressures will be enormous. Imagine a gene modification that makes for slightly better math performance. What responsible parent would ignore a technology like that, especially if they feel other parents are doing it?



Even if the United States tries to make that illegal, the motivation is too strong to stop it internationally. Once a Chinese lab, or company, starts offering the service, people from everywhere will travel there to get the procedure done for themselves. This will be very hard to stop.

(FlickR photo by jacob earl )

I’m a builder

I was devastated to read, in the latest Atlantic, that one of my favorite female writers, Sandra Tsing Loh, is divorcing.  In an excellent summary of several books she reads while reflecting on the topic, she includes this description of a new one by Helen Fisher.

Why Him? Why Her?: Finding Real Love By Understanding Your Personality Type'>Why Him? Why Her? explains the hormonal forces that trigger humans to be romantically attracted to some people and not to others (a phenomenon also documented in the animal world). Fisher posits that each of us gets dosed in the womb with different levels of hormones that impel us toward one of four basic personality types:

The Explorer—the libidinous, creative adventurer who acts “on the spur of the moment.” Operative neurochemical: dopamine.

The Builder—the much calmer person who has “traditional values.” The Builder also “would rather have loyal friends than interesting friends,” enjoys routines, and places a high priority on taking care of his or her possessions. Operative neurotransmitter: serotonin.

The Director—the “analytical and logical” thinker who enjoys a good argument. The Director wants to discover all the features of his or her new camera or computer. Operative hormone: testosterone.

The Negotiator—the touchy-feely communicator who imagines “both wonderful and horrible things happening” to him- or herself. Operative hormone: estrogen, then oxytocin.

Fisher reviewed personality data from 39,913 members of Chemistry.com. Explorers made up 26 percent of the sample, Builders 28.6 percent, Directors 16.3 percent, Negotiators 29.1 percent. While Explorers tend to be attracted to Explorers, and Builders tend to be attracted to Builders, Directors are attracted to Negotiators, and vice versa.

Although I love nothing more than cracking out the manuals to discover every feature of my new gadgets (and lord knows, I have a lot of testosterone)  I think I’m more of a Builder than a Director, the only other item on this list that seems to fit me.  I don’t go through friends quickly enough to say that I value “interesting” over “loyal”.  I mean, some of my best friends are people you probably think are pretty boring. 

What do you think?  Which are you?

Better stay away from cocaine

A new study reported on the 23andme blog shows a link between cocaine-induced paranoia and the SNP rs9387522.  Turns out that I have the TT variation of this gene, which correlates with increased odds of cocaine-induced paranoia.

Guess I better stay away from cocaine.

I find this to be a problem with many (most?) of the conclusions I get from my 23andme results.  So far most of the actionable consequences of my genes tell me to avoid behavior that my mother taught me long ago was bad anyway.  Although I’m glad I took the test, and I eagerly look for a reason to recommend the test to others, so far I haven’t found many examples of why I think you should pay $300 to see your results.

Steven Pinker on Personal Genomics

The Jan 11th edition of the New York Times Magazine has a cover story by my favorite thinker describing his experiences with genome testing, and he finds the same lack of satisfaction that I have.  He coins the term “Geno’s Paradox”, to describe how with genomics it seems that the more you know the less you know.  My experience with the 23andme test is that yes, I’m glad I tested myself, but what did I really learn?  Like Pinker, I find myself using my knowledge of myself to make sense of the test results, rather than the other way around.

Some interesting takeaways from the essay:

• He didn’t have the guts to test himself for Alzheimer’s.  [unlike me]
• Although he has the bitterness receptor (unlike me], he still enjoys brocolli and beer.  So what good does the gene do?
• He’s a libertarian! [like me]
• The company Counsyl specializes in pre-natal genetic testing, a good idea before you have kids.

