Synthetic Biology:SB2.0/Biosecurity resolutions/Perhaps a randomized malicious ordering of genes sent to companies to see if they respond appropriately

From OpenWetWare
Jump to navigationJump to search

Better late than never, but I spoke at the MIT Town Hall Meeting last month about the need for some mechanism of testing compliance of the 'code of conduct' for companies that synthesize DNA. A mechanism to do blind tests to see if companies are screening for suspect sequences. If there is no mechanism for 'naming and shaming' wrong doers who do not screen their sequences, then the only way the community could shun them (the proposal suggests that if the community does not support the company then it could not survive) would be through gossip or word of mouth.

Here I propose an outline of a yearly compliance test, please do flesh it out or fill in gaps, or if it is flawed, please say why.

Random sampling and testing of compliance

  1. Compile list of companies to test, revise this list.
  2. consult with these companies and get support for action.
  3. costs could be covered from industry/academic contributions depending on size of organization. External agencies may fund the body too.
  4. Each year the committee could select a different chair and review panel from industry/academia, to share the workload.
  5. A random selection of companies could be chosen each year,with all knowing that they 'might' be selected.
  6. suspect sequences are ordered through 3rd parties.
  7. Results are published each year/2 years, keeping the topic in the public eye, and showing that the community is taking responsibility.
  8. finally, an annual televised award ceremony 'synthetic lab of the year'

Possible problems:

  • the algorithms needed to be developed and distribute to companies to find suspect sequences could be used in reverse to search for suspect sequences. (?)
  • what are the associated risks of actually producing these suspect sequences if they are not checked?
  • creating third parties that the sequence submission would have to go through to keep it anon.

So these are just some thoughts, that suggest one solution to enforcing the checking of sequences. [John Cumbers] may 19th 2006


John, I have several concerns about this scheme in addition to the ones you outline.

  1. I believe the SB community should be proactive in identifying and mitigating possible risks, I do not believe that it is our job to police synthetic companies, in the US or any other country.
  2. The frequency of testing is too low to be meaningful
  3. There is no agreed list of what is suspect. Where is the limit? A full gene? A full genome.
  4. The scheme unfairly targets gene synthesizers and excludes oligo synthesizers

Perhaps an electronic, open source list of "sequences of concern for synthesis"(SoCS)(select agent, IP, etc) that could be accessed by synthetic groups has some merit. Like anti-virus software definitions, such a list could be updated continuously as new SoCS are identified. This definition file would be available to all DNA synthesis groups, from large companies to the future users of desktop synthesizers, and could be incorporated into online design tools.

Andrew --Ahessel 19:39, 21 May 2006 (EDT)