Defining the terrorist risk

The participants of this roundtable, including myself, agree that it is a bad idea to limit access to pathogen sequence information. However, we seem to disagree on why we agree. Iris Hunger wrote that it is unlikely that available sequences will be misused by terrorists because gene synthesis requires tacit knowledge. But what do we mean by “terrorist”?

While I agree that the role of tacit knowledge in the creation of biological weapons is very important and frequently overlooked, I also agree with Gigi Kwik Gronvall that its importance diminishes if we assume that a terrorist group either has the means to hire trained scientists or that it consists of scientists. Additionally, one does not need modern gene synthesis technology to create the genomes of many viral agents: standard recursive polymerase chain reaction will suffice and has been around for many years. (For example, the first reported creation of a pathogen, the in-vitro synthesis of a poliovirus, was not based on a novel scientific method. At the time, scientists were using more efficient and much faster ways to create large complementary DNA blocks based on sequence information. Eckard Wimmer’s poliovirus publication was simply a proof-of-concept project.)

Many priority pathogens have simple and short genomes, which themselves are infectious. The methods to create these genomes are so standard that they are not even described in the method sections of publications anymore. This means that the tacit knowledge to apply these methods is widely spread and practically speaking “for hire.”

Other agents require more sophistication. The genomes of negative-stranded RNA viruses, for instance the Zaire ebolavirus or 1918 H1N1 influenza A virus, are not infectious by themselves, but require the presence of viral helper proteins, which also have to be synthesized and present inside of a cell in the right numbers. It takes longer to create such reverse genetic systems, and a limited number of people have the skill to succeed, but the methods themselves are not any different from those routinely used by thousands of scientists and taught to students within months. Also, scientists have already created and published many RNA viruses of concern, using individual sequences obtained separately–not synthesized. Other agents, such as the variola virus and bacteria are not within reach of individuals, because the methods to synthesize them are not as widely distributed or not yet developed.

Tacit knowledge becomes particularly important for the steps after the synthesis of a virus, as Leonid Ryabikhin asserts. The methods to stabilize, coat, store, and disperse a biological agent are highly complicated, known only to a few people, and rarely published. I am not too concerned about publicly available pathogen sequences–not because I think people can’t misuse them to synthesize an agent, but because if they chose to do so and succeed they will in all likelihood get stuck during the weaponization process. Meanwhile, the scientific community uses pathogen sequences to gain tremendously important information about creating countermeasures against all kinds of threats.

Hunger’s suggestion to control the work based on publicly available sequences merits further discussion. It is often suggested that companies offering gene synthesis should screen for “dangerous” sequences. But how would such “dangerous” sequences be defined and who would establish the definitions? It would be counterproductive for ordinary infectious disease research to ring alarm bells any time scientists order a piece of a Priority Pathogen genome. Entire laboratories and institutes research individual genes or proteins of microbes without ever touching a “live” (replicating) agent. A company or, better, an outside oversight body would have to find a way to distinguish between these institutions’ orders and those that might lead to the assembly of a complete pathogen genome.

How could this be achieved? And how would a decision be made to allow one “good” laboratory to assemble a genome, but not another? Are there other ways of controlling such work? Tracking the global distribution of gene-synthesizing equipment will only be a temporary solution, so we need to broaden our thinking.