Gibson Assembly: How to Troubleshoot in 3 Easy Steps

Having left the blogging/podcasting synbio scene for over a year now, I think it’s time to get back. How better to start than by sharing how I troubleshoot the construction of DNA using my favorite method: the Gibson assembly, which literally allows me to piece together almost any piece of DNA with almost any other piece of DNA in an easy protocol. (No more digests and ligation!) The original paper describes the method in great detail; the general principles are described in this webpage hosted on JBEI, and a starter kit for getting up and running can be found on NEB’s website.

Gibson Assembly Setup

When I setup a Gibson assembly, it always starts with the PCR step. A (brief) protocol that I’ve found useful is to do the following:

  1. Primer design: 40 anneal, 20 overhang
  2. PCR conditions:
    1. Phusion HF Polymerase Master Mix + 5% DMSO
    2. 65C annealing temperature
    3. 1 min/kbp extension
    4. Bin the PCRs into <1kbp (1 min extension), 1-5 kbp (5 min extension), and >5 kbp (for however long is needed to amplify the biggest piece @ 1 min/kbp).
    5. Run the PCR overnight


If these PCRs fail, then the three-step troubleshooting protocol I follow is:

  1. Polymerase change:
    1. My favorite troubleshooting polymerase of choice is the KAPA HF 2X Master Mix.
    2. 5% DMSO, gradient annealing, program: follow provided protocol.
    3. All this done at the same time as a temperature gradient. If you have a nice PCR machine, you can set the heat block at 72C (top row), 55C (bottom row), and the 65C temperature somewhere at the 3rd or 4th row. Then, just setup 3 reactions for 72, 65 and 55C.
  2. “Gibson30”:
    1. The sequences at the junction remain the same, but instead of 40 anneal and 20 overhang, I do 30 anneal and 30 overhang.
  3. Alternate junction:
    1. Usually I don’t get to this stage, but if it turns out that I need an alternate junction just to get the necessary sequences into my plasmid, then I will do this.


Following the above protocol, and integrating it into a “batch cloning” workflow (where I try building 30-ish plasmids at one shot), I’ve managed to shorten the time I spend on DNA assembly/cloning, and increase the time I spend on functional assays in my experiments.


For all you Gibson assembly peeps out there – what’s your favorite strategy? If you’ve not tried a troubleshooting strategy before, please try this out, and let me know how it goes!