Well, hooray! I was going to jump onto this awesome example of flagrantly stupid creationist innumeracy, but I'd been putting it off (oh, my grading. My grading. It tears at me with talons like razors). This guy mangles recent measurements of human variation, making comments like this: "previous concepts that all humans were 99.9% alike were blown apart by the research conducted on 270 people of various races that confirmed that 2,900 genes could vary within people, making over a million combinations possible." I mean, seriously, how ignorant do you have to be to think that the possibility of variation in many genes somehow means the nucleotide sequences can't still be highly similar, or even sillier, to be impressed at the possibility of a million genetic variations in a human population of billions? Maybe in his day job this propagandists sets ransom demands for Dr. Evil.
Fortunately, untangling mathematical misconceptions is Mark Chu-Carroll's destiny in life, and he polished this one off today. Go read that. I'm going to read a few more student essays.
- Log in to post comments
What is unclear to me is how this could possibly be taken as evidence against evolution. It seems to me that creationists have taken to a style of propaganda where they take any new discovery about the genome, and remark, "aha! This disproves evolution," without the least effort to indicate just how it might do so.
Actually, wouldn't more variation than previously thought between humans be evidence for evolution (or at least consistent with evolution)? After all, natural selection needs variation to work upon. At the very least, this observation is certainly not evidence against evolution.
I agree with Orac and Russell on not knowing how this disproves evolution. Also this muddies the "Man was created in God's image" argument if there are supposedly so many variations in the human genome.
Someone correct me if I'm wrong, here, but doesn't the original article's claim of "over a million combinations possible" translate pretty directly into "over a million possible different people"? Doesn't that conflict rather obviously with the number of (presumably genetically unique) people in the world?
Another classic "Science is DOOMED!!!111!!one" press release from an irrelevant god-bothering organization...
Scientists are excited about this discovery, which they say is the most revealing since Gregor Mendel's initial work with the genetic code in the 1860's.
Mendel worked with the genetic code 90 years before it was discovered? AMAZING!
...
...
If there are more than a million different human genetic combinations possible, this means that, on average, each of us has upwards of 6,000 identical twins out there, at various ages.
I was imagining an unbeatable criminal defense in this, but suddenly I saw the downside: a huge mob of Paris Hiltons, roaming the streets and SINGING.
...
...
It beats having a huge mob of Britney Spears roaming the streets and singing.
Hank, Stanton - I think you both forget the horror that is Peter Andre and Jordan.
I remember when this CNP information started to be uncovered a few years back during allelic copy number analysis in cancer tissues. At the time I thought that this will be cited as a classic example of positive proof for Darwinian for years to come. It basically shows that gene duplications - probably the major driving force in the evolution of complex genomes are frequent events. Im rather surprised that CNP information is not used to counter creationist claims that mutations are always destructive since gene duplications not only can be neutral or even positive in effect but also provide the material for future evolution of entirely new genes - while at the same time preserving the original functional gene.
OMFG, a mob of Liberaces playing Nocturn No. 2 In E-Flat Major! Run for your lives!
Well, oddly enough, genetics as a science does predate the discovery of DNA, and Mendel did study it in the sense that his work basically founded it. They just didn't know what genes really were until DNA was described.
Martin: That's pretty much what I said on Mark's blog. It's ridiculously obvious that...
Yup.
There's long been a debate about whether creationists are as dumb as they appear or they're just saying these ignorant things because they know they can get away with it. I should think that this settles that debate. It's clearly the former.
Also this muddies the "Man was created in God's image" argument if there are supposedly so many variations in the human genome.
But the god in the Christian bible seemed to have a very bad case of multiple personality disorder . . .
Actually, if there are 2900 different loci of genetic variation in humans, and each of them has just 2 alleles (leaving aside for the moment the complications introduced by the sex chromosomes), that makes the number of possible human genotypes 2^2900 or about 10^873. More if any of the variable genes have more than 2 alleles.
This can be described as "more than a million" about the same way the universe can be described as "larger than a proton". Although in fact the universe is nowhere near 10^873 protons in mass, or even in volume.
Oh, wait - that's *haploid* genotypes. For diploids (still with 2 alleles per locus) it's 3^2900, or 4.5x10^1383.
Sex chromosomes complicate the calculation because males *aren't* diploid for some genes, and females don't have some genes at all (SRY and others carried on the non-crossover portion of the Y), but I think it would still come out in the same general ballpark - give or take a hundred orders of magnitude.
If there's one thing that can reliably be counted on to produce big numbers that make astronomically big numbers look puny, it's combinatorics.
The thing that still confuses me (and maybe someone can clear up) is the difference between these two statements:
I share 98% (or so) of my genes (or is it DNA?) with a chimp and,
I share 50% of my genes with my brother (of father).
What is the discrepancy and how do these two percentages relate to the number of base-pairs I share with my brother vs a chimp?
oh man that's going to kill me for a few hours.
The first statistic refers to all genes, most of which cannot be functionally altered or the organism cannot survive. The second refers to the genes that can vary between human beings - on average, siblings will share about half of the genes that can vary, and you get half (depending on what gender you are) of your genetic material from each parent.
