Team:BostonU/Attributions/Experimental Workflow

After choosing potentially viable split sites and the dimerization domains, our team set out to clone a large set constructs and test them in HEK239T cells. For each indicated split site, we created 6 split pair variants - 3 using the different dimerization domains, and 3 using the same domains but in the opposite orientation. This means that for each split protein, we set out to clone 12 different plasmids. Ultimately, our construct set included over 400 constructs! We actually ended up cloning and testing around 200 of these.

Below are the graphs generated by our MATLAB model for each protein that we aimed to split. The split sites are indicated in black.

Here are the two candidate integrase proteins. Learn more about our experimental motivation for developing dimerizable integrase proteins here.

TP901 was split between amino acids 154-155, 164-165, 250-251, 266-267, 276-277, 307-308, 325-326, and 453-454.

PhiC31 was split between amino acids 184-185, 230-231, 254-255, 269-270, 348-349, 539-540, 574-575, and 590-591.

Here are the two candidate RDF proteins. Learn more about our experimental motivation for developing dimerizable RDF proteins here.

Orf7 was split between amino acids 10-11, 27-28, 39-40, and 47-48. The split between amino acids 47-48 was specifically chosen in a hydrophobic region to serve as a negative control.

Gp3 was split between amino acids 39-40, 76-77, 109-110, 149-150, 160-161, 218-219. The split between amino acids 160-161 was specifically chosen in a hydrophobic region to serve as a negative control.

Here is the saCas9 protein. Learn more about our experimental motivation for developing a dimerizable saCas9 protein here.

SaCas9 was split between amino acids 180-181, 227-228, 253-254, 288-289, 310-311, 327-328, 439-440, 445-446, 500-501, 530-531, 558-559, 600-601, 633-634, 710-711, 750-751, 797-798, and 883-884. 2 splits were chosen in hydrophobic regions to serve as negative controls.