<img loading="lazy" decoding="async" class="alignright size-full wp-image-2521" title="cornflakes" src="https://blog-api.23andme.com/wp-content/uploads/2009/01/cornflakes.jpg" alt="" width="350" height="232" srcset="https://blogcms.23andme.com/wp-content/uploads/2009/01/cornflakes.jpg 350w, https://blogcms.23andme.com/wp-content/uploads/2009/01/cornflakes-300x199.jpg 300w" sizes="auto, (max-width: 350px) 100vw, 350px" />
Pop Quiz: What do women who eat cereal for breakfast each morning have in common?
<ol style="list-style-type: upper-alpha;">
<li>They get a full day&#8217;s supply of 11 essential vitamins and minerals.</li>
<li>They enjoy better <a id="jgop" title="sex lives" href="http://news.scotsman.com/health/Porridge-boosts-your-sex-life.4854146.jp">sex lives</a> than women who don&#8217;t eat cereal for breakfast.</li>
<li>They are more likely to give birth to male children.</li>
<li>They make buying decisions based on the advice of cartoon leprechauns, silly rabbits and mustachioed mariners.</li>
</ol>
Actually, we don&#8217;t know anything about answers A, B and D. But we do know about C. When British scientists announced last year in the <a title="Abstract of Mathews et al (2008)" href="http://journals.royalsociety.org/content/w260687441pp64w5/?p=b72119df2dd049f990eb8b531c3a467c&amp;pi=1" target="_blank" rel="noopener">Proceedings of the Royal Society B</a> that their research suggested <a title="NYT Well Blog Entry" href="http://well.blogs.nytimes.com/2008/04/23/boy-or-girl-the-answer-may-depend-on-moms-eating-habits/" target="_blank" rel="noopener">eating breakfast cereal increases a woman&#8217;s chance of having a male child</a> (among other things), they may have been falling prey to a statistical error that is of constant concern in the kinds of studies that 23andMe uses to provide information about the genetic components of disease risk. This post explains the nature of that misstep, and how we go about avoiding it.
Where did the researchers go wrong? Their research looked at the consumption of 133 different foods by 749 English mothers before and during the course of their pregnancies, and found that among all of these foods, only breakfast cereal was associated strongly with having sons. Nothing wrong with that. The problem arose, a pair of statisticians <a title="Press Release Re: Young et al (2009)" href="http://www.eurekalert.org/pub_releases/2009-01/nios-prn010909.php" target="_blank" rel="noopener">writes in the current online issue</a> of the Proceedings of the Royal Society B, when the original researchers neglected to account properly for the number of opportunities they had had1 to find a significant result.
In statistics, this is called the &#8216;multiple testing&#8217; or &#8216;multiple comparisons&#8217; problem, and it&#8217;s a very important concern for us at 23andMe. It&#8217;s pretty easy to get a feel for what can go wrong: suppose you were to flip a quarter five times, each time noting whether it came up heads. The odds that you would see heads all five times are not good &#8212; they&#8217;re 31 to one, or about 3%. But if you repeated this &#8216;experiment&#8217; 133 times, you would be very likely to see all-heads at least once, and more probably three to five times. So if you see all-heads a few times, it doesn&#8217;t really make sense to regard this outcome as unusual. In fact, it would be very unlikely not to see all heads at least a few times2. It&#8217;s something like the mathematical version of Tom Petty&#8217;s wistful refrain: Even the losers get lucky sometimes.
At 23andMe, the genetic association studies underlying our Health and Traits reports are saddled with the mother of all multiple testing problems. The basic idea behind these studies is pretty simple. They look at single-letter DNA variations known as SNPs in, say, 1,000 people who have a disease. Then they look at those same SNPs in another 1,000 people who are similar to the first thousand in as many ways as possible, except that they do not have that disease. If researchers were to look at a SNP and find, for example, that 80% of the people with the disease have the AA version, and only 30% without the disease do, then they&#8217;d be justified in suspecting that they&#8217;d found something significant.
But here&#8217;s the rub. These studies don&#8217;t look at just one SNP; they typically look at half a million to a million. A million is a lot more than 133; if association studies didn&#8217;t correct for multiple testing, every one of them would be expected to find tens of thousands of falsely associated SNPs mixed in with the really associated ones. That would not do.
Fortunately, there are many ways to correct for the multiple testing problem, ranging from dead-simple to math-degree-required. The top genetics journals will accept an association study only if the analysis has corrected for multiple testing using a standard method. Since we at 23andMe will consider a paper for use in Health and Traits only if it has come from a top journal (with some exceptions), we usually don&#8217;t have to worry about whether the studies we rely on are statistically sound. And, we go a step further by insisting that any finding used in Health and Traits must have been replicated by at least two independent groups of researchers. This policy helps to guard both against possible failure of multiple testing correction and unintended error on the part of the researchers conducting the study.
So how did the breakfast cereal-boy baby study make it into print? That remains a bit of a mystery to this reporter-slash-scientist. But anyone who is dreaming of having their own baby boy will just have to rely on alternative types of <a id="s-js" title="lucky charms" href="http://pregnancy.about.com/od/boyorgirl/a/chooseaboy.htm">lucky charms</a>.
<hr />
<ol>
<li> The original paper looked at the womens&#8217; diets before pregnancy and both in early and later pregancy, and tested each food in each time interval, so they appear to have conducted 399 (133*3) separate tests. They only report a significant association between consumption of breakfast cereal and bearing sons for the &#8216;before pregnancy&#8217; and &#8216;later pregnancy&#8217; time periods, so it is unclear whether they also tested the &#8216;early pregnancy&#8217; time period. If they only tested &#8216;before&#8217; and &#8216;later&#8217;, which seems odd, then that would mean they conducted 266 separate tests. Either number of tests, 266 or 399, suffices to cast serious doubts on the soundness of their result.</li>
<li>How unlikely? We want to know: what is the chance that in 133 5-coin-flip &#8216;experiments&#8217;, we never see all-heads? Well, per experiment, given a fair coin, the probability of getting five heads in five coin-flips is 1/32 (= (1/2)^5), about 3.1%. Therefore, the probability of not seeing all-heads is one minus the chance of seeing all-heads, or 31/32, about 96.9%. This is exactly the same as the chance of seeing at least one tail in five flips. The chance, then, that you see at least one tail in each of 133 flips is (31/32)^133, which is 1.47%. If you doubled the number of experiments to 266, the chance of never seeing all-heads (ie, seeing at least one tail in each experiment) is (31/32)^266, or 0.021%. This fits with your intuition: as the number of experiments goes up, the chance of never seeing an all-heads outcome goes down, because you have increased the number of opportunities in which to see this outcome. 
</li>
</ol>

