Human Practices
Summer STEM Outreach
Every year, the Cooper Union hosts a summer STEM program for high school students. Recently, this program started including a bioengineering section of high school students. Since they were working only a few rooms over from us, we saw this as an opportunity to share our project and findings with the community. So, we prepared a presentation for the high schoolers and allowed them to give us feedback and discuss our project. This was a valuable experience for both the STEM students and for our igem team.
This presentation was useful for the STEM students because it gave them an idea of the type of work that is currently going on in biology research labs. It also showed them that the basic lab techniques they were learning could be applied to more advanced project ideas. We hope that this presentation sparked their interest and encouraged them to continue to pursue research projects in genetic engineering and biology. This outreach also proved to be beneficial to us. Preparing a presentation on our project helped us organize the various parts of the project and improve our public speaking skills. After the presentation, we had a bioethics discussion with the students. Their feedback helped point out possible misuses of our device and what this could mean if it became readily available to the public. We followed up some of the issues they raised in our team ethics discussion. The discussion and presentation helped us improve our overall presentation while creating a memorable experience for both ourselves and the high school students.Bioethics Team Discussion
Cooper Union 2015 iGEM Discussion of the De Novo DNA synthesizer: Biosafety, Biosecurity, Copyright and Market Needs.
In Attendance: Keith Joseph, Ben Ma, Chris Lastihenos, Tushar Nichakawade, Susung Choi, John Song, Jean Lam
Moderator: Oliver Medvedik (instructor)
Background and topics for discussion:
Our focus this year was to continue working on and optimizing an enzyme based personal DNA synthesizer that would permit the fast, cheap and reliable de novo synthesis of DNA by end users. We decided to transcribe our many thoughts throughout the summer months into a more formal document that would be available to all. Specifically, we had three broad categories of questions that we wished to address, and if possible, implement into our device. We first dealt with getting the device to market and what, based on our personal experiences, was the “pain point” for the consumer that our device would address. The second question had to do with biosecurity and biosafety, namely, what, if any, safeguards should be in place in a device that would essentially be radically lowering the barrier to entry to synthetic biology. The last category touched briefly on intellectual property issues and having a device that can essentially synthesize any bit of DNA that you wanted.
Meeting Minutes:
Market NeedsOliver: DNA synthesizers have been around for over 30 years, but they are relatively difficult to use and the reagents involved are noxious. We are making something that’s smaller, cheaper, and simpler to use. In what settings do we envision the device being used? Given these proposed settings, what advantages must the device possess in order to compete with present day DNA synthesis? If this device was ready to ship, who will purchase it and why?
Jean: I anticipate that High schools would be one of our expected purchasers. Since they tend to have smaller budgets, a DNA synthesizer would fit their budget. More and more schools are getting into synthetic biology and synthesizing your own DNA, cheaply, would definitely be a cost savings. Also, small labs in general that need to generate a lot of DNA, such as start up companies.
Oliver: The cost of G-blocks from Idtdna start from $89.99 for 125bp and up to $329.00 for 2000bp. Plus, you’d have to wait about a week. So would the suggestion be that our device could either a) cost about the same to synthesize DNA, but with faster obtained products, i.e. 24 hour turn around vs. 10 days or b) be significantly cheaper to synthesize DNA but with similar turnaround and still meet consumer needs. Ideally, of course, having DNA produced both more cheaply and quickly would be the eventual target.
Chris: In the STEM 2015 program in Synthetic Biology that we taught this summer as part of our iGEM Outreach, we had students waiting over a week to get their DNA synthesized, thus significantly delaying their projects. Time was truly critical as the entire program only ran for 6 weeks. So a faster turnaround in time, even if the cost is not appreciably lowered, would still convert into a massive advantage.
Chris: Thinking of price, we spent about a few hundred dollars on the hardware to build our initial prototype. Thus, the actual DNA-synthesizing device will probably cost as much as large DNA sequence that is synthesized commercially. It could pay for itself relatively quickly.
Oliver: We should also consider that there are two halves to the hardware. The electronics and mechanical components, and the glass slides that house the reaction chamber. Currently, the glass slides are disposable. Either the slides would have to be reusable, or if disposable ones are made, we have to factor in their cost as DNA is being synthesized.
Tushar: The cost is important when thinking about how we purchase primers and G-blocks. The concentrations that we obtain means that for us we have enough to last us for months. So on one hand, you can save money be having pre synthesized DNA for your class, but you are now limited with what you already have. By synthesizing DNA faster and more cheaply, the lab experience for students would be more meaningful as they would be able to design and implement their own sequences.
Oliver: Our first proof of concept is that everything is in one reaction chamber or well and then is subsequently amplified. Right now we have one well per slide, but if we scale it all down significantly we can have a slide, potentially, that has an array of hundreds or thousands of wells where different sequences can occur. This would be a factor of driving down cost. This device is open-source. We hinge on the reversible protecting groups. These nucleotides have to be purchased.
