Hi All,
I've been using Relion to produce 2D class averages of antibodies negatively stained with UA. Basically we just want to see what conformations the antibodies take to have an idea of their flexibility (so that we can compare different forms of the same antibody). I have been getting what look like quite nice results (we can see differences between more flexible/rigid forms) but I feel that I dont really understand the 2D classification process enough to know if what I am doing is the correct way to go about it.
I have attached a ppt with an example of a typical sample. I have been following the steps in the tutorial (without CTF correction). I manually pick and do 2D classification to produce templates for the autopicking.
After autopicking I usually have between 5 000 and 20 000 particles. With the 2D classification I usually use #100 classes initially (should I use more if I have nearer to 20 000 particles or is that just overkill?), then I select all the good/semi-good class averages, extract the particles and re-do the 2d class averages with the number of classes > number of class averages selected. Is there a benefit to doing this? From just looking at the class averages produced I'm not sure. My thinking behind this is that if I select the good/semi-good class averages and extract all the particles that make up these classes and then re-do the Class2D with more classes I will be able to weed out any fuzzy/not well defined particles...........
Any thoughts very welcome!
Thank-you for your time,
Sonya
########################################################################
To unsubscribe from the CCPEM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCPEM&A=1
|