23andMe Blog

23andMe Blog Team

Cereal Killer: The Tragic Tale of a Fatal Statistical Flaw

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2501" title="431px-melisstokerijmkroniek" src="https://blog-api.23andme.com/wp-content/uploads/2009/01/431px-melisstokerijmkroniek.jpg" alt="" width="300" height="416" srcset="https://blogcms.23andme.com/wp-content/uploads/2009/01/431px-melisstokerijmkroniek.jpg 300w, https://blogcms.23andme.com/wp-content/uploads/2009/01/431px-melisstokerijmkroniek-216x300.jpg 216w" sizes="auto, (max-width: 300px) 100vw, 300px" />
At the Blog we love to hear how scientists are using our DNA to unlock the mysteries of our ancestors.   In fact, hardly a week goes by when we don’t report on the latest <a id="fbjx" title="discovery" href="../2009/01/08/one-if-by-land-two-if-by-sea-new-genetics-study-indicates-multiple-paleo-indian-migration-routes/" target="_blank" rel="noopener">discovery</a> in the field of genetic ancestry.
Occasionally, however, researchers manage to uncover some mystery of the human past using the DNA of another species. Take the stomach parasite <a id="z2u4" title="Helicobacter pylori" href="http://articles.latimes.com/2002/feb/25/science/sci-ulcer25">Helicobacter pylori</a>, for example, or the ever-popular <a id="yqcb" title="louse" href="https://www.nytimes.com/2007/03/08/science/08louse.html?fta=y">louse</a>.
Now an interdisciplinary team of scholars from North Carolina State University, led by assistant professor of English Dr. Timothy Stinson, has developed a means of using cow DNA to determine the origins of medieval manuscripts.

