Gene patents won't impede whole genome sequencing, law professor says

In a working draft titled "Will Gene Patents Impede Whole Genome Sequencing?: Deconstructing the Myth that 20% of the Human Genome Is Patented," University of Missouri associate professor of law Christopher Holman argues that the fear that gene patents will have a substantial negative effect on genetic testing is unfounded.

Holman takes as a backdrop to his discussion the case of Association for Molecular Pathology v. PTO where, during oral arguments at the Federal Circuit, Judge Bryson asked the patentee's attorney whether the patent would be infringed by the sequencing of an individual's genome. Underlying that concern was the widespread belief that 20% of the human genome is patented, which Holman traces to a study by Jensen and Murray published in 2005. Holman deconstructs that study and explains that there is no evidence supporting its conclusion.

The Jensen and Murrary study, Holman points out, was not based on patents claiming human genes. Rather, it considered issued US patent in which a human gene sequence, or the protein enoded by a human gene sequence, was merely mentioned.

Upon reviewing a random group of the gene patents used in the Jensen and Murray study (533 out of 4270), Jensen found that many would not be infringed by DNA sequencing and few, if any, include "claims that a court would necessarily find valid and infringed by all forms of DNA sequencing."

Holman explains that patent claims covering isolated forms of DNA molecules are unlikely to be construed broadly because if they were, they would facilitate invalidity attacks based on the novelty, written description, or enablement requirement.

In addition, sequencing may avoid infringement based on the differences between cDNA and genomic DNA. Most gene patents are based on the isolation and sequencing of cDNA. Genomic DNA, however, contains introns which would be present when a genome is sequenced thereby avoiding infringement.

DNA sequencing technology also usually avoids infringement because it does not involve isolating the full-length gene. Rather, only fragments of the full-length gene are physically sequenced and then pieced together.

Holman concludes that the Jensen and Murray study provides no basis to infer that 20%, or any defined percent of human genes, are covered by patents that would be infringed in the course of gene sequencing.