He doesn’t use the Taleb term “narrative fallacy”, but that’s what he means when describing the need that people have to explain why they turned out the way they did—even if it’s untrue.

some good quotes:

The most prominent finding of behavioral genetics has been summarized by the psychologist Eric Turkheimer: “The nature-nurture debate is over. . . . All human behavioral traits are heritable.” By this he meant that a substantial fraction of the variation among individuals within a culture can be linked to variation in their genes. Whether you measure intelligence or personality, religiosity or political orientation, television watching or cigarette smoking, the outcome is the same. Identical twins (who share all their genes) are more similar than fraternal twins (who share half their genes that vary among people). Biological siblings (who share half those genes too) are more similar than adopted siblings (who share no more genes than do strangers). And identical twins separated at birth and raised in different adoptive homes (who share their genes but not their environments) are uncannily similar.

[this is obvious to anyone who has children]

Although Pinker is clearly doubtful about the short-term promise of genetics testing, I actually think the situation is even more complicated.  What if much of our “environment” is determined by all those genes from the bacteria inside our bodies, some of which are inherited, some of which just arrives through whatever accidents life presents us.  In that case we’d have something with a genetic component (bacterial genes) combined with an environmental one (how we picked up the bug).  How in the world would you ever be able to analyze that amount of complexity?   But like he says:

Personal genomics is here to stay... People who have grown up with the democratization of information will not tolerate paternalistic regulations that keep them from their own genomes…There are risks of misunderstandings, but there are also risks in much of the flimflam we tolerate in alternative medicine, and in the hunches and folklore that many doctors prefer to evidence-based medicine. And besides, personal genomics is just too much fun.

What’s in my genes? Results from 23andme

My DNA test results arrived much sooner than I expected. They tell you six to eight weeks, but it was more like two or three. The company that I chose for the test, 23andme, recently dropped their price to only $400, making it much more affordable. (By the way, I paid the original $1,000 price, but they generously refunded the difference--a hint that they care about good customer satisfaction).

I hope to post much more about this as I get time to analyze my data, but meanwhile enough customers are out there now that it's possible to get a sense for the direction this technology is headed. Mark Fletcher is an early user who posted his experiences, and there are several others who have even posted their entire results on line. One of my readers suggested I check out Promethease, a Windows/Mac program that will run through your results and tell you everything that's known about your SNPs. Since my test shows 500,000 SNPs, there is a lot to analyze and I'll need as much help as I can get.

So let's open the envelope, please, and tell me the results!

First, no surprises on my ancestry. Although I found out through the $99 Genographic Project's test that my grandmother is likely descended from native Americans, no such traces exist on either my mitochondrial or Y-chromosomal DNA. I am about as pure-blooded a north European as the test shows it is possible to be. For example, here's a distribution of the people who share my haplogroup, H1*, the one I got from my mother.

On my father's side, R1b1c9, the results are similar:

Note that these results are consistent with the very different conclusion from my Indian grandmother, who gave me 1/4th of my overall genes, but not the ones from this test. Remember, my mitochondrial DNA comes exclusively via my mother, just like my Y-chromosome DNA comes exclusively from my father. Since Grandma (a female) had no Y to pass down, my Y comes from my father and grandfather. Similarly, my mitochondrial DNA is 100% from my Lithuanian mother's side (as you can clearly see from the diagrams above).  All the more reason to test your own grandparents while they’re still alive.

Okay, so that’s ancestry.  What about the rest?  23andme provides some basic analysis on (as of today) about 90 specific genetic conditions, with varying degrees of scientific certainty or idle interest, depending on how well-studied these are.

In my case the following areas turned up good:

• Very low chance of Parkinson’s Disease (unlike Google co-founder Sergei Brin)
• I have the same mutation as that Jamaican Olympic sprinter!
• No Crohn’s, no Celiac, no lupus, no gout.
• No male infertility (hmmmm)
• Higher odds of living to 100
• IQ:  breast-feeding would have raised my IQ 4-5 points.  (thanks for nothing, Mom)
• Heart disease:  lower than average risk.  (14.5 out of 100 versus 17.7 out of 100 for other white males)
• Arthritis:  lower than normal (1.1/100 vs. 4.2)
• Diabetes: almost no chance whatsoever of getting Type I and lower than normal odds on Type II (13 vs. 21/100)
• Lactose tolerance:  what do you expect from a farm boy like me?