Traditionally, scholars rely on handwriting and subtle differences in dialect to identify the origins of medieval manuscripts, which have often proven unreliable. But there&#8217;s another clue: the parchment itself. Parchment, unlike paper, is actually made from animal skins.   From the earliest civilizations in Eygpt, Greece, and Rome, scribes often used parchment for their written records.   During the Middle Ages, the use of parchment extended to religious texts and other kinds of manuscripts.   Sheepskin, calfskin, or goatskin would be stretched thin and dried, cut into appropriate sizes, and bound together in book form.
That&#8217;s harsh treatment for a molecule as delicate as DNA. Yet when scribes prepared these strips of parchment, the DNA from the skin of the animal — be it cow, sheep, or goat, was not easily removed.   Instead, some parts of the animal’s DNA — specifically, the mitochondrial DNA, remained on the piece of parchment.   Using state-of-the-art DNA extraction technology, Stinson and his colleagues have extracted and analyzed the DNA from two 15th-century French manuscripts. Their initial goal was to test whether they could in fact retrieve any DNA at all.   If so, they hoped to learn what that DNA might tell them.
Initially, the DNA tests revealed that the pages all seemed to come from the same animal: Bos taurus, otherwise known as the domestic cow.   But when the scientists analyzed the DNA more closely, they found that the different pages of the same manuscripts each shared the same mitochondrial DNA.   Their conclusion: it’s likely that each manuscript was made of hide from the same animal (or at the very least, very closely related individuals).   This means that the pages of parchment might have been prepared around the same time, rather than pieced them together from a variety of different sources.
More than simply identifying the species of animal used in creating these pages of parchment, there are many other benefits that can be gained from DNA analysis of medieval manuscripts. Stinson hopes that genetic testing of these manuscripts can help scholars discover exactly when and where they were produced.
Soon, Stinson hopes that genetic testing of manuscripts “will also allow us to trace trade routes of parchments” throughout medieval Europe.   Traditional theories on the origins of individual manuscripts can finally be put to the test using genetic evidence, and medieval scholars would have a new tool to use when tracing the evolution of the medieval book industry.
It is clear that, just as the rise in prominence of human DNA analysis has revolutionized the fields of archaeology and anthropology, so too can the DNA analysis of medieval manuscripts revolutionize the fields of history and literature.   Dr. Stinson discusses DNA testing of medieval manuscripts at the Bibliographical Society of America in New York City on January 23.

An Unlikely Use for Mitochondrial DNA Analysis

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2489" title="stevenpinker1" src="https://blog-api.23andme.com/wp-content/uploads/2009/01/stevenpinker1.jpg" alt="" width="350" height="233" srcset="https://blogcms.23andme.com/wp-content/uploads/2009/01/stevenpinker1.jpg 350w, https://blogcms.23andme.com/wp-content/uploads/2009/01/stevenpinker1-300x200.jpg 300w" sizes="auto, (max-width: 350px) 100vw, 350px" />
For an insightful and thought-provoking essay on the present and future of personal genomics, as well as the role that 23andMe is playing in its advancement, check out Steven Pinker&#8217;s essay &#8220;<a href="https://www.nytimes.com/2009/01/11/magazine/11Genome-t.html?em" target="_blank" rel="noopener">My Genome, My Self</a>&#8221; in this week&#8217;s New York Times magazine.
Pinker, who is a 23andMe customer himself and also a participant in <a href="http://arep.med.harvard.edu/gmc/" target="_blank" rel="noopener">George Church&#8217;s</a> <a href="http://www.personalgenomes.org/" target="_blank" rel="noopener">Personal Genome Project</a>, recommends taking genetic data with a grain of salt. Because as we often point out on The Blog, genetics is not fate and the science is still evolving.
Pinker, for example, has discovered that he has a genetic variation that increases his chances of going bald. Considering the man&#8217;s trademark thick and curly tresses, that one genetic toggle is not the whole story.

Pinker recommends personal genomics if you&#8217;re curious about how the countless genetic discoveries that are being made almost daily may apply to you personally, and might be relevant to your health down the road. And, we might add, becoming part of the 23andMe community also gives you the opportunity to help advance genetic <a href="https://www.23andme.com/research/" target="_self" rel="noopener">research</a> into diseases, conditions or traits that interest you. Combined with our community features, which let you find people with whom you share elements of nature AND nurture, personal genomics has great potential to become a tool not just for utilizing scientific advancements but making new ones as well.
As Pinker makes it clear however, personal genomics is not about to supplant, or even supplement, preventive medicine any time soon. &#8220;If you want to know whether you are at risk for high cholesterol, have your cholesterol measured,&#8221; he rightly recommends. (For some other takes on the Pinker piece and the state of personal genomics, check in with our colleagues at Gene Expression and Genetic Future.)
One of the most fascinating, albeit technical, parts of the essay is Pinker&#8217;s prediction that personal genomics will have more success elucidating the genetic underpinnings of personality than intelligence. Assuming from an evolutionary standpoint that it&#8217;s better to be smarter (it also kind of depends on how you define intelligence, but we won&#8217;t dwell on that), then over the millennia natural selection has done the work of deleting any genetic variations that substantially decrease intelligence but left intact hundreds or thousands that have negligible influence. Thanks to their tiny effect, those variations will prove very difficult to identify.
Evolution appears to foster large personality differences, on the other hand, because different types of people tend to thrive as the environment changes. Consider the recent history of the world&#8217;s financial markets – it&#8217;s easy to see how both pessimists and optimists could do well, though not necessarily at the same time. And it&#8217;s also easy to see how the number of bulls and bears could oscillate as first one side, then the other, got the upper hand. If you&#8217;re smart (and lucky) enough, you can make money in the market using either approach.
Evolutionary biologists call it balancing selection when two opposing traits are maintained in this way. And geneticists have had considerable success identifying genetic variations that have been subject to balancing selection, just as they have ones that make some people more adventurous, or prone to depression.
&#8220;It’s still a messy science, with plenty of false alarms, contradictory results and tiny effects. But consumers will probably learn of genes linked to personality before they see any that are reliably connected to intelligence,&#8221; Pinker contends.

