Thursday, June 5, 2008

Small events, big pictures

One of the hidden dangers in the way much of biological and medical science is being conducted in the early 21rst century is that we tend to draw too many conclusions from a reductionist approach to the scientific method. Allow me to illustrate:

It’s late Christmas night, you are tired beyond description from a day filled with the nagging details of other people’s lives (your in-laws) and in the glow of your favorite computer monitor you decide that it is time to figure out what makes that 5.1 megapixel digital camera really work. If you are a male, like me, you will not look for the six pound instruction manual – it’s written in French anyway – and you will not be satisfied by messing around with the buttons on the back of the little gizmo. No, for this task a tiny Philips screwdriver is a must! Four impossibly small and soon completely stripped screws later you have the camera apart and you are gazing with wonder at the array of circuits, photosensors, lenses, and motors that make up the modern camera.

Next you grab your trusty dissecting microscope – you don’t have one? No worries, I have two – and begin to really look at the motor that automatically used to focus your little camera. It has some small gears and tracks, but is generally electronic in nature. Just to the right there is a tiny chip that is connected to the motor. Closer examination reveals that this chip has little pathways painted in a very intricate, almost petroglyphic design. Whoever made this must have believed in a God of order and simplicity, buried in the complex mechanism of the camera.

Now, if you are really careful you can see how this thing works. Take your scalpel, (or borrow mine) and scratch a little nick in one of the lines on the petroglyph. Be careful not to touch anything else as you reassemble the camera. Finally, turn the power back on and see what happens. If you have done your work correctly and only nicked one line, everything should function perfectly except perhaps one feature. Maybe the perimeter of the photo is now out of focus. From this you conclude that the circuit line you cut is responsible for sharply focusing one specific aspect of your photos.

This approach is really the same one most biologists use to attempt to figure out what each of the genes in the human genome is actually responsible for doing. Essentially, the nucleus is the circuit and the genes coded in the DNA are the painted lines on the circuit. When a biologist wants thinks she has isolated a gene that causes, say diabetes, she will create a mouse (or rat) that has had that specific gene cut out. These little animals are called knockout (KO) mice because they have had a gene, quite literally knocked out of their genome. If our scientist is correct in guessing the function of the gene, the mouse gets diabetes, becomes obese, and watches infomercials all day while the scientist celebrates, publishes a paper in a Nature, gets tenure, and watches infomercials all day... or something like that☺

But that’s not how it always happens. As often as not the camera stops working – the mouse is non-viable. In other words, the gene that she was studying controlled more than just diabetes and the mouse does not live. When this happens, primary investigators curse, post-docs cry, and graduate students get to “stick around a little longer”. There is no paper in Nature, no tenure… weeping, gnashing of teeth… get the picture?

To be sure many interesting and significant discoveries have been made using the knockout approach to biology. I am not diminishing this approach to discovery, but I do think we need to place it into context. When conclusions in medicine are made on the strength of the evidence in KO mice, there are bound to be problems in more complex and highly interactive systems like humans. Early studies may indicate that a gene is responsible for a disease, only to give way to later evidence that contradicts the first papers. This happens all the time in science and is part of the process: part of the search for scientific truth. Trouble is, people who report for the Wall Street Journal and the Marion Chronicle Tribune do not want to wait for the longer more complex studies and we get premature science published in the general media as fact.

Reductionism (looking for answers in the infinitely small) alone is a flawed approach to science and to life. To be understood completely, genes must be viewed in the context of their regulators, and the things they regulate. In science, this has led to the anti-reductionism or “whole animal” movement in recent years. So, the next time you see a study that claims “ the gene for schizophrenia has been identified in KO mice” be skeptical, wait a while, and ask for the context.

No comments: