Mailund on the Internet

Years back, when I was studying computer science, I took a course on virtual machine design by Lars Bak.  At the time he had just returned to Denmark but was still working at SUN and he managed to get a VP from SUN to give one of his lectures for him (I forget who it was, as I said it was many years ago).

That particular lecture wasn’t about building object oriented virtual machines but about building successful software companies.  (No snide remarks about someone from SUN talking about that, please).

This was during the .com bubble or just while it was bursting, and the advice he gave was: “when everyone is digging for gold, you get rich by selling shovels”.

If you build the basic infrastructure that everyone needs, it might not be as glamorous and if you are selling commodity products you won’t get rich over night, but if you are selling something that everyone needs, you won’t loose your market over night.

Personal genomics and medicine shovels, anyone?

I’m telling this story because I just read this post at Genetics Future. It concerns genetic testing and how it will soon change with complete re-sequencing which will be cost-effective Real Soon Now(tm).

The post ends:

There are ruthless economies of scale in the human disease genomics business, both in terms of sequencing infrastructure and the costs of assembling reliable knowledge bases for interpretation, so it will be increasingly difficult for smaller companies to stay competitive.

The personal genomics and genetic testing field is another gold rush (although one where small garage companies aren’t quite in on the game yet).  Right now there’s plenty of testing labs, but with resequencing we’ll probably only get a few large companies, at least until the price for resequencing drops significantly.

I don’t want to compete here.  I’m sure I’ll lose.  I would absolutely love to be selling shovels to the gold diggers!

What will all these companies need?

Of course they will need IT infrastructure to manage their data and statistical methods to correlated genotypes with phenotypes.

The question is, of course, whether it will be possible to sell bioinformatics to such companies, or whether they will want to build all their informatics in house.  Some, they want to, of course, as that will give them a competitive advantage, but surely there will be some commodity software they will want to buy somewhere else.

They won’t build their own OS or database system, but probably their own specialised statistical models. Somewhere in between, there is money to be made, if I can only figure out how…

Google Health is up and running. See a review here.

My first thought was: “who would use such a service?”

I don’t have much of a “medical history” to keep track of.  I’m not eating any regular medicine and never had, and in any case it is not something I want to try to keep track of myself.

I trust my physician and would want him to figure out what medicine I need, if any.  I want him to keep track of my medical history.  He is the expert, that is why I go to him.  I don’t want to second guess experts in a field where I have very little knowledge.

So really, what’s the point in using Google Health?

Today my blog received a lot of traffic about this post from yesterday about the relative risk of disease genes. I wrote that the relative risk (RR) of the genetic variants we have discovered recently using genome wide association studies are rather small — 1.1 to 1.5 — and that such a small increase did not matter that much, all in all, and that I doubted that it would have much of an impact for us to know we have a gene that increases our risk that little.

All of this I stand by. The numbers for the relative risk are also consistent with the papers I’ve read, but I have not done a proper survey to see the actual distribution of the RRs. I am thinking about doing that now, but it is not quite as simple as it sounds to figure it out. There is something called “the winners curse” that essentially means that our estimates of the relative risk tends to be higher than the risk really is, because we estimate the risk from a biased sample: the sample where we discovered the risk in the first place. See Zöllner and Pritchard: Overcoming the winners curse for more on this.

I gave an example, however, where I said that increasing the risk of cancer from 0.1% to 0.15% — a relative risk of 1.5 — would have no consequence what so ever. Those numbers I just made up. I intentionally picked very small numbers to make a point, but it is a bit dishonest. I don’t know what realistic numbers would be, to be absolutely honest, but these are probably way too small for any “interesting” disease.

If the risk of a disease, without “risk genes”, is 0.1% I don’t think we would bother with it in the first place. It would be pretty hard to find enough cases for a study anyway.

Realistic numbers might be 5% to 7.5% or 10% to 15%. I don’t think it changes my point: people are not going to change their habits for such an increase in risk when the do not change their habit for much larger risks such as diet, exercise, smoking, drinking, etc. As Genome Technology Online puts it: That’s Because Risk Is Small and Inertia Is Great.

Anyway, I shouldn’t have made up numbers like that — even as an informal example to make a point — and I wouldn’t have if I knew this many people would read it…

Now I should probably go figure out some accurate numbers so I don’t make the same mistake again.

This post is inspired, but not really related to this blog post: In what we trust. It is an interesting read, so go read it, but the one sentence that I want to focus on is this (emphasis mine):

The headlines seem inevitably to contrast starkly with the output of government and industry that seeks to quash our fears and to emphasise how doubling a tiny, tiny risk is no big deal.

With personal genomics getting big, I’ve been thinking about the impact of knowing your genes and the disease risk they carry. With the validated findings we have, the relative risks are all very small. Somewhere between 1.1 and 1.5, most of them. Having one variant over another might increase your risk of a certain kind of cancer from 0.1% to 0.15%. Is that really going to matter for you?

Considering that smoking has a massively larger relative risk, and that so many people still smoke, do you really think that they will change their lifestyle if we tell them that they are in a risk group due to their genes, if the increased risk is really that small? Do you think they people will change their diet if their increased risk of diabetes is that small?

I seriously doubt it.

What we learn from association mapping about the genetics of diseases is important for our understanding of those diseases, but — human nature being as it is — I don’t think it will matter much for the individual to know his personal genetic risk.

Here’s an interesting story: Insect Attack May Have Finished Off Dinosaurs.

Apparently, a lot of disease carrying insects appeared at the time of the dinosaur mass extinction. If the dinosaurs’ immune system was not up to the task of defending the host against this, that might be what killed them all off.

“We can’t say for certain that insects are the smoking gun, but we believe they were an extremely significant force in the decline of the dinosaurs,” Poinar said. “Our research with amber shows that there were evolving, disease-carrying vectors in the Cretaceous, and that at least some of the pathogens they carried infected reptiles. This clearly fills in some gaps regarding dinosaur extinctions.”

Personally, I know nothing about dinos and cannot judge if this is a reasonable theory or not, but I did find it an interesting read.