Steven Pinker on Personal Genomics

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2476" title="istock_000001076277xsmall1" src="https://blog-api.23andme.com/wp-content/uploads/2009/01/istock_000001076277xsmall1.jpg" alt="" width="300" height="199">
It seems like new discoveries about the peopling of the Americas are a dime a <a id="c6ex" title="dozen" href="../2008/05/23/peopling-of-the-americas-times-two/" target="_blank" rel="noopener">dozen</a> these days. &nbsp; Without a doubt, the journey those first Americans took from the frozen wastelands of Asia down the Pacific coast into the Americas has been an active research subject for many decades. &nbsp; Archaeologists, linguists, and now geneticists have all analyzed the data in their respective fields, and while we have seen progress in figuring out the overall timing and migration routes across the <a id="ki2q" title="Bering Strait" href="https://en.wikipedia.org/wiki/Bering_Strait" target="_blank" rel="noopener">Bering Strait</a> during the height of the Ice Age, many questions remain unanswered. &nbsp; For example, there is still disagreement over whether there was a single wave of migrations into the New World around 18,000 years ago (a scenario generally favored by geneticists), or whether there were several separate migrations, each bringing across the Bering Strait its own distinct culture and languages (more popular among linguists).
Now, an international team of <a href="http://www.eurekalert.org/pub_releases/2009-01/cp-faa010509.php" target="_blank" rel="noopener">geneticists</a> has added to the debate by trying things a bit differently. &nbsp; While the majority of genetic studies have focused on the four most common mitochondrial DNA (mtDNA) types, or &#8216;haplogroups,&#8217; among Native Americans, these authors switched things up a bit. In study published online Thursday by Current Biology, they focused instead on two of the most rare and localized mtDNA haplogroups in the New World: D4h3 and X2a.

Mitochondrial DNA is passed down from mother to child exclusively. So by comparing the mtDNA of different populations, geneticists can estimate where and when their female lines diverged from one another.
Haplogroup D4h3 is usually found along the Pacific coast of South America, while X2a has been found only in north-central North America. &nbsp; The authors sampled a total of 55 individuals who fell into either one of these two groups. &nbsp; They sequenced the entire mitochondrial genome for each, thereby adding to the considerable lack of knowledge on these haplogroups; this was the first time that anyone completely sequenced representatives of either D4h3 or X2a.
The analyses of haplogroups D4h3 and X2a revealed two distinct genetic histories. That difference suggests they may have come from separate regions of Asia and expanded in the New World in very different directions, even though they both may have arrived around the same time.
Specifically, the authors argue that that, even though it appears that both D4h3 and X2a individuals arrived in the New World at about the same time period — between 14,000 and 17,000 years ago — they took very different routes to get there. The authors argue that D4h3 individuals crossed from Asia to the Americas via a coastal route; the same path the ancestors of most people bearing the major haplogroups are believed to have taken. Then, they continued down the Pacific coast, settling in various places along the way.
However, X2a individuals seemed to have embarked on a different journey. &nbsp; Avoiding the coast entirely, this group of people traveled through a small inland corridor between two major North American ice sheets, about 15,000 years ago. &nbsp; Then, they continued into the heart of North America, settling in what is now central Canada, where their descendants still reside today.
The results presented by the authors provide — perhaps for the first time — clear evidence that at least two separate routes were used by the earliest Paleo-Indians as they left East Asia and entered the Americas. &nbsp; But, and maybe even more importantly, it shows that the migrants who took these routes may have come from two different source populations. &nbsp; We can only hope that future research on these two haplogroups can reveal where in East Asia they originated, and maybe even what made them cross the frozen landscape in the first place &#8211; by land and by sea.

