Team:Johns Hopkins-Software/Safety

With regards to safety, one potentially concerning aspect of our software project is the vast amount of information a user has direct access to. This data comes in the form of many thousands of features with associated DNA sequences. But, our database contains well-known, hand-curated features, and we do not believe that these commonly-occurring features present a direct safety concern for the design of unsafe or harmful pathogenic plasmids, for example. Our features come from the Saccharomyces Genome Database, the E. coli K-12 database supported by the University of the Wisconsin Madison, and the Registry of Standard Parts.

Though we do have a database of pathogen genes to allow the user to scan their DNA, for potential pathogenic sequences, the user does not have direct access to the details of this database. For instance, users may not view the sequence of a particular pathogen. Our software simply informs a user if their sequence is potentially pathogenic without providing any further details. Our list of features were obtained from the Virulence Factors Database maintained by the State Key Laboratory for Moleclular Virology and Genetic Engineering, Institue of Pathogen Biology, CAMS&PUMC.

One feature of AutoGene that can pose safety issues is the ability to add features into a user’s personal stand-alone database. This may present safety hazards if the user adds pathogenic features, for instance. However, through constant updates in the software, we can attempt to filter out suspicious use of our software by denying access and usability of harmful genes. In particular, one way we can develop this type of filter is by matching the user’s imported features against our pathogenic database as an initial screening. Over time, our pathogenic database will grow in size, and this will significantly help in detecting unsafe user-imported features. These unsafe features will not be permitted for use in our software.

AutoGene does not produce any new biobrick parts, and therefore presents no hazards or safety concerns in this regard.

Johns Hopkins has an Institutional Review Board but our project does not require IRB approval.

Since one of AutoGene’s modules is capable of quickly checking plasmids for features, this module can be extended to provide users with the ability to screen their sequences for potentially hazardous genes. We believe that producing simple, intuitive screening tools for hazardous sequences for future iGEM teams can help to reduce the chances of adding such parts to the registry. With efficient bioinformatics alignment algorithms, and the speed of cloud computing, this process can be made so that iGEM teams can effectively scan their newly created parts against a large database of pathogenic sequences with ease; in result, allowing iGEM to have screening as a major component of the competition.

While it is certainly true that this technique is applicable to the iGEM teams, this is also one of the main ideas we had for contributing to biosafety engineering. Through a pre-screening detection process similar to that found at an airport, new biological parts, devices, systems, can be scanned for unsafe components. Since DNA is the primary mode by which biological information is represented, a simple pre-screening process as described above is a reasonably easy and effective way of pointing out safety issues.