Standards in Synthetic Biology: Part 1

Standards

What are standards and why should we bother developing them? I suspect to many people including my fellow scientists and perhaps even some engineers, that standards are nether a rewarding nor worthwhile endeavor. Standards however are the invisible hand that increases our industrial and scientific productivity and helps reduce costs. Can you imagine a world where every workshop produced its own design of thread on a nut and bolt, or every power company generated electricity at a different voltage, or every internet provider used a different protocol for sending email, or every part of the world used a different way to represent time… the list goes on. The world would be an intolerable place without standards. Ultimately standards improve efficiency, productivity and above all reduce costs.

Developing standards however is never easy. The main issue is usually not the technical details but the social issues that have to be addressed. Developing a standard is usually slow and costly. Setting standards is often accompanied by political, commercial and personal struggles. The different stake holders in the enterprise will have opposing requirements and there will be professional maneuvering to win favor or career advancement. Many outside the standardization process will not be able to understand why the process takes so long.

I have been involved in developing two standards and have observed the development of a number of others. The two standards I am referring to are SBML (Systems Biology Markup Language) and SBOL (Synthetic Biology Open Language). SBML was started in 2000 in order to help users exchange computational models of biochemical networks between different software tools. Ten years later SBML has proved to be a very popular standard in the systems biology community with over 200 tools supporting SBML. The second standard, SBOL, was started in 2008 with a workshop I instigated in Seattle and sponsored by Microsoft. Here I want to briefly discuss SBOL and it potential role in synthetic biology.

I am assuming that most readers know what synthetic biology is, if not here is a brief definition from the Royal Society in the UK:

“Synthetic biology is an emerging area of research that can broadly be described as the design and construction of novel artificial biological pathways, organisms or devices, or the redesign of existing natural biological systems.”

The FAQ at syntheticbiology.org is also a useful resource.

What would a standard in synthetic biology be for? The area SBOL is currently focusing on is developing a standard way to represent synthetic biology designs. An immediate application would be the ability to easily extract a design from a publication. At the present time if you try to extract a design from a published paper it is unlikely, without much effort, that you’ll be able to reproduce the design. The original repressilator is a good example; try to reproduce the repessilator from the published description and you will fail. The reasons for this are simple – describing the design of a synthetic circuit using English prose means you’re very likely to miss something critical from the description. Imagine trying to describe an electronic circuit or even a computer algorithm using just plain English, it is very likely it will not work plus it would be very tedious to use because each author is likely to have their own way of describing the circuit including how they built and tested the circuit or algorithm. The way engineers get around this is to develop standards. In electronic design we have a standard way to visually depict a circuit or special languages such as Verilog or VHDL.  What we need then is something similar to either a standard visual presentation (such as SBOLv) or a special language that describes an engineered cell. This is the purpose of SBOL.

In a later blog I’ll describe SBOL itself in more detail, its philosophy and future direction.

This entry was posted in General Science Interest, Molecular Biology, Pathways, SBML, Synthetic Biology. Bookmark the permalink.

2 Responses to Standards in Synthetic Biology: Part 1

  1. hsauro says:

    Hi Vincent, to answer you immediate questions:

    – Has the synthetic biology community reached out to other relevant communities to learn from existing standards ?

    I suppose you could say yes to this. I was a founding member of the SBML development team and we have Chris Myers who is an SBML editor and an active member of SBOL mailing list. We also have people who were involved in the early development of semantic technologies and a number of ontologies, for example OPB at http://www.bhi.washington.edu/research/SemBioProcess/OPB.htm and the Foundational Model of Anatomy ontology (FMA). We have also reached out to other standards efforts including Sequence Ontology and we are aware of many other standards that we might be able to leverage. It is important however as you suggest to be aware of other efforts, particularly when there is significant overlap. One example of this is SBML and CellML. SBOL will need a modeling component and there is no point in inventing our own given how long it took to develop the existing ones.

    – What was so unique about the needs of Synthetic Biology to justify working on a completely new DNA sequence description standard ?

    This is a good question and has been asked by a number of people. What we found with other standards is that these are highly specific to just describing DNA sequences and they are not designed for easy extensibility. Our plan is to develop a standard that is extensible by others in the community. Other than DNA sequence information, which ultimately will be only a small part of SBOL, we must be able in future to represent circuit diagrams, assembly, modeling, host context, part characteristics, distributed component and devices repositories and so on. Many of the existing DNA standards are not suitable for this kind of extensibility and of course even if they were, we would still have to persuade the community that controls the standard to include the changes we propose. Interestingly Sequence Ontology is open to suggestions for include new terms in their ontology that are applicable to the synthetic biology field.

    – Will this independent Synthetic Biology initiative create a domain specific standard when Biology is now moving towards cross-discipline integration ?

    I am not sure what you mean here but the role of SBOL is to support engineering in biology, particularly at the cellular level.

    – So many well established, and essential to a synthetic biologist, Bioinformatics tools work from other DNA sequence standards. Is it the end-game to be able to run a BLAST using a SBOL description ?

    This is another good question. The libSBOL library that is being developed by Mike Galdzicki, will, if not already, have a plugin architecture that will allow the core SBOL object model to be translated or into other more common formats. For example we have at present the ability to read and write GenBank files. I imagine that other developers will be able to add addition translators to the library.

    I hope that answers some of your questions. Feel free to contact me if you’d like to join the current mailing list, particular since you were at the original Seattle meeting.

  2. vrouilly says:

    Hi,

    I was lucky enough to be invited to the first SBOL meeting you organized in Seattle. I could not agree more with you about the importance of standards to stimulate innovation across all engineering fields.
    And, as you mention, synthetic biology requires to embrace standards to move from expert biological master pieces to mass-produced biological machines.

    I really like the point you make about the fact that setting-up a standard is not a technical feat, but a social challenge to bring to a compromise all the necessary actors.

    This is where my worries start about SBOL. It seems to me that SBOL is defining yet another way of describing a DNA sequence. Other biology communities have come across this issue, and have developed standards (http://www.genomatix.de/online_help/help/sequence_formats.html).

    From my point of view, it raises interesting questions about the SBOL effort:
    – Has the synthetic biology community reached out to other relevant communities to learn from existing standards ?
    – What was so unique about the needs of Synthetic Biology to justify working on a completely new DNA sequence description standard ?
    – Will this independent Synthetic Biology initiative create a domain specific standard when Biology is now moving towards cross-discipline integration ?
    – So many well established, and essential to a synthetic biologist, Bioinformatics tools work from other DNA sequence standards. Is it the end-game to be able to run a BLAST using a SBOL description ?

    The intention here is not to undermine the current process that I support. It is more about making sure that it will have the impact that we are all hoping for.

    I am very much looking forward to reading your next post, so that I can learn more about the current driving forces behind SBOL.

    best,
    Vincent.

Leave a Reply

Your email address will not be published.