One if by Land, Two if by Sea:  New Genetics Study Indicates Multiple Paleo-Indian Migration Routes

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2467" title="strait_of_gibraltar" src="https://blog-api.23andme.com/wp-content/uploads/2009/01/strait_of_gibraltar.jpg" alt="" width="300" height="310" srcset="https://blogcms.23andme.com/wp-content/uploads/2009/01/strait_of_gibraltar.jpg 300w, https://blogcms.23andme.com/wp-content/uploads/2009/01/strait_of_gibraltar-290x300.jpg 290w" sizes="auto, (max-width: 300px) 100vw, 300px" />
Less than eight nautical miles wide, the Strait of Gibraltar is an easily surmountable barrier between northwest Africa and Spain. Military invaders such as Hannibal of Carthage and the caliphs of the Islamic Empire have had no trouble traversing this sliver of sea in the process of invading Europe.
But in addition to those legendary military campaigns, countless unsung migrants have also crossed between the two continents via the so-called “Pillars of Hercules.” In fact, archaeologists have long speculated that people have been moving across the Strait — in both directions — for thousands of years.   Recently, genetics has also supported a prehistoric connection between the two continents.
Though most scientists who study North African genetics have focused their attention on connections to the Near East, recent studies have shown that the majority of North African genetic types, or haplogroups, suggest European ancestry.   The most recent evidence comes from a team of geneticists on both sides of the Mediterranean who collected samples from from more than 300 people living throughout Tunisia and analyzed their mitochondrial DNA (mtDNA). Passed directly by mothers to their offspring, mtDNA traces ancestry back thousands of years along an exclusively maternal line.

The researchers were especially interested in so-called ‘Andalusians,&#8217; who are descended from Spanish Muslims, or Moors, who were expelled from Spain in the 16th and 17th centuries after the <a id="sq8x" title="Reconquista" href="https://en.wikipedia.org/wiki/Reconquista">Reconquista</a> and <a id="db1q" title="Inquisition" href="https://en.wikipedia.org/wiki/Spanish_Inquisition">Inquisition</a>.   Because the ancestors of the Andalusians came to Spain from North Africa with the Muslim invasions of the 8th century, only to return to North Africa 900 years later, the researchers hoped their mtDNA might reveal a very interesting pattern of genetic ancestry.
After analyzing the mtDNA samples, the authors noticed some very interesting results indeed. Most of the samples belonged to the very diverse and distinctly European haplogroup H.   In some Tunisian communities nearly 45% of individuals belonged to that haplogroup.
Because haplogroup H is so diverse and widespread across Europe, the authors wanted to dig a bit deeper to find out if they could pinpoint the origin of the Tunisian haplogroup H lineages. They found that the most common sub-groups of haplogroup H in Tunisia were most commonly found in Western Europe, specifically Spain and Portugal. The sub-groups of H most often found in Eastern Europe and the Near East, on the other hand, were quite rare. Further analysis showed that people bearing these sub-H groups almost certainly entered North Africa via the Strait of Gibraltar.
The authors speculate that the migration occurred after the end of the Ice Age, perhaps 10,000 years ago. At that time, people began streaming out of the Iberian peninsula, where they had been confined for millennia by the mile-thick sheet of ice that covered northern and central Europe. These results show that — even before invading armies crossed the Strait of Gibraltar heading north, people carrying the H haplogroup passed through in the opposite direction.
But what about the Andalusians?   Does their mtDNA distinguish them from other Tunisians?   Can we see any genetic footprints of their recent expulsion from Iberia? Even after examining the mtDNA of Andalusians in fine detail, the authors were unable to distinguish them from other North African communities. It appears that the majority of the European influence on the genetics of North Africans occurred many thousands of years ago, when glaciers began to retreat and people expanded from Spain; both north into the heart of Europe, and south into Africa.

