Produced by design

In the brilliant and disturbing 1997 movie Gattaca, set in the "not very distant future," it has become possible to genetically screen for... just about everything.  So most couples conceive using in vitro fertilization followed by pre-implantation genetic diagnosis (PGD), to screen out susceptibilities to common diseases and to screen in high IQ, a tendency to physical strength, good looks, and emotional stability.

The movie is about a young man, Vincent (played movingly by Ethan Hawke) whose dream of becoming an astronaut is thwarted by the fact that he is an "in-valid" -- conceived the old-fashioned way, and thus a mix of varying traits from his parents, both good and bad.  While there is an official policy of non-discrimination on the basis of genes, the hiring requirement of a urine test for drug screening allows potential employers to find out what you're made of -- and to find a pretext not to hire you if it turns out that you have a potential toward early heart disease or higher aggression.  When everything about you -- your fitness, intellect, personality, even your suitability as a mate -- can be discerned from a fingerprick, a cheek swab, or a stray hair, there is no longer such a thing as "private information."

I won't tell you further about it because it's well worth watching, and Vincent's struggle against being held back by the genetic cards he was dealt is deeply inspiring.  But it's interesting to look at whether the "not very distant future" has panned out in the twenty-two years since the movie was made.

Like much of the cutting edge of science, pre-implantation screening hasn't progressed quite as fast as people expected.  The conditions we can reliably screen out are fairly few in number, and the idea of screening in highly complex traits such as IQ or personality or appearance has turned out to be elusive at best.  Our potential to use genetic information even to predict something a good bit simpler -- adult height -- was dealt a significant blow by a paper that came out in Cell last week, called, "Screening Human Embryos for Polygenic Traits Has Limited Utility," by a team led by Ehud Karavani of the Hebrew University of Jerusalem.

What they did is pretty impressive.  Using aggregate data from no less than 700,000 people, they came up with a protocol for generating a genetic "score" that seemed to correlate with adult height, and then tried to see if applying the score to individuals from 28 different families who were not part of the original study worked the other way -- if the score reliably predicted adult height.

To say that the results were disappointing is a bit of an understatement.  It predicted the tallest person in the family in only seven of the families -- a hit-rate of 25%.  Worse still, five of the people it identified as tallest turned out to be shorter than average for the family.  The method would almost certainly be even less accurate if you tried to apply it to a trait that shows greater variability and (likely) less overall dependence on genetics -- such as IQ.

"There’s still a great deal of variability that’s not accounted for by the genes they were analyzing or simulating in their polygenic score," said Susanna Haga, geneticist at Duke University, who was not part of Karavani's team.  "Therefore, you’re still going to see a wide distribution of height or IQ points."

While I'm fascinated with genetics and all for learning more about how our DNA works, I'm not entirely sure this is a bad thing.  Pre-implantation genetic diagnosis to allow a couple who are carriers of (say) cystic fibrosis to conceive a healthy child is one thing; screening an embryo to try to boost its eventual intelligence crosses a line.  It's edging way too close to eugenics for my comfort, and in any case, we haven't even been able to come up with a consistent definition for what we mean by intelligence, so the idea that we'd screen embryos on that basis is dodgy right from the start.

So while the results of the Karavani study are a setback to our understanding of how DNA affects our development, in one way it's actually kind of a relief.  We humans don't exactly have a stellar track record for applying scientific discoveries in a positive or humane fashion, and the number of ways this kind of information could be misused is astronomical.  I know of people who are passionately interested in genealogy but have refused to sign up for Ancestry or 23 & Me genetic testing because they're afraid of what might be done with the data, and while I can understand the concern, it's nice to know that the ability to surreptitiously abstract from our DNA information about our appearance, personality, intelligence, and aptitudes is not possible...

... yet.

*******************************

Long-time readers of Skeptophilia have probably read enough of my rants about creationism and the other flavors of evolution-denial that they're sick unto death of the subject, but if you're up for one more excursion into this, I have a book that is a must-read.

British evolutionary biologist Richard Dawkins has made a name for himself both as an outspoken atheist and as a champion for the evolutionary model, and it is in this latter capacity that he wrote the brilliant The Greatest Show on Earth.  Here, he presents the evidence for evolution in lucid prose easily accessible to the layperson, and one by one demolishes the "arguments" (if you can dignify them by that name) that you find in places like the infamous Answers in Genesis.

If you're someone who wants more ammunition for your own defense of the topic, or you want to find out why the scientists believe all that stuff about natural selection, or you're a creationist yourself and (to your credit) want to find out what the other side is saying, this book is about the best introduction to the logic of the evolutionary model I've ever read.  My focus in biology was evolution and population genetics, so you'd think all this stuff would be old hat to me, but I found something new to savor on virtually every page.  I cannot recommend this book highly enough!

