# IGEM:Paris Bettencourt 2012/Notebooks/Semantic group

(Difference between revisions)
 Revision as of 04:53, 29 June 2012 (view source) (→Going further)← Previous diff Revision as of 05:23, 29 June 2012 (view source) (→Making a Choice between Serine and Tyrosine)Next diff → Line 26: Line 26: ==Making a Choice between Serine and Tyrosine== ==Making a Choice between Serine and Tyrosine== - As we may not have time to test the efficiency of our system with either Serine (S) or Tyrosine (Y), we have to choose. In order to discriminate which of these two amino-acids, I first check their codon usage in E coli K12 ([http://openwetware.org/wiki/Escherichia_coli/Codon_usage| Codon Usage]). It turns out that S has 57,88 codons over 1000 codons when Y has 28,59 codons over 1000 codons. Here it would be more interesting to use the S, as we want our genes to contains more amber codon, so the more S we have, the more amber codon we will have. Second I checked the possibilities of the amber codon to reverse to a S, Y or any similar amino-acids. For that I made a program whose an outline is showed on figure 1. The program calculates a score. The higher is the score, the more likely the mutation(s) will maintain the function. So as we don't want to recover the fonction, we want the lowest score between S and Y. + As we may not have time to test the efficiency of our system with either Serine (S) or Tyrosine (Y), we have to choose. In order to discriminate which of these two amino-acids, I first check their codon usage in E coli K12 ([http://openwetware.org/wiki/Escherichia_coli/Codon_usage| Codon Usage]). It turns out that S has 57,88 codons over 1000 codons when Y has 28,59 codons over 1000 codons. Here it would be more interesting to use the S, as we want our genes to contains more amber codon, so the more S we have, the more amber codon we will have. Second I checked the possibilities of the amber codon to reverse to a S, Y or any similar amino-acids. For that I made a program whose an outline is showed on figure 1. The program calculates a score. The higher is the score, the more likely the mutation will maintain the function. So as we don't want to recover the fonction, we want the lowest score between S and Y. I used 3 types of [http://en.wikipedia.org/wiki/BLOSUM| blosum] matrix, blosum62, blosum80, blosum100, all gave the same relation between S and Y. I used 3 types of [http://en.wikipedia.org/wiki/BLOSUM| blosum] matrix, blosum62, blosum80, blosum100, all gave the same relation between S and Y. Therefor S seems to be less robust to mutation, meanings that if a single substitution occurs in the amber codon, it will be more likely that the fonction of the amino-acid will be change, resulting in an inefficient protein, ''a priori''. Therefor S seems to be less robust to mutation, meanings that if a single substitution occurs in the amber codon, it will be more likely that the fonction of the amino-acid will be change, resulting in an inefficient protein, ''a priori''.

## Revision as of 05:23, 29 June 2012

Semantic containment project

# The idea

We need to prevent our genetic construction from being used by other organism. Since horizontal gene transfer (HGT) can be perform either by conjugation, by transduction, or by transformation, and none of these system is only dependent of our organism, we cannot assume the fact that HGT is fully avoidable. Semantic containment [1] means that our bacteria won't be able to "speak" with other organism, since they don't speak the same language. The language being DNA. Here, it won't be all DNA that we are going to change, but just 1 stop codon that we are going to change in a normal aa codon, let say the aa 'X', for OUR GMO bacteria. So, in case of HGT, the gene transferred won't be able to be translate correctly, since it has many stop codon instead of the aa 'X'.

# What we can use from others

Church Lab already engineered a strain in order to remove the rarest stop codon in E coli MG1655, which is TAG (amber stop codon, 314 occurrences), and replace it with the most common stop codon which is TAA [2]. We should ask them for this strain. Moreover it already exists a tRNA amber suppressor gene, named supD, that replaces amber stop codon with a serine amino-acids [3]. This system used by Anderson et. al has already been used by different IGEM teams, with either Serine (supD) or Tyrosine (tyrS), including Paris and Pekin University.

# What we have to do

## About amino-acyl synthetase of this tRNASerTAG

I didn't find any information about this enzyme, therefor I supposed that this enzyme is not specific of the tRNA amber suppressor. Since it's still a supposition, we will still have to look for this information, because, a specific enzyme would really increase the robustness of the system.

## Making a Choice between Serine and Tyrosine

As we may not have time to test the efficiency of our system with either Serine (S) or Tyrosine (Y), we have to choose. In order to discriminate which of these two amino-acids, I first check their codon usage in E coli K12 (Codon Usage). It turns out that S has 57,88 codons over 1000 codons when Y has 28,59 codons over 1000 codons. Here it would be more interesting to use the S, as we want our genes to contains more amber codon, so the more S we have, the more amber codon we will have. Second I checked the possibilities of the amber codon to reverse to a S, Y or any similar amino-acids. For that I made a program whose an outline is showed on figure 1. The program calculates a score. The higher is the score, the more likely the mutation will maintain the function. So as we don't want to recover the fonction, we want the lowest score between S and Y. I used 3 types of blosum matrix, blosum62, blosum80, blosum100, all gave the same relation between S and Y. Therefor S seems to be less robust to mutation, meanings that if a single substitution occurs in the amber codon, it will be more likely that the fonction of the amino-acid will be change, resulting in an inefficient protein, a priori.

## Proof of principle

We should design an experiment that will show that this system is functional. For instance, we could imagine a transformation test, in which we have 2 plasmids. One (say p1) with a resistance gene with 1, 5 or maximum amber codon instead of Serine codon, and another plasmid (say p2), with the tRNASerTAG. We will then transform either both plasmid or only the first one (p1), and then plate the 2 transformations on an antibiotic. We're expecting to have no colonies in the second case since the resistance gene can't be expressed. Here it would be a binary test and the leakiness of the system can't be appreciated. In order to quantify the leakiness of the system, we can use the β-galactosidase reporter, with an ONPG test for example.

## Going further

Every single synthetic gene we want to construct have to be coded with amber codon instead of any serine codon (TCN and AGY, N being A,T,C,G and Y being T or C, so 6 codons). Moreover, the aminoacyl-transferase, if it is specific to that tRNASerTAG, should also have TAG codon instead of normal serine codon. This would enhance massively the robustness of the system, if it's possible.

# References

1. Marliere, P. The farther, the safer : a manifesto for securely navigating synthetic species away from the old living world. System and Synthetic Biology 3, 77-84 (2009). Paper
2. Isaacs, F.J. et al. Precise manipulation of chromosomes in vivo enables genome-wide codon replacement. Science (New York, N.Y.) 333, 348-53 (2011). Paper
3. Anderson, J.C., Voigt, C. a & Arkin, A.P. Environmental signal integration by a modular AND gate. Molecular systems biology 3, 133 (2007).Paper