There are, unfortunately, a few misconceptions in Mark's post, which I think should be cleared up:
First, the figure 99.9% most probably refers to the density of polymorphism in the human genome - there's a number of ways of measuring this, the most reasonable of which is average heterozygosity, that is, the fraction of positions that will be different, on average, in the genomes of two random individuals in the population. This number is 0.0004, meaning the true figure above should be 99.96%.
Second, the article the creationist in question was referring to deals with copy number variation, which means that there are some gene duplications that are segregating in the human population - that is, if I have one copy of gene Q on chromosome 4, you might be carrying a duplication of gene Q, and have two near-identical (probably adjacent) copies of gene Q on chromosome 4. The *number* of genes falling within detected copy number variant regions is 2,900 (some regions span more than one gene - the total number of detected regions was 1,447), covering in total some 360 megabases (12%) of the genome. The original Nature paper is here.
This is a pretty substantial result, but we should be clear on what it means: a gene duplication is still a single event; it doesn't imply that estimates of human population variation were wildly wrong. It's comparing apples and oranges to say that a copy number variant represents 10K base-pairs of difference between two humans - it's still the product of a single event and should be considered accordingly. In any event, single-nucleotide polymorphisms and point-mutations between species are still a much simpler and more consistent way of looking at evolutionary histories.
As to the guy who asked about 50% above, that 50% represents the number of chromosomes, on average, you share with your sibling - about half the ones you get from your dad and half the ones you get from your mom will be in common. However, it's important to note that even two "different" homologous chromosomes are not really that different in composition - they're largely identical. But not *exactly* identical, which is what that 50% number is referring to.
Well, oddly enough, genetics as a science does predate the discovery of DNA, and Mendel did study it in the sense that his work basically founded it.
Certainly Mendel worked on genetics, and was the first to derive an understanding of certain modes of heredity. He did no work, initial or otherwise, on the genetic code, having absolutely no knowledge of DNA, RNA, codons, or base pairs. It's just another passage that demonstrates the abysmal understanding many creationists have of science -- they seem vaguely aware of the jargon, but have little sense of what the jargon actually means. They want to sound intelligent, however, so they go on using words they don't understand. Maybe it impresses other creationists, but they just turn out looking like idiots to people who do understand.
----There's long been a debate about whether creationists are as dumb as they appear or they're just saying these ignorant things because they know they can get away with it----
They Are Cuturally Conditioned By Their Parents.
They are not smart enough to see through the fasade of ignorance.
Saurabh: that's what you call "clearing [something] up?" Remember that you're most likely talking to a layman (if you're specifically responding to Mark).
Well, I *did* try. Not sure what I actually accomplished, though.
There are some very important distinctions that are sometimes lost in everyday language. There is a difference between a gene and a gene sequence - 'gene' refers to a section of DNA that codes for some biochemical trait, and the sequence refers to the nature of that trait. A blue-eyed person and a brown-eyed person will have different gene sequences, but they'll both have sections of DNA that affect eye color. These different sequences are called 'alleles' - they're variations in the gene that can give rise to different traits.
Creationist Innumeracy is deplorable but even reputable scientific magazines don't always do their fact checking very well. Here is a case in point.
I know this is a bit off topic but illustrates the fact that we should all be a bit more careful with numbers...
According to data available at http://www.mongabay.com/brazil.html between 1988-2006
60-70 percent of total deforestation in the Amazon could be attibuted to cattle ranches.
This is not exactly the same as what is claimed in an article entiled "FAO report creates a stink over farm animals" that can be found in the Science News section posted today at Scientific American.com Reprinted from Reuters.
http://www.sciam.com/article.cfm?chanID=sa003&articleID=673DE96B04FA3AD…
Here is an excerpt:
"Besides the threat to the climate, the growth of livestock farming had added to water pollution and the reduction of forests to make way for grazing. About 70 percent of Amazonian forests had been turned into grazing land, it said."
I beleive the actual total amount of rain forest that has been lost is closer to 25%, not an insignificant amount, but certainly nowhere near the 70% stated in the article.
Now if I hadn't personally been to the Amazon a few times I might have been tempted to beleive the 70% figure. I read it in Scientific American after all so it must have been checked, right?
saurabh: I believe Mark's intent was to take the creationist misinterpretation on face value and show that their claim, even though it was based on intentionally misunderstood and distorted premises, is still wrong when you work out the math. ;)
saurabh:Thanks for clearing things up?
Joshua: Thanks for the compliment, but I am neither devious nor clever enough to have intended what you said. You are right though in that maybe my intent was that even though I sort of knew the answer, someone more devious than I could exploit such a glaring contadiction.
In summary:
Me and chimp - ~98% (of base pairs)
Me and average human - 99.96% of base pairs)
Me and my brother - 99.9(99?)6% (of base pairs)
Me and my brother - 50% (of chromosomes)
Correct - and in addition, you and your brother share 50% of your alleles, on average.
Darn you, 386sx! I had a moment of beautiful hope that your link was indeed to a youtube video of a group of Liberace impersonators playing the nocturne in unison! Now I'm sad.