Biggest worries:  [note to insurance companies: please stop reading here]

• Stomach cancer:  I have a mutation (Rs2294008) that has been shown in Japanese people to correlate with 4x higher rates of diffuse stomach cancer
• Age-related macular degeneration:  12.5 out of 100 (vs. 7.7 in other europeans) according to some studies.

There are many more mutations covered by the test, but these are the most relevant to me.  As you can see, the results are mixed: some “good” news, some “bad”.  But look more closely and you’ll see why I’m not sure yet if this really tells me much.  For example, I have a mutation (rs1051730) associated with nicotine dependence.  Okay, guess I better not smoke.  Same thing goes with similar mutations associated with heroin, HIV, and noroviruses:  I’m at risk.  Big deal.  My mother tells me these things even without viewing my genetic results.  Did I pay $400 for this?

Of course, if some of these results had gone the other way – and I’m less susceptible to addiction, HIV, or other preventable conditions – would that make me consider taking up different behaviors?  No, of course not.  There are lots of good non-genetic reasons to avoid these things.

I also question the reliability of some of the science.  For example, one of my mutations is associated with higher-than-normal risk of becoming obese if I eat more than 30% of my calories as fat.  That’s absurd to anyone who knows me and my eating habits: I’ve been skinny since childhood, seemingly regardless of what I eat.  But if I were overweight, I’d look at the same result with a big “aha” and think I know something revealing about myself.

So my bottom line is that although this is fascinating to somebody like me who is willing to invest the time and skeptical energy to interpreting these results, I’m not sure others will benefit much yet.  A cursory glance will tell you a few things, but it will take much work and analysis to uncover whether the results are truly interesting, or whether they merely confirm your own preconceived biases about yourself.

Family reunion at the Mayflower

We brought the kids yesterday to Plymouth, Massachusetts to see the Plimouth Plantation, and the full-size replica of the Mayflower. My wife's family can trace itself back to John Howland, one of the original 104 passengers back in 1620 and the oldest surviving pilgrim (he died in his 80s, having fathered ten children). There were no Spragues on the Mayflower itself, but several of the passengers either later married Spragues or had daughters who did, so although as far as I know there is no similar well-documented genealogy on my side of the family, I'm almost certainly a direct male descendent.  Genetic testing on my grandmother last Fall shows that I might be descended from some of the native people as well.

That's less of a big deal than you think. In fact you, my dear reader, are almost certainly a direct descendent of the Mayflower passengers too. It's obvious from simple mathematics. Four hundred years have passed since the Pilgrim days. If you assume that each of your ancestors had children somewhere between age 20 and 40, then you are something like 10 or 20 generations removed from then. Each generation has a mother and father, of course, so the total number of your direct ancestors who were alive in 1620 is somewhere between 2 to the 10th and 2 to the 20th power, or anywhere from 1,000 to 1 million people. You are a direct descendent of every single one of them, so don't be fooled by your last name or what you think of as your ethnicity. 

This is true even if you think your ancestors are from China or India, or Jewish, or African. Think about it: you had as many as a million ancestors alive and breeding back in 1620. Only one supplied you with your last name, yet each of the others contributed just as much of your genetic material as he did. That's a lot of people, and a lot of years for all kinds of mixups: orphans abandoned and rescued because of war or disease, "non-paternity" events from rape or infidelity, kidnappings. All it takes is one, just one of those events and you are part of another lineage even if it's not preserved in your name or looks.

Welcome to the family!

Count me in

The new August 2008 issue of Wired describes the Personal Genome Project, an upcoming ambitious attempt to collect DNA and more from 100,000 individuals.  Of course I immediately signed up!  Funny to see that my hero, Steven Pinker, also signed up as one of the first 10.