Engineering

We attempted to design experiments based on ANOVA statistics to be applied after protein predictions made in software. We discovered this design was too ambitious and not suited to our resources.

Initially Dry Lab hoped to have the cooperation of a protein-folding expert. Considerable time was spent running OpenFold, but the expert had to leave for another country due to a personal emergency and never joined the team. Thus Dry Lab did not discover which proteins should be run through OpenFold.

Furthermore, Dry Lab believed that we could get a Taguchi-based experiment with the following dimensions:

1. pH
2. temperature
3. amount of enzyme (mass in grams)
4. amount of EG (mass in grams)
5. amount of TPA (mass in grams)

However, that assumption turned out to be entirely unrealistic. In the first place, multiple enzymes would be necessary. In the second place, the physical constraints of the laboratory limited the types of experiments that could be done, and the Taguchi design was never clearly explained to anyone in the Wet Lab. Thus Wet Lab moved forward with entirely different ideas.

Some of the prototype simulation programs from January are shown on the timeline page.

We made a more practical attempt to proceed with enzymes and simple mathematical models particularly comparing a Michaelis-Menten model with a Hill Coefficient model.

Christian led the Wet Lab team to successfully engineer microbial cell factories.

Building on the success of the June experiments, Nico proposed an innovative Python pipeline which is reported at the software page.

Preston designed a clinostat which is reported at the hardware page.