- 1 General
- 2 Protein domains
- 3 Assembly of protein domains
What's this new part type that you call a translational unit?
A translational unit is a part that includes both translational initiation and translational termination. Thus, it begins with a ribosome binding site (or equivalent in other organisms) and ends with a stop codon. In other words, in its simplest form, a translational unit is an RBS followed by a protein coding sequence.
Now natural systems can be quite complicated. In some cases, the stop codon of one protein coding sequence can overlap the start codon of the next. In such situations, the translational unit encompasses everything from the RBS to the final stop codon (and thus includes multiple protein coding sequences). Note, however, that operons are different! Most operons typically have multiple ribosome binding sites, so thus they have multiple translational units.
What's this new part type that you call a translational start?
A translational start is a part that includes both the ribosome binding site and Head domain. Studies have shown that the ribosome binding site, first, second and third codons of the protein coding sequence can impact the rate of translational initiation. Thus, we've defined a new part type called a Translational start that spans the RBS and Head domain in a single part. Thus, Translational start occur at the beginning of translational units.
Why are you splitting up protein coding sequences into multiple protein domain parts?
Proteins are often composed of multiple protein domains, each of which has a distinct biological function. For example, the lac repressor has a protein domain whose function is to bind to DNA and a separate domain whose function is to enable lac repressor to oligomerize. In synthetic biology, a part encodes a basic biological function as a nucleic acid sequence. Thus, since protein domains often have a particular function, it is logical to specify protein domains as a separate basic parts that are assembled together.
What are head domains, internal domains, and tail domains?
Proteins are not just arbitrary sequences of amino acids, but rather they have some structure to them. Proteins have a beginning, middle and end. Certain domains only belong at the beginning of a protein, others only in the middle and still others only at the end. To better describe the underlying structure of protein coding sequences, Tom wrote up BBF RFC 13 called "Rethinking the boundaries and composition of coding regions". Head domains occur at the beginning of protein coding sequences, internal domains in the middle, and tail domains at the end.
- In its simplest form, head domains are composed of a start codon. However, we typically recommend that the head domain include not only a start codon but also codons 2 and 3 of the protein coding sequence since the first three codons are known to impact the rate of translational initiation. Head domains should begin with an
- Internal domains occur in the middle of protein coding sequences. They include neither a start codon nor a stop codon. Multiple internal domains can be strung together. Multiple internal domains are often separated by linkers, a special type of internal domain.
- Tail domains occur at the end of protein coding sequences. In its simplest form, a tail domain is just a stop codon. By convention, Tail domains should end with a double
I'm confused, there aren't any Translational start parts available.
You're right there aren't. However, you can design some new parts and make them available via the Registry.
Assembly of protein domains
Tom's BioBrick standard, Assembly standard 10, doesn't support in-frame assembly of parts. What do I do?
You're right it doesn't. After giving this problem a lot of thought, the Registry has decided to adopt a policy that you can construct a protein coding sequence from protein domains any way that you want. You can use direct synthesis, PCR, or any one of assembly standards 21, 23, 25, and 28 that support in-frame assembly. The only rule is that after construction, the resulting protein coding sequence or translational unit must comply with Assembly standard 10, the original BioBrick standard. The advantage of this approach is that
- Your protein coding sequence or translational unit is compatible with the large majority of parts in the Registry.
- You can use whichever construction method is most suitable for your protein coding sequence or translational unit. If you have a choice, we recommend direct synthesis so that you avoid any scars altogether.
Shouldn't we just pick a new assembly standard that support in-frame assembly?
There is a fair bit of debate on this topic in the synthetic biology community. There's a tension between two competing goals. The first goal is to have everyone use the same standard, because it makes it easy to share and reuse parts. Plus, the number and quality of the part collection grows fastest if everyone can build on each other's work. The second goal is to use the best technical standard available, particular since synthetic biology is still in relatively early days in terms of figuring out how best to engineer biology. The Registry shares this same dilemma. On one hand, the Registry is actively adding infrastructure behind the scenes to add support for other assembly standards. On the other hand, it is impractical for the Registry to support each and every assembly standard that the community develops from here on out. In practice, the Registry will likely have to select a small number of assembly standards to support though it is not yet clear how best to choose which assembly standards to support.
For now, we've decided to adopt a policy that translational units and protein coding sequences can be constructed using whatever method or standard works best for you, but that the resulting part must adhere to Tom Knight's assembly standard 10, the original BioBrick assembly standard. We chose this option for several reasons.
- Approximately 90+% of the parts in the Registry adhere to Assembly standard 10.
- There is no consensus in the community currently as to which new assembly standard to choose.
- The Registry does not currently have funds to remake all existing parts to adhere to a new standard.
Do I have to make all my protein domain parts adhere to the BioBrick standard (Assembly standard 10)?
No. As long as the assembled protein coding sequence adheres to Assembly standard 10, the individual protein domain parts can adhere to any standard. Similarly, you can use any method you like to assemble the protein domain parts -- restriction enzyme digestion and ligation, PCR sewing, direct DNA synthesis etc.