Randomly Generating a Set of Cas9 Sites to Use For Potential BAC Linearization - 2 Methods

20220406

Another method of linearizing the BAC before transfection would be to cut it with Cas9. For this, we would have to generate a random set of 23 bases and a PAM site to have synthesized in our BAC, so that it can split open in the correct place (between the left homology arm and the Amp resistance gene).

  • I did not know how exactly to do this, but in Geneious I created a new folder with DiNV, pBACe3.6, pBeloBAC11, the GFP gene, and the AmpR gene and cat promoter (still not 100% on if I got that sequence correct)
  • Then I picked a random region of DiNV: 57,074 to 58,001, and used the program to find Cas9 sites in that region
  • The program should have checked those other files in the folder to see if there were any off target cuttings possible
  • Then I picked 3 Cas9 sites that have the highest activity score to potentially use (however if the activity score takes into account surrounding DNA, this will not mean anything in the BAC context)
    • 1: 5’ CAAAAGCTTATATTTCAATGCGG 3’ activity 0.692
    • 2: 5’ AAAGATTGCGAATTTAACGGTGG 3’ activity 0.683
    • 3: 5’ CCAGTCACACCAAAAACACTGGG 3’ activity 0.628

20220427

After talking with Rob, we decided that a random species DNA should be used to insert a fragment into the BAC with a sgRNA site. We used the narwhal (Monodon monoceros) which is incredibly different then a fly or a virus so there should be no chance of homology. Additionally, this sequence shouldn’t be included in the BAC-DiNV so the DNA content shouldn’t matter.

  • Seached Monodon monoceros in NCBI Gene and picked the 1st gene: myoglobin Gene ID: 114885572
  • Downloaded it into Geneious in a folder with:
    • DiNV sequence
    • GFP gene sequence
    • Ampicillin resistance gene sequence
    • pBeloBAC11 sequence
    • pBAce3.6 sequence
  • The gene is ~14,000 bases long, so I picked a region between 7,001 and 7,542 bp to look for Cas9 sites
  • Used the Geneious program to look for sgRNAs, hopefully checking everything in that folder for off target hits
  • Picked the top 3 high scoring sgRNAs:
    • 64: 5’ ACTTGGGGTATTGAGTCACGGGG 3’ activity score 0.758
    • 61: 5’ GATGTCTTGAAGGAGACTTGGGG 3’ activity score 0.735
    • 70: 5’ GGATCAGAGAGAACTGCCTGGGG 3’ activity score 0.717
  • These are all within ~90 bases of each other
  • We probably want a 100-150bp sequence around these sites to add into the BAC?
  • Extract out the sequence between 7,310 and 7,440
  • Saved here
  • I’m not sure exactly if we’ll try one sgRNA to linearize or all 3, I’m not sure how I would be able to tell if the linearization happened properly with the BAC because it will just go from circular to linear…