Prehistoric Connection between Iberia and North Africa

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2454" title="fatmouse" src="https://blog-api.23andme.com/wp-content/uploads/2009/01/fatmouse.jpg" alt="" width="350" height="237" srcset="https://blogcms.23andme.com/wp-content/uploads/2009/01/fatmouse.jpg 350w, https://blogcms.23andme.com/wp-content/uploads/2009/01/fatmouse-300x203.jpg 300w" sizes="auto, (max-width: 350px) 100vw, 350px" />
New research in mice suggests that people who get lucky and inherit a genetic variant that has been shown to reduce the risk of type 2 diabetes are not totally in the clear. What they eat could turn the protective variant into a risky one.
PPARG is a protein involved in fat storage. The relatively rare Pro12Ala variant of this gene (which changes the 12th amino acid of the protein from a proline to an alanine) has been associated with reduced weight gain, improved insulin sensitivity and a lower risk for type 2 diabetes in several studies. But these associations are not as strong in obese subjects. Some researchers have actually found that among obese subjects, the Pro12Ala variant is associated with increased — not decreased &#8212; weight gain.
A team of scientists from France, Finland and the United States turned to genetically engineered mice to gain insight into this conundrum. Their results, published online today in Cell Metabolism, show that animals with two copies of the Pro12Ala variant are in fact leaner than animals without the variant. They also have improved insulin sensitivity and blood lipid profiles.
But when the mice consumed a diet that provided 58% of its calories as fat, compared to the normal 10% fat diet, the benefits of the Pro12Ala variant disappeared. In fact, the Pro12Ala mice were slightly heavier than the mice without the variant, although the difference was not statistically significant. For comparison, the U.S. Department of Agriculture recommends that humans keep their fat intake between 20 and 35% of total calories; recent <a href="https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5304a3.htm" target="_blank" rel="noopener noreferrer">CDC data</a> suggest that Americans consume on average about 30 to 35% of their calories as fat.
The mice with the Pro12Ala variant also live longer than the non-variant mice, but only on the normal (not high-fat) diet. The authors note that previous research in humans had suggested that this variant might impact longevity. They suggest that further research be done to investigate the impact of drugs that inhibit PPARG on lifespan.
(23andMe customers can find out if they have the Pro12Ala variant using the Browse Raw Data feature. Each copy of the G version of rs1801282 is a copy of the Pro12Ala variant.)
The researchers suggest that diet affects the PPARG protein’s interactions with hormones and other molecules involved in metabolism. When an animal (mouse or human) eats a normal diet, the proline version of the PPARG protein is more efficient at storing fat than the alanine version. But in animals that take in more energy than they expend, the alanine version of the protein ends up being more pro-fat.
“These data … may provide avenues to better, possibly Pro12Ala genotype-dependent, treatment strategies for insulin resistance in T2DM and the metabolic syndrome,” the authors write.
The results may also help people prevent these disorders in the first place.
“Through dietary counseling, carriers could be informed that they really need to watch out for high fat in their diets,” said the study’s senior author Johan Auwerx in a statement.

