Gene Expression Watch? Overcounter thinks he can carve out a niche in the blogosphere by becoming "Gene Expression Watch". He is of course welcome to try. I welcome constructive criticism, but Mr. McCrory's latest salvo is not constructive. In response to my post on testosterone's effects on behavior, he says:

Still, that doesn't change the fact that gc promised us that there is "substantial biological evidence that criminality is partly influenced by genetic factors." We are still waiting for that evidence, since what he has given us is evidence that "aggressive behaviour and other feelings of hostility" are related to "testosterone levels." The two are not the same. Here's some advise, gc. Learn to operationalize your concepts like good scientists do, so we can avoid these annoying little confusions.

I was a bit piqued by his implication that I didn't know what I was talking about, so I sent him the following email:

Dear Jeff, If you want me to engage in a conversation with you, I would prefer that you refrain from personal insults or implications that we're bad scientists/hateful racists/etc. If you have an open mind and are willing to listen, I don't mind talking to you. If you're going to act like Atrios, whose criticism consists of assertions, e.g "Well, actually, no. Not really. In fact - not at all. Not even close.", then there's no point in having a discussion. Compare the comments in these two posts and tell me if you find us unreasonable. Basically, I think comments like "Learn to operationalize your concepts like good scientists do, so we can avoid these annoying little confusions" or insinuations that we are "zog-hating loonies" are unwarranted. Ok. Now on to your specific contention - that the biochemistry of testosterone tells us nothing about genetics. The connection between biochemistry and genetics is pretty obvious to a biologist, but I don't mind spelling it out. I do, however, mind being called a "bad scientist" because I don't give a disquisition on how genes exert their effects in every email. I don't know how much you know about biology, biochemistry or genetics. But let me give you a really quick overview of what's relevant here:

• Some segments of DNA (not necessarily contiguous) are genes.
• Some genes contain the information for protein synthesis.
• Some proteins are enzymes, molecules that catalyze chemical reactions.
• Many compounds essential to life can only be produced through the action of these enzymes.

Now, compounds like testosterone are important information carrying molecules with copy numbers under tight genetic regulation. Testosterone is part of a subclass of molecules known as "steroid hormones", which are information carrying molecules synthesized from a cholesterol backbone. The different steps in the synthesis process are mediated by enzymes, and the information to produce these enzymes comes from genes. If you damage or delete genes that make enzymes involved in testosterone synthesis, you get phenotypic irregularities because the synthesis cannot be completed. Here's a quick schematic of the testosterone synthesis pathway: cholesterol -> pregnenolone -> progesterone -> 17-alpha-hydroxyprogesterone -> androstenedione -> testosterone Each arrow here represents a chemical reaction that only takes place in the presence of an enzyme. These reactions can be shortcircuited, sped up, or slowed if you alter the section of DNA that produces the enzyme that controls the conversion. Two quick examples: 1) 3-beta-hydroxysteroid dehydrogenase deficiency This alteration of DNA means that the pregnenolone -> progesterone conversion is stopped. This genetic disease is accompanied by salt excretion in the urine and early death. 2) 17-alpha-hydroxylase deficiency This alteration of DNA means that the progesterone -> 17-alpha-hydroxyprogesterone conversion is stopped. This genetic disease prevents the synthesis of sex hormones (e.g. testosterone) and leads indirectly to hypertension. The afflicted patient is phenotypically female but unable to mature. The above were two examples of alterations in genes that eliminated the function of the corresponding enzyme and shortcircuited the testosterone synthesis pathway, causing substantial phenotypic consequences. Genetic alterations do not need to be major to have major effects- often they are due to single base pair alterations (copying errors) in the parental sperm & egg. Now, not all DNA is composed of genes that code for proteins. Some DNA is composed of *control elements* that determine the time that a gene is turned on and the quantity of protein that is made. These control elements read in information from the cell and determine when to turn genes on. Combinations of these control elements form genetic circuits - one of the hottest research areas right now in biology. Generally speaking, the control structure of even a simple genetic circuit is quite complicated, but there are some universal properties. For one thing, information carrying molecules like testosterone are always under feedback control; the levels of testosterone themself are fed back to the DNA control elements, which use them to determine whether more testosterone synthesis enzymes should be produced. The reason for this is that information carrying molecules can wreak a lot of havoc if there's too much or too little. [1] There's usually a tight time-space interval in which these molecules are supposed to exert their effects. Ok, so what's the point? The point is that the *levels of testosterone* (meaning copy number of testosterone molecules along with time & place of synthesis) are under tight genetic control. When you combine this with the behavioral effects of testosterone, it's clear to the geneticist that behavior itself is genetically controlled. If you want, I can go into oodles of detail on the genetic control of testosterone levels, but I'll mainly be pointing you to references. The upshot is this: biochemistry and genetics are tightly related , and it is not at all unwarranted for me to claim that testosterone levels are genetically controlled.

[1] Many other things are under feedback control, but major information carriers are typically subject to the most complex and tight regulation. Addendum: I just noticed that Atrios made a little remark that Overcounter saw fit to post:

[Debating GC is] like playing chess with a 4 year old who isn't actually aware of the rules and is armed with 700 extra pieces.

Of course, Atrios' quip is more revealing than he imagines. I'm not constrained by his PC rules, and I have a comparative surfeit of knowledge...