Team:The Tech Museum/Project

Overview:

We created a pool of plasmids designed to produce wide hue diversity in bacteria. Variation in promoter strength randomizes the relative expression levels of red, yellow, and cyan color reporters in each plasmid. In this way, we can create bacterial ‘pixels.’ Theoretically, the hue of each resulting colony should represent a particular combination of reporter protein concentrations, similar to how an RGB LED operates.

Museum visitors are guided through the transformation of e.coli with this plasmid pool to generate plates with a rainbow of bacteria colonies. Next, they take those petri dishes to an interactive scanning station. We developed software that uses digital imaging and computer vision to analyze the color, intensity, and rarity of the bacteria colonies on the visitor’s plate. A dynamic visualization of our team’s aggregate color data is then updated in real time with each participant's individual contribution to our iGEM team.

Which promoter-RBS-color combinations will fail? Which colony hues will we not ever see? Which will dominate? Our software and the participation of museum visitors is designed to find that out.

Project Details and Documentation:

Biology

To generate random color diversity in bacteria, we designed and had assembled libraries of tri-color plasmids (thanks to the generous sponsorship of DNA2.0). The base arrangement of plasmid functional elements was modeled on the three-color expression plasmid pZS2-123 designed by Cox et al. (2010). Our plasmids contain three different reporter proteins that are each controlled by combinatorially-inserted promoter-rbs pairs of varying strengths. Nine previously described promoter-rbs pairs that span a range of expression levels were selected from the literature (Kosuri et al. 2013; see notebook for specific sequences). When paired with the 3 reporters, the output of this assembly strategy is a plasmid pool with 729 unique combinations. Theoretically, each unique plasmid should drive a specific ratio of the three different colored reporter proteins to produce a defined hue.

In total, 4 plasmid pools were generated. Two reporter protein versions of the library were made: one with fluorescent proteins (PaprikaRFP, KringleYFP, CindyLouCFP) and one with chromogenic proteins (red, yellow, blue). Additionally, each of these was also made with both a low and high copy ORI. We tested and optimize all four of these plasmid pools in our visitor-accessible museum wetlab setup. After optimization of e. coli transformation and plating conditions, we determined that the low copy fluorescent plasmid pool was ideal, so it was used for all subsequent exhibit prototyping and data collection with museum visitors.