The Supreme Court decision on patentable genomes

I'm shocked. Just totally surprised. And it was unanimous — the Supreme Court determined that human genes cannot be patented. This is excellent news.

Why is it a good decision? Because medical DNA analysis was turning into a patchwork of competing landgrabs. Sequencing technology is coming along so nicely that more and more diagnostic tools are available, that can analyze big chunks of the genome for, for instance, known dangerous mutations. But at the same time, many stretches of DNA were 'owned', or patented by various companies. A company called Myriad had the patents on the genes BRCA1 and BRCA2 which, when defective, are associated with a higher frequency of breast cancer. Another company which might have a tool for analyzing a piece of chromosome 17, where BRCA1 was located, would have to intentionally mask their analysis, hiding the sequence of the BRCA1 gene, or they'd have to pay royalties to Myriad.

This is an increasingly ridiculous situation. Imagine if 50 competing meteorological forecasting companies each had rights to the weather above a different state, and a weather service in Louisiana had to pay the weather service in Florida for the right to examine clouds and wind and pressure to the east, and you couldn't have a national or worldwide weather analysis without paying a thousand petty weather barons. That's where we've been in genetics, with an increasingly balkanized genome and a welter of companies expecting payment if you looked at the DNA sequence in an individual patient.

Francis Collins, director of the National Institutes of Health, who has long argued for limiting private control of DNA data, said today that he was pleased with the ruling. "Our position all along has been that patenting DNA in its natural state does not provide any benefit to the public. There have been concerns that you might have a $1000 genome sequence, but a $500,000 royalty fee to use it. We can breathe a big sigh of relief that this will no longer threaten to inhibit the progress of DNA research."

So, smart move, Supremes. For once they made a decision that didn't simply back corporate interests.

One complication, though. They made this decision based on the logic that the genetic sequence wasn't an invention of the company — it was just what they found there — making that unpatentable. But they also made a decision that cDNA was patentable, which is a little weird.

cDNA isn't exactly an invention by the company. Here's what it is: the genomic sequence of a messy human gene is a cluttered mess. There are regions called exons which code for the proteins of the gene product, but they're broken up by intervening sequences called introns. What the cell will do is copy the whole messy DNA sequence into RNA, and then enzymes come along and snip out the introns and splice together the exons into one continuous sequence. It's like finding an interesting magazine article in which every other paragraph is interrupted by an ad, so you cut it up, throw away the ads, and tape the story together into one complete, uninterrupted flow of text. It's a tedious exercise, but your cells do it all the time.

So this processed RNA is simply the coding part of the sequence, with all the useless bits cut out. Most of our genes are more intron than exon, so this is a fairly significant task; the BRCA1 gene, for instance is made of 24 exons, so those 24 chunks are splice together to make the final RNA molecule.

Your cells do not naturally produce cDNA, so the judges are sort of right to recognize it as an artificial process. To make cDNA, that spliced-together RNA is processed by a reverse transcriptase in the lab, making a complementary sequence of DNA. It gives you a new chunk of DNA without all the introns cluttering it up, which you can then insert into a bacterium, for instance, and put it to work making the full RNA/protein for you.

I guess it's a reasonable compromise to say cDNAs are patentable. There is some specificity to it: you might be selecting a particular splicing variant (there are 38 different kinds of RNA produced from different patterns of cutting and splicing BRCA1 RNA, for instance) with a specific mutation, producing a particular molecular construct that is useful for diagnostics or for experiments. In that case, you have used the sequence to build a useful probe or tool — it seems fair to say your tool is a patentable creation, especially since the underlying genetic sequence is not patented, so someone else could come along and build their own tool from scratch.

There's still one troubling thing about the decision, and it was Scalia who pointed it out.

Although the court's opinion was unanimous, Justice Antonin Scalia added a divergent view. While he agreed with the decision, he could not personally stand behind the "fine details of molecular technology" cited by his colleagues, he wrote, because "I am unable to affirm those details on my own knowledge or even my own belief."

So the judges came to an acceptable decision in this case, but truth be told, none of them are trained in molecular biology and genetics, so they weren't actually competent to make that decision. This is a problem that's only going to grow worse and worse as biology becomes more powerful and more esoteric. It's also a little worrying that Scalia thinks mere belief might have been a useful barometer in making a decision — but the case was so far beyond the bounds of what he understands that I suspect he and the other judges based their decision entirely on the recommendations of the lawyers presenting briefs for their scientist clients.