Re: Human genome project (was Re: Why the future needs everyone!

From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Sun Mar 19 2000 - 21:54:15 MST


On Sun, 19 Mar 2000, Michael S. Lorrey wrote:

> a) the government project has been at it for ten years. Venter started
> two or three years ago and has already sequenced the fly genome, which
> shares about an 80%+ commonality with the human genome, and will finish
> the human genome one or two years before the government project, and at
> a cost that is a small fraction of what the government has spent. I'd
> say thats beating the pants off of them.

I have spent the last 8 years of my life actively following the HGP
(because I *wanted* the genes responsible for aging). I also spent
a great deal of my own money attempting to build DNA sequencers based on
the government-funded university research that indicated the sequencing
could be done much faster. (Because at that time it looked like Perkin
Elmer wasn't trying to commercialize the research because it would have
undercut the market for their then pokey sequencing machines (ABI 373s)).
I've attended either 3 or 4 of the TIGR Genome conferences that are held
each year in Sept. My reading day-to-day reading list includes Science,
Nature, Nature Biotechnology, Biotechnology, the DOE Genome Newsletter,
the HUGO newsletter, and a host of similar sources not worth boring you with.

I humbly suggest that I am the best informed individual on this list to
comment on this topic.

The government project is taking 10 years because (a) when they started
they didn't have the technology to do it at the projected cost. They
had to fund the research to develop the technology first; (b) the
people managing the project were biologists who thought that you had
to produce a very accurate map before you could do the sequencing;
(c) nobody knew that computer power would advance sufficiently
to allow you to assemble a ~10 million piece jigsaw puzzle.

The only reason that PE/Celera is able to do this is that they recognized
the technology shifts that occurred a bit quicker than everyone else.
They also had on their team a very very bright mathematician who showed
that it really could be done. [Side note -- this show the importance
of multidiciplinary teams!]

> The human genome is something that I do agree all human beings share an
> equal ownership in. The government apparently has no intention to give
> Venter any patent rights at all, and is pretty pissed that he's making
> them look like idiots.

If Celera or the other companies meet the patent criteria, I fully
expect they will get patents. It will be up to the courts to determine
how "widely" claims may be interpreted. People within the Govt (and
outside of it) are annoyed, because the fundamental issue (regarding
patents) is whether or not any "invention" has taken place. If
Celera (or before them TIGR) had wanted to patent a process for
high throughput genome sequencing and assembly you wouldn't have
heard a complaint from anyone (except those who might have thought
that the approach was "obvious"). That was the "inventive" part
of this whole effort. Sequencing the genome is just grunt work.

> Considering that any government information that
> isn't classified for defense purposes is required to be made available
> to everyone freely, the government is breaking its own law by refusing
> to provide Venter (who is one of us) with any of their own results.

Hmmm, do you have a reference for this claim? If I look at
 ftp://ncbi.nlm.nih.gov/genbank/genomes/H_sapiens/
there are a huge number of files that were updated as recently
as a week ago. If they are trying to seal the most recent
information to prevent Celera from finishing early, I would
probably applaud the effort. If you view the genome as "public"
property (say like old growth timber on public lands), then private
companies should have to pay for the right to access this resource.
The "public" on the other hand is free to go explore this resource
so long as the leave it behind for others to do the same. In cases
where someone develops a public resource (e.g. ski areas) there are
some pretty complex regulations about their rights and obligations.

> Not only that, but they are trying to get Venter to share his results
> without any sort of propreitary use agreement. When he balked at this,
> they got all huffy and started making him look like the bad guy.

>From the perspective of some individuals, he has been the bad guy
from day one. I would argue anyone attempting to claim a public
resources for his own *is* a bad guy.

>
> So far as I know, all Venter is interested in doing with the genome
> information is to package it and sell it to other researchers.
>
Yes, and when there is value added in that information or how it is
presented, then companies will gladly pay for it. This has been true
with HGS, Incyte, and presumably in the future, Celera.

> The fact that he added value to the information by actually sequencing
> it is the value and investment he has a right to a return on.
I'm not sure whether this is a good analogy -- Did Columbus "add value"
to America by discovering it? Certainly *not* to the people who
possessed it at the time. If PE had been selling the sequencing
machines one for Celera, one for the Gvmnt, (given that the machines
are built based on the technology derived from the Gvmnt investment),
then I would grant a level playing field. However, what Celera has had
in this case is an unfair monopoly on an essential resource.

> He has no problem with other companies doing their own sequencing
> and selling that information. What he has a problem with is other
> companies getting for free his results that the government wants him
> to provide to them. As far as I am concerned, he is totally justified.

This would be correct. If Incyte combined the Celera
data with their database, they could have a substantial leg up
on Celera. It *is not* in Celera's interest to see their data
become public. If the government cannot sufficiently "hide"
the Celera data from the public, then Celera is justified in
keeping it private.

To understand this debate, you need to understand the process involved.
The goverment researchers want a really *accurate* sequence (1 error in
10,000 bases or even much lower). Celera wants an accurate sequence too
(because it makes for good marketing). To get a really accurate sequence you
have to oversample (to the tune of 9-11x coverage (30 billion+ bases)).
Both groups can get an accurate sequence if they spend the required $$$.
If on the other hand, one group can get the other to hand over its data,
when each have 4x coverage, then the group that puts it together can
announce the "finished" human genome and get a public relations "win".
In reality its smoke & mirrors.

What is interesting is the fact that Incyte most likely *has* a relatively
complete database of the expressed genes already and the government isn't
talking with them. So the "fight" is over information that I would view
is of very marginal value. For Celera to get any patents, they would
have to discover genes and figure out a patenting strategy that
Incyte has somehow missed.

At issue isn't really who has the information (both groups *will* have
a rough sequence). At issue is who has the "correct" sequence. And even
that is inaccurate because the variation from individual to individual
means that there is no "correct" sequence. So what is really at stake
is the claim that one group has the "best" consensus sequence.

> > He is not patenting innovative medicine, or proteomic interventions,
> > or even _functional_ variants on the genome.
>
> he isnt trying to patent anything that I have heard.
>
> >
> > Instead, he has filed patents on thousands of _fragments_ of code. Raw code.
>
> Since when?

I have seen a claim in print, that Celera has filed over 7000 patents.
The stock took off like a rocket when that information became public,
so you could probably find it by looking at the stock price changes
and examining public news archives.

Now, the debate is pretty *POINTLESS* because a patent application
isn't worth squat if the patent office doesn't grant it (as happened
with Incyte). Even a granted patent may not be worth squat if
the "claims" aren't for something of value. You patent a gene, but
I later patent that gene for something really *useful* that you
didn't claim and when the lawyers sit down to negotiate royalties,
the guy with the best use gets the biggest chunk of the pie.

Robert



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:27:32 MST