Now that we’ve reached mid-August, most of us MA students are returning from summer fieldwork and re-settling into life at Columbia. (Or, if you were hanging around NY this summer like me, you’re getting excited to have the department bustling again.) We’re now looking ahead to the fall: application season and thesis-writing extravaganza. But, the summer’s been pretty productive.
I’ve almost finished sequencing the glucocorticoid receptor gene in 24 African starling species. (I’ve had to re-design primers for an exon that sequenced poorly, despite clear bands on the gel.) Thanks to the photography help of the lovely Laura Booth (an undergraduate in the Rubenstein lab who has blogged about her work this summer here), I’ll show my workflow from DNA to sequence data.
Step 1: Copy DNA (PCR)
Using the primers I designed last spring, I ran many, many PCRs (Polymerase Chain Reactions) to amplify tiny DNA fragments that range from 100-1200 base pairs in length. To prepare a PCR, I first pipet the template DNA, and then add a “mastermix,” which is a combination of molecular water, Tris buffer, magnesium chloride, primers, Taq polymerase and dNTPs (deoxyribonucleotides, or the actual “ATCGs” that the polymerase needs to copy the template DNA). Then, I load the PCR onto a thermocycler and wait for 1.5-2 hours.
Step 2: See If Step 1 Worked (e.g. Gel Electrophoresis)
Next, I run that PCR product out on an agarose gel to see if I actually copied the DNA. The gel is an agarose sugar suspended in 1xTAE buffer, which creates a grid through which the DNA “runs.” On each gel, I add a ladder that shows bands of specific sizes; by comparing the position of each band to the ladder, I can check that my product is the correct size (e.g. 585bp for Exon 8 of NR3C1). I also check that each band is bright and clear, to be sure that there is enough PCR product and that all of that product is the same size. Running gels is pretty satisfying: once you get the hang of loading the dye-PCR mixture into the wells without touching the gel bed or introducing bubbles, it becomes pretty robotic.
Step 3.5: Do Step 1 Again, and Hope Hope Hope It Works
Given that I only had about 40 individuals and 8 exons to sequence, I hoped to finish this part of my thesis by early July, but science rarely works out like that. I’ve run three to four rounds of re-dos (which is normal), in order to finish 7 of the 8 exons. I’m still working on Individual 5-3 in Exon 8, a Lamprotornis regius that hasn’t been cooperating, but that guy should be sequenced tomorrow. I’ve also had to re-design primers for Exon 3, and I’m prepping the sequencing reaction for the finicky 5-3 and another round of Exon 3 in a few minutes.
Step 3: Sanger Sequencing
I use a three-step process for sequencing: two clean-up steps on either end of a cycle sequencing reaction. First, I use two enzymes (Exo and SAP) to clean the PCR product to dephosphorylate and remove free dNTPs and single-stranded DNA. Then, I mix the cleaned PCR product with more primers, sequencing buffer, molecular water and Big Dye (a stress-inducing expensive dye) and load this reaction onto the thermocycler for 3 hours. Finally, I spin that cycle sequencing product down through a hydrated matrix of Sephadex powder to remove any unused dye-labeled nucleotides. Then, I’m ready to take it down to the ABI 3730 sequencer at the museum and wait for my data.
Since the Big Dye is light-sensitive, I do this reaction in the dark (because I’m overly cautious and don’t want to risk losing precious DNA due to a failed sequencing reaction). So, I’ll leave it to you to imagine what the steps listed above might look like (complete with my requisite lab-hunching that I worked pretty hard to correct in the pictures you’ve seen – don’t worry, I have plenty of pictures of what I actually look like in lab.) I’m now off to hide out in the dark again, and hopefully, if the Science gods are kind, finish the very last round of sequencing!