Using M to Refine Membrane Proteins

[Euan William Pyle]

Hi all,

I'm currently refining a membrane protein using M, and wanted to crowd source some opinion about the best way to do this. 

Currently, I'm using the Refine half maps from RELION, in combination with a Mask generated by RELION where I include the membrane (I just threshold the structure from a masked postprocess job, and use that).

I was wondering about using other strategies with this however. As per a previous thread on here (several years ago) there was a suggestion to use OccuPy to reduce the membrane signal in the reference itself (I assume the half maps). I was wondering if anyone has tried this?

The other option I was considering was to simply not include the membrane in the mask. The protein has a pretty large cytosolic domain (it's ATP synthase) so I assume that's enough signal. 

Interested if anyone has any thoughts, or suggestions!

Thanks

Euan