Mouse Research Offers Insight on Human Diet

By Joyce Tung and Amy Kiefer
<img loading="lazy" decoding="async" class="alignright size-full wp-image-2445" title="istock_000002575226xsmall" src="https://blog-api.23andme.com/wp-content/uploads/2008/12/istock_000002575226xsmall.jpg" alt="" width="282" height="426" srcset="https://blogcms.23andme.com/wp-content/uploads/2008/12/istock_000002575226xsmall.jpg 282w, https://blogcms.23andme.com/wp-content/uploads/2008/12/istock_000002575226xsmall-199x300.jpg 199w" sizes="auto, (max-width: 282px) 100vw, 282px" />
It&#8217;s been an exciting seven months since we launched <a href="https://www.23andme.com/research/" target="_blank" rel="noopener">23andWe</a>, the arm of 23andMe that gives people an unprecedented opportunity to collaborate with us on cutting-edge genetic research. Since May, the amount of data we’ve collected has grown at a fast and furious pace. For those of us who are used to the difficult and painfully slow accumulation of data in academic research projects, this information explosion has been nothing short of amazing.
From our first baby steps with “Ten Things About You” in May, to our three latest surveys – &#8220;Health Habits,&#8221; &#8220;Where Are You From?&#8221; and &#8220;What Do You Do?&#8221; – 23andWe has undergone some serious evolution. Almost every month, we have published more surveys and developed more features to help make the survey-taking experience simpler, more interesting, and more rewarding. We want to make it easy for our customers to provide truthful, good quality data, as that is the first and most important step towards doing high quality research. A big thank you to all our survey takers–we pledge to constantly work on improving this feature so we can keep you coming back for more.
We’re starting to look at genetic associations with the traits we ask about in our surveys, and we expect to have some exciting ones to report soon. But we’ve already learned some interesting things just by looking at the survey responses themselves. For example, while a few sources suggest that a higher percentage of men are left-handed than women, our data so far suggest that once you control for age this is not the case. It seems like our society is becoming more accepting of us female lefties! We’ve also seen that handedness does indeed significantly correlate with footedness. That is, left-handers are more likely to be left-footed, and right-handers are more likely to be right-footed. Similarly, handedness significantly correlates with ocular dominance, as left-handers are more likely to be left-eye dominant, and right-handers are more likely to be right-eye dominant.
And proving mom right once and for all, we&#8217;ve found that a sweet tooth does lead to more cavities. After controlling for sex and age, you&#8217;re more likely to report having many cavities (as opposed to few or none) if you reach for either something sweet or something sweet and salty when it&#8217;s time for a snack.
How is this kind of information going to usher in the era of personalized medicine? Handedness may seem like a relatively trivial trait, but it is correlated with risk for learning disability, schizophrenia, exceptional mathematical talent and other relevant traits. Understanding the biological underpinnings of what makes us choose one hand over the other for all our most delicate tasks may help us better understand the basis for these other complex traits.
As 23andWe matures we plan to start focusing more directly on health-related traits. Look for surveys in the very near future that ask about various medical conditions whose genetics is not yet understood. By combining the information customers provide in their survey responses with data from our custom chip, we can look throughout the genome for DNA variations linked to many different traits. This method can help us find genes that no one thought would be involved with a particular condition. For example, genome-wide studies on age-related macular degeneration (a leading cause of blindness) recently surprised researchers by identifying associations with genes that make components of the innate immune system. This gave scientists a whole new pathway in which to search for treatments.
We can’t guarantee that we’ll find something useful or interesting with every analysis that we do; science is a game you have to play a lot of times in order to win. But we can guarantee that we will strive to do the best research and that we will share our findings with the scientific community. By contributing to the body of knowledge on human genetics, we believe we can help bring the dream of personalized medicine a few steps closer to reality.
And all we need you to do is take some surveys.

Joyce Tung

23andWe: The First Annual Update

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2381" title="istock_000004001127xsmall" src="https://blog-api.23andme.com/wp-content/uploads/2008/12/istock_000004001127xsmall.jpg" alt="" width="300" height="199" />
My surname – Holden – has gone through many incarnations since it originated in England nearly 700 years ago.   Letters were added, then dropped.   Some branches of my family added an extra &#8220;u&#8221; in the middle, while others changed the pronunciation entirely.   Then, when my ancestors arrived in America over 200 years ago, the name went through a whole new set of changes.   It seems my surname has been in a constant state of change since its inception.
But the story of my surname is not unique.   Millions of Americans have similar stories about ancestors who, upon arriving in the New World, actively changed their names to sound more “American.” German immigrants named Blum became Bloom, KÃ¼sters became Custers, and KÃ¿fers became Coopers. Immigrants from Italy, Sweden, France, and countless other countries underwent similar transformations.   After just a few generations, the original spelling or pronunciation was lost.
Just as our surnames have changed over the centuries, little by little, so too has our DNA.   In fact, some regions of the human genome acquire mutations in such a way that researchers can trace the changes back through time — much like <a href="https://www.23andme.com/topics/ancestry/family-tree/">tracing a surname</a> back for generations in a family tree.   And one region in particular, the Y-chromosome, happens to be passed down from father to son, the same way surnames are inherited in Western culture. That provides a wealth of opportunities for scientists from a variety of disciplines to use the Y-chromosome to unlock history&#8217;s secrets, unravel family trees, and even solve crimes.

