"The Origin of Specious Nonsense"

IT-Guy wrote: »

I'm a lay person in terms of science but to take dead one's question further, once a complete understanding of the human genome is achieved, would it be possible to predict how a mutation on a certain gene sequence would affect a person or animal?

Short answer cos I'm on my phone: pretty much, yeah.

I'll try a longer answer once I'm near a proper keyboard.

IT-Guy wrote: »

I'm a lay person in terms of science but to take dead one's question further, once a complete understanding of the human genome is achieved, would it be possible to predict how a mutation on a certain gene sequence would affect a person or animal?

Longer answer:

Pretty much, yeah. It could be a while though. The human genome has been sequenced, but that doesn't mean they know which genes do what yet. We don't even know just how many genes are present, exactly. Genes are much more fiddly than people usually realise. The dominant/recessive Mendel stuff most secondary school students end up learning is a gross oversimplification, for example, as it suggests that the recessive gene is inactive when it's still producing proteins that interact with other proteins. And those interactions are very complex. Change one thing and it can cause a massive cascade of changes. Which is, incidentally, of great use in designing drugs and antibiotics; create a substance that affects one little important gene, and it screws with every other gene it interacts with. Easy way to kill bacteria or tumours, knock out the protein with the most interactions.

It's not just a simple case of one protein interacting with another, though. some interactions are chemically strong, some weak. Some proteins interact with several others (some strongly, some weakly), some only interact with themselves, some change what other proteins interact with. some bind together to make different proteins not actually coded for in the genome. We're still not sure how many genes there are in humans because of several factors.

One of those is pseudogenes: generally genes that are damaged so they don't get expressed, or are expressed differently. Typically a small mutation will cause a new "stop" code in the middle of the gene (Or the removal of an existing one) so the protein ends up too short or too long, or the insertion of a new nucleotide or two will cause a frameshift mutation and entire sections of the protein are suddenly made of something completely different. but they can sometimes still have a function; some are accidentally transcribed because they're situated next to a fully functioning gene; Some are identified incorrectly because actually determining a gene's start and stop involves very complex algorithms which don't always get it right; Some are still functional, but the mutations cause the protein it codes for to assume a different shape which might cause it to stop interacting with some proteins or interact with new ones. Most of the time it just renders a gene (and some of the genes it affects) non-functional, but sometimes it doesn't.

There are other factors, like some regions of DNA being overlooked because they code for a protein so small it seems unlikely to do anything, but it still does. Figuring out the genetic sequences is one thing. Figuring out which bits of that sequence actually do stuff is another entirely, and requires everything from discrete mathematics to advanced organic chemistry to quantum physics.

Here's a picture of a small protein-protein interaction network for the human genome. It includes 1,700 proteins. Pretty complicated, isn't it? Note how many grey points denote proteins that are known to do something, but we don't know exactly what yet.

Now consider that current estimates suggest that human DNA codes for somewhere between 20-25,000 proteins. The complexity of such a network is hard to fathom. But, and this is the important part, it's not impossible to understand. Science has only scratched the surface of this stuff really, and we're starting to get a grip on some of it. As usual, it starts with the useful and profitable bits, so we know most about the sections of the genome that are involved in cancer and genetic disorders, and the results of that research has been invaluable in saving lives with new therapies.

It's like putting a really big jigsaw back together, when the jigsaw has a million pieces and it's a picture of something repetitive like a plate of beans, and when you slot two pieces together they change shape. In the last 50 years or so, what we've done is essentially find a corner piece that we can work outwards from. It'll take a lot of time, but eventually we'll figure out where all the pieces go and how they affect all the other pieces. Once that's done, predicting the changes caused by a given mutation or selective pressure will be pretty straightforward.

As with most of science, it's only a matter of time. Religious loons keep shouting "SCIENCE DOESN'T KNOW X!". They keep forgetting the last word, the most important word, in that sentence. Science doesn't know X yet.

So dead one hates blowjobs? That would explain a fair amount of... Well, yeah.

... Are you coming on to me?

At this stage, in this thread, with my reputation...

I'll stick to the bacteria, thanks. Far less fiddly and no jackasses from the ethics committee calling round to ask awkward questions about the dead test subjects in the boot of my car.