[Note: if you purchase this book using the image/link below, part of the proceeds goes to support Skeptophilia!]

The master cookbook

Every year, I look forward to teaching my biology classes the basics of molecular genetics, because it's just so cool.

All organisms on the Earth contain a master recipe book -- DNA -- that contains all of the instructions necessary to create them.  Each of those recipes is deciphered, through a pair of processes called transcription and translation; the first produces a temporary copy of a single recipe (called RNA), and the second takes that RNA and uses it to build a protein of some sort.  So to extend the analogy of DNA-as-cookbook; transcription would be photocopying a single recipe, and translation would be reading that recipe and using it to make lasagna.  (The lasagna, if you don't mind my stretching the analogy to the snapping point, would be the protein.)

The problem is, as with most things in life, it's not quite that simple.

Your DNA contains a lot more than recipes used in a straightforward fashion to the building of a protein.  Between twenty and seventy percent of your DNA -- depending on whom you believe -- is junk DNA, which are essentially evolutionary leftovers.  Genes that got damaged, lost promoters (promoters are, more or less, universal "on" switches), or were moved somewhere in the genome that they couldn't be activated.  Some researchers think that junk DNA provides a sort of backhanded benefit; it gives us a larger target for mutations.  Mutations in the junk DNA have no effect, so it makes it less likely that any given mutation will kill us.

But there are other complications, too.  Some DNA (called "non-coding DNA") doesn't actually produce proteins directly, but acts to control the activity of other genes -- so it's pretty critical even though it's not specifically making your lasagna for you.  Some of these are "riboswitches" -- bits of DNA that are transcribed into RNA, but the RNA then binds to other pieces of RNA and alters the rate at which they're translated.  Another example are the telomeres, which form the ends of the chromosomes and act to protect them from degradation -- the decreasing size of telomeres is thought to play a role in aging.  A third, more mysterious example are the VNTR (variable number tandem repeat) regions, which are regions made of the same pattern of bases repeated over and over -- it's been made useful in the technique of DNA fingerprinting in forensics, but their function in the living organism is unknown.

With all of this complexity, it's been an ongoing source of contention as to exactly how many genes we have.  As you can see from the admittedly brief description I've given, it's not completely clear whether something is a functional gene in the first place, so how could you hope for an accurate count?  Estimates have run up to 6.7 million genes in the human genome -- and it certainly seems like something as sophisticated as we are must surely be the product of a huge number of individual instructions.

But the more people have looked into it -- starting with the Human Genome Project in the 1990s -- the smaller the estimate has become.  Just last week, the most recent revision was released, and it's pretty startling; a team led by Steven Salzberg at Johns Hopkins University has come up with a tally of 21,306 coding genes (ones that directly produce proteins) and 21,856 non-coding genes (bits of DNA that act to control the expression of other genes).

Which, considering that we're made up of trillions of cells interacting in countless different ways, is really a pretty small number when you come to think about it.

Salzberg is up front that these estimates could still be revised.  He, and study co-author Mihaela Petrea, write:

We aligned all human genes from NCBI's RefSeq database to the Ensembl gene set in an attempt to explain the differences, but although the total counts differ by less than 300, there are several thousand genes in each set that do not map cleanly onto the other, many of them representing genes of unknown function.  Our personal best guess for the total number of human genes is 22,333, which corresponds to the current gene total at NCBI.  We prefer this to the slightly higher Ensembl gene count both because the NCBI annotation is slightly more conservative, and because recent compelling arguments support an even lower gene total.  This number could easily shrink or grow by 1,000 genes in the near future.  However, recent analyses make it clear that even if we agree on a complete list of human genes, any particular individual might be missing some of the genes in that list.  The genome sequence is complete enough now (although it is not yet finished) that few new genes are likely to be discovered in the gaps, but it seems likely that more genes remain to be discovered by sequencing more individuals.  Additional discoveries are likely to make our best estimates for this basic fact about the human genome continue to move up and down for many years to come.

So the exact count of recipes in our DNA cookbook is still a matter of contention, but the whole thing is fascinating -- to think that such a (relatively) small number of sets of instructions could produce something as complex as we are.  As for me, this whole discussion has left me hungry, for some reason.

I think I'm going to make some lasagna.

***************************************

This week's recommended read is Wait, What? And Life's Other Essential Questions by James E. Ryan.  Ryan frames the whole of critical thinking in a fascinating way.  He says we can avoid most of the pitfalls in logic by asking five questions: "What?"  "I wonder..." "Couldn't we at least...?" "How can I help?" and "What truly matters?"  Along the way, he considers examples from history, politics, and science, and encourages you to think about the deep issues -- and not to take anything for granted.