Bioinformatics-GregoryMaurer

Download Report

Transcript Bioinformatics-GregoryMaurer

Subject Matter Patentability for
Bioinformatics Patent Applications
Principles & Practice
Gregory L. Maurer
Klarquist Sparkman, LLP
AIPLA Spring Meeting 2008
Biotechnology/Emerging Technologies Committees
Subject Matter Patentability for
Bioinformatics Patent Applications
•
•
•
•
•
Importance of subject matter patentability
Differences from “regular” software
Kinds of claims available
Two common pitfalls: One solution
Impact of recent cases
Theme: Spectrum of §101 positions
• Zealous representation is great, but . . .
• Include conservative §101 positions.
Importance
• Failure can be disastrous
– Shipwrecked case
– Narrow coverage
• Advocacy Issue
– Unique issue requiring attention/preparation
– Law is in constant flux – subjective tests
– Good advocate makes a difference
Importance: Advocacy
•
•
•
•
•
Is software patentable?
Imagine approaching the issue in 1965.
Any changes between 1965 and 1988?
Any changes between 1988 and today?
Any lessons to be learned?
Importance: Advocacy
• Allowed claim from case filed in 1965:
In a data processing system including a plurality
of magnetic tape units for serially storing data
signal combinations and having means for
reading and writing signals during reeling
thereof in a forward direction . . .
a first iterative control loop means having
means for initiating operation of said sort
performing means to sort sets of said data
signal combinations. . .
Importance: Advocacy
• Allowed claim from case filed in 1988:
In a data processing system including sensing means
for sensing an image and converting said image into
input image data, preprocessing means connected to
receive said input image data for filtering noise from
said input image data, and data conversion means
connected to receive said filtered input image data for
converting said filtered input image data into output
data, . . . , a data processing method of converting said
input image data into said output data when said filter
means is n-sized comprising the steps of:
Practice Take Away
• The law will change, but . . .
• Useful, innovative inventions can still be
protected if presented properly.
Difference from “Regular” Software
• Different Art Unit
– Technology Center 1630
• Cases tend to be huge
• Patent practitioner has more responsibility
– No other person may completely understand
• “Cutting edge”
– Describe practical applications in detail
• Careful: May be seen as mental process
Typical “Regular” Software Claim
A method of compressing a digital image
comprising:
determining a recurring pattern of values in the
digital image;
storing the recurring pattern of values for a first
occurring occurrence of the recurring pattern of
values; and
for subsequent occurrences of the recurring
pattern of values, storing a reference to the
recurring pattern of values in place of the recurring
pattern of values.
Bioinformatics: Mental Process?
A method of determining a set of codependent genes comprising:
identifying a set of one or more genes
having related gene expression data; and
removing at least one gene from the set of
one or more genes based on a surplus
information relationship between the at least
one gene and other genes in the set.
Would adding “computer-implemented” save?
Bioinformatics: Practical Application?
A method of determining a set of codependent genes comprising:
identifying a set of one or more genes
having related gene expression data; and
removing at least one gene from the set of
one or more genes based on a surplus
information relationship between the at least
one gene and other genes in the set.
Practice Take Away
• Not everyone loves software patents, so . . .
• Be prepared for § 101 brick wall. Have
backup positions.
• Not everyone is familiar with your subfield, so
• If “cutting edge,” understand how invention
fits into bioinformatics ecosystem. Be
prepared to limit to identified practical
applications.
Kinds of Claims
•
•
•
•
•
•
•
Method (of finding gene relationship)
Apparatus (computer programmed to . . .)
Beauregard (computer-readable media)
User Interface (to accept commands)
Data Structure (for storing data)
Means-plus-function (Aristocrat)
Others (assay, kit, API, business aspects, etc.)
Practice Take Away
• Claim diversity is advised, but . . .
• It can be expensive.
Pitfall: “Floating” Claim
A method comprising:
generating a quad tree from gene
expression data for respective genes in a gene
set;
identifying a most heteroskadastic gene out
of the quad tree; and
removing the most heteroskadastic gene
from the genes in the set, yielding a reduced set
of genes.
More Solid Version
A method of identifying an outlier gene in a set of
co-determined genes comprising:
generating a quad tree from gene expression data
for respective genes in the set;
identifying a most heteroskadastic gene in the quad
tree as contributing zero information to codetermination;
removing the most heteroskadastic gene from the
genes in the set, yielding a reduced set of genes; and
identifying an outlier gene via application of
applying a fast Fourier transform on gene expression
data for respective genes in the reduced set of genes.
Even More Solid Versions
• Add “outputting” a gene identifier
• Add language about an assay/gene chip
• Add language about purpose of assay
Pitfall: “Parameter” Claim
A method comprising:
generating a first data structure from gene
expression data for respective genes in a gene set;
for a plurality of genes in the gene set, determining
a first parameter for respective genes out of a set of
genes and storing the first parameter in the first data
structure as associated with its respective gene;
based on a gene having a highest value for the first
parameter, storing an identity of the gene having the
highest value in a second data structure; and
for a gene identified by the second data structure,
performing an operation on the set of genes, whereby
the gene set is reduced in size.
More Solid Version
A method of identifying an outlier gene in a set of
co-determined genes comprising:
generating a quad tree from gene expression data
for respective genes in the set;
identifying a most heteroskadastic gene in the quad
tree as contributing zero information to codetermination;
removing the most heteroskadastic gene from the
genes in the set, yielding a reduced set of genes; and
identifying an outlier gene via application of a fast
Fourier transform on gene expression data for
respective genes in the reduced set of genes.
Practice Take Away
• If there is no clear practical application . . .
• The claim is in trouble.
(“Practical” application changes as the field evolves
and is relative to the bioinformatics ecosystem.)
Impact of Recent Cases
• In re Nuijten (Fed. Cir. Sept. 20, 2007)
– “Signal Claim” invalid
– Four categories and “Vacuum” rationale
– Take away: Include definition/examples of
“computer-readable media” or Examiner may
allege it covers a “signal.”
Impact of Recent Cases
• In re Comiskey (Fed. Cir. Sept. 20, 2007)
– “Mandatory arbitration resolution”
– “Mental process” not patentable
– Claim seems to have more than mere mental acts,
but no “machine.”
– Take away: Make sure specification describes that
actions are performed by machine or “tool.”
Impact of Recent Cases
• In re Bilski (Fed. Cir. en banc arguments May 8, 2008)
– “Series of market participant transactions
balances the risk position”
– 5 Questions
– “Technological arts” test?
– “Physical transformation” test?
– Take away: Be prepared for adverse decision or
adverse application of decision.
Thank You
Resources
• MPEP § 2106 (Subject Matter Eligibility)
• 2005 Examiner Guidelines (Subject Matter)
• www.uspto.gov/go/og/2005/week47/patgupa.htm
• Federal Circuit Decisions
• http://www.uspto.gov/go/com/sol/fedcirappeals.htm
• Listen to the Oral Arguments
• http://www.cafc.uscourts.gov/oralarguments/