Like many neuroscientists, I’ve been following news of the Brain Activity Map (BAM) project since it was first announced in the New York Times last month. Most news pieces have unfortunately shared very little new information. But as I was reading this piece by John Hewitt, a link caught my eye. It said ‘patent’. This was something I hadn’t seen before. Clicking on the link led me here.

George Church, one of the leading players in BAM, holds a patent for a ‘Nucleic Acid Memory Device’. Below are excerpts from the patent. I’ve bolded key words for ease of comparison to excerpts from the first BAM paper. The patent opens:

This invention pertains to methods of imparting information onto nucleic acid sequences. In specific embodiments, the present invention  provides site-specific recombinase systems and error-prone polymerase systems to alter nucleotide sequences such as DNA

It later reads:

The present invention is directed to the use of nucleic acid polymers …as storage media for memory. Information can be recorded onto the nucleic acid polymers in vivo or in vitro.

Compare this to what was written in the first BAM paper, published in Neuron last year (also available here):

DNA polymerases could be used as spike sensors since their error rates  are dependent on cation concentration. Prechosen DNA molecules could be synthesized to record patterns of errors corresponding to the patterns of spikes in each cell, encoded as calcium-induced errors, serving as a ‘‘ticker-tape’’ record of the activity of the neuron. The capability of DNA for dense information storage is quite remarkable.

The patent doesn’t specifically mention recording spiking activity in neurons. But it seems clear we’re talking about the same underlying technique here. The BAM project now has 3 publications to its name in Neuron (2012), Science (2013) (also available here), and ACS Nano (2013). The nucleic acid recording technique is presented, especially in the Neuron paper, as a cornerstone of the project, and a technology which BAM will help to further develop. And yet none of these papers cites the patent, which was filed in 2007 and published in 2010. (According to Google Scholar, the patent has only been cited twice by articles not related to BAM and written by different authors.) At the end of the ACS Nano paper it reads: “The authors declare no competing financial interest.”

This got me thinking about federal funding and patents in general. I am not a lawyer, nor an expert in patent law or federal funding laws. So I could be way off base here. But isn’t holding a patent on a technology your research project proposes to further develop considered a competing financial interest? You are asking for money, some of which will presumably come from tax dollars, for a project that, if successful, will improve your patented technology and possibly put it into wider use. Both of those results could lead to personal financial gain for the patent holder. My questions is, is this legal? Most information I’ve found relates to patents arising from federal funding, but does not address receiving related funding after one has obtained a patent. I suspect it is legal, as long as proper disclosure about the patent is made at the time you apply for funding. But what types of disclosure are required? Is disclosure only necessary when applying for funding? Or, is it also necessary on any publications related to the project? When is not disclosing permissible?

Let me be clear: I am not accusing Church of any wrongdoing. His website clearly lists all his patents. I’m sure he discloses them properly when applying for funding. He’s also not the only researcher involved in BAM to hold a patent that may fall under the scope of the project. Karl Deisseroth has multiple patents related to optical stimulation of neurons – techniques proposed as ideal to meet BAM’s goal of creating “tools to control the activity of every neuron individually”, and particularly emphasized in the ACS Nano paper. John P. Donoghue has patents on inventions to “decode brain signals” and a “neural interface system” designed, among other things, to record electrical signals from many neurons simultaneously. Sotiris Masmanidis and Michael L. Roukes hold a patent for “micromachined neural probes” –  exactly those proposed for use by BAM in the Neuron and ACS Nano papers. Rafael Yuste has a patent on technology that can be used to establish connections between neurons using flourescent indicators. Other authors of the ACS Nano paper also hold patents, many of which are related to nanotechnology – a primary area on which BAM intends to focus.

While the authors do cite their published articles related to these technologies, none of the patents is referenced nor disclosed in any of the three BAM articles. I also found no mention of the patents in news releases or media interviews of the authors (If someone knows of another piece in which the patents are mentioned, please let me know in the comments). Again, this does not indicate wrongdoing on the part of any BAM authors.

But I do see it as a pretty big transparency failure.

[Note: I would greatly appreciate it if experts would comment about the laws regarding federal funding and patents. However, it should be understood that nothing herein is to be construed as legal advice.]

Advertisements