The 2015 iGEM Undergrad Winners

The 2015 iGEM Grand Prize for an undergrad team went to the College of William and Mary – one of the oldest universities in the United States.


Here is the link to the team’s top-notch wiki page. They worked on noiseBut not the familiar kind that we can hear. They were interested in the noise describing the variability in the performance of a machine – the way its output fluctuates randomly. Genetically Engineered Machines – the GEM in iGEM – show that kind of variability, too.

What kind of noise are we talking about here?

Take, for example, a common device used in GEMs: a protein expression cassette.


The output of this device is the protein produced, and the amount of protein is not the same at every moment in time, but varies depending on a number of factors. One source of variability is the promoter, which is the DNA part where transcription starts.

A promoter works like a switch. When it’s ON, the protein downstream from it gets made. When it’s OFF, there is no protein. At least that’s the way most of us think of it. Also, some promoters are stronger than others, which means they result in more protein. So, on a graph, two promoters of different strength could be pictured as following.


But this is only an approximation. The amount of protein produced is not constant, as implied by the graph above. In reality, this amount varies around a baseline.


But what causes this noise?

While essential to transcription, a promoter is just one of the elements needed to make the protein. Actually, the process is quite complex.

To get things started, the segment of DNA acting as a promoter, together with the DNA sequences surrounding it, must first recruit and assemble the transcription machinery which is made up of a bunch of proteins (the RNA polymerase and transcription factors in the image below)


illustration by OpenStax College

Then, this machinery must find and string together – one by one and in the correct order – the letters for the new messenger RNA molecule – the RNA transcript. And when it reaches the end of the gene, the same machinery must go back and start all over again – numerous times.

Transcription elongation

illustration by

Therefore, even if one considers only the transcription (just one of several processes needed to make a protein), the steps and elements involved are many, which makes it inevitable that some variation in the amount of protein will occur. So yes, there is noise associated with promoters, and the team decided to take a closer look at this.

Why study this noise? 

The intricacy of Genetically Engineered Machines is increasing. New GEMs contain multiple genetic circuits, which must function in a coordinated fashion. Ripples from one “noisy” promoter may affect a genetic web in unexpected and significant ways. Taking into account this noise is therefore not just useful but smart. It will allow us to build the most efficient, safe, and reliable GEMs possible. And, from an engineering perspective, it will gain us additional control.

What came out of this project?

As far as I can tell based on their wiki, the team’s main accomplishments are the following:

  1. a ready-to-use system for measuring the noise associated with any given promoter
  2. quantitative data on noise generated by 3 common promoters from the iGEM Registry: R0010, R0011, and R0051
  3. a new set of tools for the regulation of 11 of the most frequently used promoters from the iGEM Registry
  4. education resources for synthetic biology

In the remainder of this post I will elaborate a little on each of these accomplishments.

  1. The ready-to-use system is based on a composite part the team created: the galK Integrator cassette. It’s a part that can be used to integrate any piece of DNA at a specific location (named galK) on the E. coli chromosome. Using it accomplishes the following:
  • Limits the copy number of the DNA of interest to 1 or 2. This is important for the accuracy of the noise measurements. If the DNA were carried by a plasmid, the number of copies per cell would be hard to control, and noise measurements would be less reliable. This conclusion was drawn from mathematical modelling done by the team.
  • Inserts the new DNA piece into a particular site on the bacterial genome, which has been tested and validated for such experiments. This is important because it keeps the DNA sequences that surround the promoter always the same. These sequences have an effect on how strongly the transcription machinery binds to the promoter, thus affecting the transcriptional noise – by integrating promoters in the same place in the bacterial chromosome one can control this variable.

2. The quantitative data. While all three promoters tested here are categorized as “strong”, the team wanted to know if they were similar in terms of noise, too. Promoters differ in their DNA sequences, which affects how they interact with the transcription machinery. This causes some promoters to be noisier than others – that is, intrinsically. On a graph, two equally strong promoters could look like this.

Slide2How did the team figure out the intrinsic noise of each of the three promoters?

For each promoter they made two constructs, which produced two different fluorescent reporters: Cyan Fluorescent Protein (CFP) and Yellow Fluorescent Protein (YFP). For instance, the two constructs for the R0011 promoter looked something like this.


Then they integrated each of the two constructs into the E coli genome using the galK integrator cassette. Since their identical location allowed one to assume they were identically regulated, the two genes could be regarded as “two distinguishable variants of the same reporter”.

By analyzing the fluorescence data collected from cells with the integrated CFP along with data from cells with the integrated YFP, the noise intrinsic to the promoter at hand could be determined based on a correlation coefficient. The way this was done employs more complicated mathematics, but the following quote may make some light:

A good analogy for intrinsic noise is the correlation between two signals – in the example of transcription, if a promoter has high intrinsic noise, then the signals between two fluorescent proteins under this same promoter should be less correlated than if the promoter has low intrinsic noise.

The results showed that the R0011 promoter is less noisy than the other two promoters by an order of magnitude. This means that the expression level of proteins placed under the control of the R0011 promoter is significantly more accurate and stable than it would be if R0010 and R0051 were used.

3. The new set of regulatory tools. The team created a new regulatory kit aimed at 11 commonly used iGEM promoters. The kit is based on the novel and powerful DNA editing tool known as CRISPR/Cas9.

The kit consists of molecular devices which provide effective blockage of transcription from those 11 promoters. It uses a mutated version of Cas9 (dCas9 or dead Cas9) which binds but doesn’t cut the DNA. The team changed the sequence of dCas9 to enhance its expression in E. coli bacteria. The kit also contains a set of vectors coding for 11 different guide RNAs (gRNAs) whose sequences are exact complements to the sequences of the 11 promoters.

To block transcription from a particular promoter, the cells must receive the part coding for the matching gRNA along with the vector for the expression of dCas9 in E coli. The kit includes an extra gRNA whose sequence doesn’t match any promoter, and which can be used as a negative control.

4. The education resources. For Human Practices, the team held workshops aimed at educating children, youth, parents, and the general public about synthetic biology and their project. They also created a syn bio learning resource ( a booklet) with 24 activities for students of different ages. The team’s wiki says that this resource is free and can be requested by email.







Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s