Qestion on MCore models

[Hamidreza Rahmani]

Dear all,

I am working on the whole CZI dataset for the particle-picking package our collaborator is developing. I imported the Aretomo alignments successfully, however I had to convert them to xf and the local alignment info is gone. 

I have a few questions:
1. Is there a way of reading the full local alignment? This would be very helpful since I am trying to compare results. 

2. If I use MCore to improve my results (I am hoping to use all of them, or the best targets) the xml files will change and the Tilt-series alignment will improve. Is this correct?

3. Can I make copy the original xml files and revert back to the pre-M alignment by putting the old ones in the warp_tiltseries/ directory ion case I needed that?

PS. I noticed that WarpTools doesn not maintain extra columns in starfiles like Warp did, this would be very nice feature to add, but I can also do this in python and it is not high priority. 

Thanks so much, 

Best,

Hamid

[Reika Watanabe]

Hello! 

My questions might be somewhat related to Hamid's question. 

I have several independent maps from a single WarpTool processing folder ready for M refinement.

Since they are independent structures, I plan to run MCore step individually.

After starting MCore, xml data from both tilt-series and individual frames are modified. 

For tilt series xml files such as TS_01.xml, older versions are saved in versions folder in warp_tiltseries folder. This means, we can come back to previous status.

But for frames xml files, such as Mar18_XXX.xml, are just updated after M, but I could not find older versions in the warp_frameseries folder.

Does this mean that once we modify frame data in MCore, no more coming back to the original pre-m status?

Like Hamid asks, do people save the original frame xml elsewhere and replace it when you run another independent status pre-m status?

Maybe I do not have to worry about reverting to the pre-m status to perform M for the following structures? Or can one perform M refinement of several structures simultaneously without compromising the outcome?

I would like to hear your idea, recommendations and experience.

Thank you very much for your kind attention in advance!

Reika