Biosecurity and biosafety
Oliver: Should there be any safety mechanisms in place to prevent the production, inadvertent or otherwise, of DNA sequences that encode select agents or toxins? If so, what safety mechanisms should be implemented and how can they be implemented with minimal disruption of the user experience?
Keith: Thinking back to how IDT checks sequences by not synthesizing sequences that may harm individuals or society, I think it’s important we think about how this type of system could be implemented into our device. Liability issues may arise if someone who owns our device can design and create a harmful DNA sequence.
Ben: The thing is, our device is open-source. People have the ability to change the code that our hardware utilizes. If we are selling it as a private device, then a “sequence checker” makes sense. But it wouldn’t work as open source. The check would therefore have to communicate with a security server.
Oliver: Would a passive block be good enough? In other words, most biotech companies will not sell reagents and equipment unless it is going to a commercial or non-profit entity such as a school, rather than a residential address. There is thus some accountability where the device is sold to those that seem trustworthy, i.e, companies, schools. The risk of course is that something may be synthesized for bad purposes, but that is a risk that society would have to be prepared to accept. Maybe we could also have these devices on a lease. So if there is a situation where the lab closed down, instead of ending up on an online auction it would be returned. For example, Ford’s electric car was leasable in the 1990s.
Jean: Maybe have a Deposit that would then be returned to end users as an incentive to return them?
Ben: I see this as all very relatable to 3-D printers. You can print whatever you want. People can print weapons, but the 3-D printer developers don’t account for this. Our passive block should be enough.
Jean: Is there a way to make our system half-open source and half-closed source?
Oliver: Can there be any other checks in place to prevent easy misuse? What if this device is used to make the sequences for the toxin ricin?
Chris: There would be a long process for further downstream processing of the protein. That in of itself could be a suitable barrier to easy misuse.
Oliver: In the case of a very potent toxin, all our plasmid DNA needs is a transformation step. Small amounts are needed so no extensive purification would be required. Can we prevent this?
Ben: People can also just make their own devices if they really wanted to do something harmful. It’s relatively simple.
Oliver: Even if someone is able to recreate the device, which is an intricate system, the reagents are the choke point. It would not be trivial to synthesize the reversible protective group nucleotides. Also, a crude open source version of the device can be hacked, but to be really accurate, much more work would need to be put in for the finished device.
John: So then the difficulty of reverse engineering the actual device itself would be a check in of itself, even if it were made open source.
Ben: So all the pieces are pretty much regulated since the reagents can only be given to labs/institutions. I think these cover the trust that we should have.
Oliver: High levels danger is difficult to combat. For example(1) Someone really intent on doing something nefarious probably has their own “Dr. Evil” type lab or is sponsored by a government or other organization. This is probably too difficult to thwart at our level. However, let’s say that someone wants to make GFP and they are sabotaged by someone who uploads instead a gene encoding a toxin.
John: Even if we screen for the sequence, users can still circumvent the system by synthesizing smaller pieces that are beyond the limits of detection.
Oliver: But then we have another passive block in that extensive ligations are required. Ben: We could have the sequences sent to our server to be checked prior to the device receiving a signal to go ahead and synthesize. We can have a server where the code is sent to it, which then reports back to the coder whether it's feasible to synthesize.
Tushar: Or sent to the FBI if they are on the select agents list.
Oliver: We can make the argument that if you want to synthesize something, you have to presently be online anyway, so this wouldn’t be something new and it shouldn’t be a problem to have this feature i.e. a device connected to the internet. Probably isn’t a hindrance that would stop people from buying. Also, a device such as this is really useless to anyone without a lab, since you would need to be doing something with your sequence downstream.
Chris: I’m not worried about someone trying to kill one person, but making a disease that can harm many. Like anthrax.
Oliver: The only paper I know of in similar context is the making of the polio virus through oligos and a lot of downstream work was required. Fortunately, it is not easy to say the least to make your own pathogen from scratch.
Chris: In that case then having a software check implemented to verify the sequence prior to synthesis could work.
John: It wouldn’t be easy to hack and circumvent. Some work would be required so it would prevent the easier forms of misuse.
Copyright
Tushar: One last thing to consider is intellectual property and liability. What if people are sequencing sequences that are patented. We don’t want to be held accountable.
Chris: We just have to specify that we aren’t liable such as through an End User Agreement since many other devices can be used to copy information, such as digital recorders, but it is the end user that is responsible for not using it to infringe copyrights.
Summary
After discussing all of these issues, we arrived at the conclusion that as a first step we can implement a software based screen for select agent sequences into our device and then assess the end user experience.