Bioperl: Re: Open Source and Bio questions

Mon, 06 Dec 1999 13:22:01 -0600

I was at a CHI Meeting on Human Genetic Variation, SNPs, etc. in Boston
a few months back and one of the most interesting talks was by a patent
lawyer covering what you could and couldnt get a patent for and how
these patents might be applied. I have her email address if you are
interested in contacting her further.

The very basic gist of it was this (she was talking about SNPs,
primarily):
[Im regurgitating the lecture handout, my comments are in square
brackets]
----------------------------------------------------------------------------------------------
You find a sequence, you notice a difference between your sequence and
the wild type.

This constitutes an Invention.

You can file a patent application at this point even though you dont
know the significance of the sequence difference.

Patentability of Sequences:
Novelty - the SNP sequence must be new, ie. not available in any public
database or publication.
[This caused a lot of discussion centered around "If I make my computer
generate all possible 100bp sequences and put them in a publicly
available database, does this prevent anyone else from patenting these
sequences or subsequences as they would no longer be novel?"]
Obviousness - a new nucleotide sequence is never obvious
Utility - a new nucleotide sequence is always useful

What can you claim?
"A recombinant polynucleotide *comprising* the sequence [gagga...your
disease sequence here ...gatatatg]"
However, the broadest claim you can make should be the smallest sequence
that contains your nucleotide sequence difference.
[the point being that I can use a subset of your sequence and avoid the
claim, I cant use a sequence that entirely contains your sequence as
this infringes the claim, hence you make the sequence as short as
possible]
Perform a 'prior art' seach against publicly available sequence
information to determine the shortest sequence containing the sequence
difference that is new.

So you might end up with a claim:
"A recombinant polynucleotide comprising the sequence GATCGATC" 
[the middle G is the SNP]

What does this cover?
The 8 nucleotide sequence surrounded by any other nucleotide sequence.
[Yes, this is rather broad]

[She then set out a number of cases where people may or may not be
infringing the patent]

#1. Company A has the patent on the SNP sequence, Company B is using a
full length gene as a drug target which contains the SNP sequence.
Company B is infringing the patent - they are using a polynucleotide
that contains the SNP sequence within it.

#2 Company C is selling a DNA array of 10,000 sequences, at least one of
which contains the SNP sequence. Company C is infringing the patent,
they are making using and selling the sequence.

#3 Company D is selling a kit with a DNA probe of the sequence GATCGAT.
They are not infringing the patent as the patent is for the full 8bp
sequence. Where the allegedly infringing product contains LESS THAN what
is claimed, there is no infringement.

#4 Company E has the SNP sequence in a database for sequence comparison.
They dont infringe the patent as the patent reads 'recombinant
polynucleotide comprising...' and the sequence on paper doesnt infringe
the claim to a polynucleotide.

#5 Company F tests populations of nucleic acid sequences for
polymorhphisms on a DNA array, then hybridises, detects and sequences
the population of sequences. They find they have a sequence which
contains the SNP sequence - they are infringing the patent, they use the
polynucleotide in their process.

----------------------------------------------------------------------------------------------

There were a few other examples but the basic point was that if you
actually had a polynucleotide sequence that contained the patented
sequence anywhere within it, then you infringed the patent. I would
think that for long sequences, things start to become rather hard to
enforce: I patent a 100bp sequence, you use only 99bp in your assay and
you are fine. I would think a sequence would have to be a reasonable
length in order for it to be novel and the longer it gets, the greater
the chance that you can use a subset of that sequence to do your assay
and avoid the patent.

She was also only talking about US patents, from what I understand the
EU may not be allowing the patenting of sequences so if you move your
operation to Europe all bets are off.

That was my understanding of the talk, I know nothing about these things
myself, Im just reporting what I heard! Please dont bombard me with
questions about interpretation of these points. :)

Simon.

Paul Gordon wrote:
> 
> Of course, this is all a little off topic, but still interesting.
> Lookup the patents in the US Patent Office database.
> 
> http://164.195.100.11/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2Fsearch-bool.html&r=0&f=S&l=50&TERM1=gene&FIELD1=TTL&co1=AND&TERM2=sequence&FIELD2=TTL&d=curr
> 
> It seems that not only are sequences patentable, but so are specific
> oligonucleotide primers and more.  From what I can tell, companies just
> making blanket patents, and waiting for someone to challenge them.  In
> doing so, they can lay first claim to most anything associated with the
> gene.  To repeat what has been said before, I don't think there have been
> many challenges or uses of these patents so far, so there meaning is
> unclear.  It will take a legal battle to determine what the patents mean.
> 
> I always though patents were for inventions, not discoveries. Just my
> $0.02 worth.
> 
> =========== Bioperl Project Mailing List Message Footer =======
> Project URL: http://bio.perl.org/
> For info about how to (un)subscribe, where messages are archived, etc:
> http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
> ====================================================================

-- 
--------------------------------------------------
Simon Twigger, Ph.D.

Laboratory for Genetic Research,
Medical College of Wisconsin,
8701 Watertown Plank Road, 
Milwaukee, WI, 53226

http://legba.ifrc.mcw.edu/~simont/

tel. 414-456-4409               fax. 414-456-6516
--------------------------------------------------
=========== Bioperl Project Mailing List Message Footer =======
Project URL: http://bio.perl.org/
For info about how to (un)subscribe, where messages are archived, etc:
http://www.techfak.uni-bielefeld.de/bcd/Perl/Bio/vsns-bcd-perl.html
====================================================================