The Genetic Legacy of the Vikings
The histories of Scandinavia and the British Isles have been entwined since Vikings from Norway and Denmark landed on the eastern coast of England in the year 792.   Successful raiding parties eventually led to settlements along the eastern half of England.   Today there are remnants of Viking settlements in this region in the form of place names, unique vocabulary, and even surnames.   Last year, <a id="e6_-" title="British geneticists" href="http://mbe.oxfordjournals.org/cgi/content/abstract/25/2/301" target="_blank" rel="noopener noreferrer">British geneticists</a> took surname information from an area formerly settled by Vikings to see if men living there today who had Scandinavian surnames also had evidence of Scandinavian (aka Viking) genetic ancestry.   They analyzed the Y-chromosomes of several hundred men, and, not surprisingly, found that those with Scandinavian surnames did indeed have Scandinavian DNA, at least on the Y-chromosome.   Similar studies of Irish men have also found a modest connection between surnames and Y-chromosome types.
Surname DNA Projects
As we <a id="e_zg" title="reported" href="../2008/11/12/building-roots-from-the-ground-up-genealogy-20/" target="_blank" rel="noopener noreferrer">reported</a> several weeks ago, the field of genealogy has been invigorated by the increasing use of genetic testing to fill in the missing branches of a person’s family tree.   Genealogists are now comparing their Y-chromosomes to those of others with the same surname, to see if a shared surname is also an indication of the shared ancestry.   Within the past few years, <a id="xr16" title="surname projects" href="http://www.dnalist.net/" target="_blank" rel="noopener noreferrer">surname DNA projects</a> have sprung up all across the world &#8211; with hundreds of genetic genealogists digging deep into their genes as they piece together their detailed family trees.
Surnames and Forensics
By far one of the most interesting applications for surname and Y-chromosome comparison is in the field of forensic science.   In 2006, British geneticists found that — for some of the more rare surnames such as Maloy or Rivis, there was a strong connection between surname and Y-chromosome haplogroup.   The authors reasoned that, if DNA were to be recovered from a crime scene, forensic investigators might be able to narrow down the possible perpetrators to a specific subset of surnames.
However, there are several limitations to this idea — namely the fact that most men in the UK have rather common surnames, such as Smith, Green, and Adams.   Men with these surnames have a wide range of Y-chromosome DNA types, so it would nearly impossible for investigators to use the Y-chromosome to locate a suspect.   However, on principle this idea has merit, and further advances along these lines may someday allow investigators to exploit the DNA-surname connection.
One final note: 23andMe customers need not worry that their data will be used in this way – our research database does not include surnames and our <a href="https://www.23andme.com/about/privacy/" target="_self" rel="noopener noreferrer">terms of service</a> do not allow us to share data with law enforcement unless we are legally compelled to. And even if such a situation did arise, we have publicly <a href="http://cyberlaw.stanford.edu/node/5922" target="_blank" rel="noopener noreferrer">committed</a> to resisting legal requests for customer data.

What’s in a Name:  Surnames and the Y-Chromosome

<img loading="lazy" decoding="async" class="alignright size-full wp-image-2390" title="lungs" src="https://blog-api.23andme.com/wp-content/uploads/2008/12/lungs.jpg" alt="" width="301" height="223" srcset="https://blogcms.23andme.com/wp-content/uploads/2008/12/lungs.jpg 350w, https://blogcms.23andme.com/wp-content/uploads/2008/12/lungs-300x223.jpg 300w" sizes="auto, (max-width: 301px) 100vw, 301px" />
When the toxins in cigarette smoke are inhaled, lung cells don’t just sit there and take it. A complex network of proteins is activated with every puff. The proteins break down, modify, or pump out as many of the hundreds of cancer-causing chemicals in tobacco smoke as possible.
But when this system fails or falls behind, toxins have a chance to damage cells. New results published online this week in Cancer show that variants in the genes for ABCB1 and ABCC1, two important pump proteins that transport a variety of toxins out of lung cells, increase the risk of lung cancer.
Haijian Wang and colleagues studied 500 lung cancer patients and 517 healthy controls from Nanjing City and the surrounding area in southeast China. They found that, overall, having specific variants in the ABCB1 gene was associated with increased odds of lung cancer.
The researchers also found a variant that increased the odds of lung cancer in women of all ages by 2.57 times and by 1.5 times in those younger than 60.  In those people with a history of cancer in a first-degree relative (parent, sibling or child), the researchers found a variant that increased the odds of lung cancer by 1.91 times.
A final correlation the authors found was for a specific type of lung cancer called adenocarcinoma. The authors say that the rates of this type of cancer have been increasing in the last few decades, as has the amount of a specific carcinogen in cigarette smoke that is dealt with by the ABCB1 protein.
“Because tobacco smoking is the leading preventable cause of cancer and the cancer-prone genotypes of these genetic components are relatively prevalent in the human population, our findings have important implications for the prevention of tobacco smoking-related cancers,” the authors write.

Latest Posts — Page 199

Cereal Killer: The Tragic Tale of a Fatal Statistical Flaw

An Unlikely Use for Mitochondrial DNA Analysis

Steven Pinker on Personal Genomics

One if by Land, Two if by Sea: New Genetics Study Indicates Multiple Paleo-Indian Migration Routes

Prehistoric Connection between Iberia and North Africa

Mouse Research Offers Insight on Human Diet

23andWe: The First Annual Update

What’s in a Name: Surnames and the Y-Chromosome

Variants in Genes for Carcinogen Transporters Linked to Lung Cancer

Stay in the know.