More Recent Comments

Friday, May 25, 2007

SCIENCE Questions: Why Do Humans Have So Few Genes?

"Why Do Humans Have So Few Genes?" is one of the top 25 questions from the 125th anniversary issue of Science magazine [Science, July 1, 2005]. The complete reference is ...
Pennisi, Elizabeth (2005) Why Do Humans Have So Few Genes? Science 309: 80. [Text] [PDF]
Elizabeth Pennisi is a news writer for Science magazine. She has been publishing articles there for at least ten years. She had previously written about genes and genomes, including earlier articles about the number of genes in the human genome.

Pennisi begins with the usual mythology about how surprised scientist were to discover that humans had fewer than 30,000 genes [see Facts and Myths Concerning the Historical Estimates of the Number of Genes in the Human Genome]. She continues by using most of the standard excuses for the Deflated Ego Problem [The Deflated Ego Problem].
That big surprise reinforced a growing realization among geneticists: Our genomes and those of other mammals are far more flexible and complicated than they once seemed. The old notion of one gene/one protein has gone by the board: It is now clear that many genes can make more than one protein. Regulatory proteins, RNA, noncoding bits of DNA, even chemical and structural alterations of the genome itself control how, where, and when genes are expressed. Figuring out how all these elements work together to choreograph gene expression is one of the central challenges facing biologists. [Numbers 1,2,5,6,7]
It's downhill from then on. Pennisi goes on to briefly describe the leading contenders for solving the imaginary problem. Not once does she mention that these have all been challenged in the scientific literature and not once does she mention that they are not specific to humans even though they must be if they're going to get you out of the pickle.

Is this one of the "right" questions that I talked about earlier? [SCIENCE Questions: Asking the Right Question] Nope. Not even close. In fact, it's a very "wrong" question that reflects an ignorance of the scientific literature and a profound misunderstanding of evolution, developmental biology, and gene expression. Humans have exactly the number of genes that we expect. They don't need to have many more genes than fruit flies or worms because a small number of unique genes are all that's required to make significant differences in development. They don't need to have special complexity mechanisms to "explain" anything because there's nothing that needs explaining. Human genes are fundamantally the same as those in Drosophila melanogaster (fruit fly), Caenorhabditis elegans (nematode worm), and Arabidopsis thaliana (a small flowering plant).

This "top 25 question" illustrates exactly the problem that I alluded to earlier. You don't recognize the important questions in science by polling science writers and editors. The "right" questions are the ones being asked on the frontiers by the creative experts who are thinking outside the box. This is an "inside the box" question and very few of those ever turn out to be important.

The right question would have been "Why Were You Surprised?"

10 comments :

Carl said...

Maybe it's my bias as a science writer myself, but I don't understand why it's so bad for science writers to be involved in forming these questions. Elizabeth Pennisi is a very experienced science writer, and in the course of her work she has been talking to lots of scientists about what is known and unknown in their fields. The alternative you propose, if I understand correctly, is to find that one prophetic scientist who actually understands the true questions of a given field. How exactly would that one scientist be selected, particularly since scientists often have fiercely held, clashing views on such issues?

I should also mention that I wrote another one of these question articles (read here), but the question was presented to me by the editors. Still, I thought it was a good question. I'll wait to see your reaction...

Larry Moran said...

Carl Zimmer asks,

Maybe it's my bias as a science writer myself, but I don't understand why it's so bad for science writers to be involved in forming these questions.

Well, I've done two questions so far. What do you think of them?

As far as I'm concerned those two questions would never have made it past a group of biologists sitting around a table. Don't you agree?

We'll get to your question. It's clearly one of the legitimate top 25 questions. Do you recall that I asked you about these question two years ago because I was curious about where they came from? I thought that the writers had selected their own questions but it turns out that many of them were just assigned by the editors.

PZ Myers said...

To be fair to the science writers, there are a lot of biologists who would let such questions stand. I just panned a poor old paper from James Valentine that was an attempt to rationalize a different metric, cell type number, that put humans at the top of the heap. And Valentine is a sharp guy, usually.

SPARC said...

I do appreciate the availability of the human genome sequence, but I guess one reason for the persistence of the question about human gene numbers resides in the craving for recognition of some researches pretending that working on the human genome is something special and of utmost importance. This may have helped to launch the HPG. However, sequencing other species first would have been more reasonable. E.g., sequences of mice are of much greater value because one can investigate gene function in them.
Another point may be that many of the sequencing guys come from microbiology, molecular biology, biochemistry, physics and engineering and have only little knowledge in genetics, developmental and evolutionary biology. Thus, they may have been unaware that gene numbers have been estimated by genetic means 20 years earlier.
In addition, one may ask if philosophy behind mega-projects like GPH or random mutagenesis approaches in mice or other organisms is really scientific since because they are not hypothesis driven. Still, they require other legitimations like the gene number question which is rather a guess than a hypothesis, even if it was presented with the authority of a Nobel laureate.
Finally genome sequencing payed off but it only did and still does because hypothesis driven research asks the right questions.

Larry Moran said...

PZ says,

To be fair to the science writers, there are a lot of biologists who would let such questions stand. I just panned a poor old paper from James Valentine that was an attempt to rationalize a different metric, cell type number, that put humans at the top of the heap. And Valentine is a sharp guy, usually.

I agree, and it's a serious problem in my opinion. When I said "a group of biologists sitting around a table" I meant a particular group of biologists who have thought about biology. And I was trying to indicate that the more biologists you ask, the more likely it is that there will be some skeptics who can think independently.

We're about to put it to the test 'cause I'm going to ask the blogging biologists to come up with their own list of the top ten questions in biology. Let's see if we can do a better job than the editors of Science.

Larry Moran said...

Martin says,

Finally genome sequencing payed off but it only did and still does because hypothesis driven research asks the right questions.

There's value in data collecting and the investment in genome sequencing is well worth it as far as I'm concerned. I don't think all science has to be hypothesis driven.

However, there is a problem in modern science, as you pointed out. While not all science has to be hypothesis driven, it turns out that very little of modern science is hypothesis driven in any meaningful sense of the term. Most of what's published today is "me too" science designed to add a little bit more evidence to the prevailing paradigms. This sort of science masquerades as genuine hypothesis driven science and that's a problem.

It's too risky to work on experiments that attempt to falsify the current models and it's too much of a career gamble to invest time and effort in a new hypothesis.

T Ryan Gregory said...

Like you, I disagree with people who desperately wish for humans to be the top ranked organism on some pseudo-objective scale and who make ad hoc attempts to salvage that notion in the face of genomic information. There is some middle ground, though, which is that 20,000-25,000 human genes is initially surprising, but that it simply means we need to do more work to sort out how phenotypic complexity is generated in general. Thus, I actually don't have a problem with Pennisi's question, given that it is merely aiming to take a post-genomic look at pre-genomic assumptions about genetics.

Emil Nilsson said...

Interesting! Now, I wonder what you make of this text (http://www.sciencemag.org/cgi/content/summary/325/5945/1194) published in Science by the same author. Personally, I am confused by the bold statements made by the advocators of epigenetical inheritance. It would be interesting to hear what you think!

Anonymous said...

In response to Nilsson on Kammerer:

How is the midwife toad thing not an example of genetic assimilation?

I don't get why that's so controversial.

Amy Blackwelder said...

I would have thought other lifeforms would have had more genes than humans -such as the chameleon ...it can change colors and blend into the environment. Seems like something much more complex then we can do. So, I wasn't surprised.