Cryosparc doesn't check axis order when importing maps?

Import and Export
1

Hi,

A colleague noticed something odd when importing two maps into Cryosparc. One map had been generated by another instance of cryosparc; the other had been modified at some point with phenix (I believe phenix auto sharpen).

Loading both maps into Chimera they are well aligned. But when they are imported into CS they are completely off, not aligned at all.

Looking at the headers of CS MRCs, they have the axis order “XYZ”, while MRCs generated by phenix have the axis order “ZYX”. It seems that Chimera recognizes this, while CS assumes that the axis order is XYZ even if it is ZYX, resulting in the observed behavior.

I realize this is a pretty niche case, but perhaps worth adding a header check for such cases.

Cheers
Oli

2 Likes
2

Why does PHENIX change the axes around? I’m curious (some historical XRD justification?) as unless explicitly asked to reorder axes, I’d consider flipping them around a bug in PHENIX…

Nonetheless, a sanity check on axis order is a good idea.

3

No idea what the origin (no pun intended!) of that discrepancy is. Suspect it is a CCP4 vs MRC format thing (as phenix technically writes in .ccp4, not .mrc format) but not entirely sure… this old thread from the Chimera bb has a potentially useful discussion of the issues, as well as the rationale in Chimera for how to deal with them:

https://www.cgl.ucsf.edu/pipermail/chimera-users/2006-September/000961.html

But in general I think it would be good to check the header of the file so it is displayed as expected by the user.

1 Like
4

Thanks. :slight_smile: 2006 is a bit before my time! Shame no pun intended, it was a good one. :wink:

Completely agree it’s worth checking. Been running into a surprising number of unexpected compatibility woes recently (most not cryo-EM related) even between software produced by the same company!

That said, might be worth mentioning to the PHENIX devs as well (e.g.: “please maintain input axes order”?) as it doesn’t seem like expected behaviour? Although it would explain some of my woes with PHENIX with some maps/models.

1 Like
5

Not that much of a niche case!

I have had many times the need to do this and also twice I observed this behavior. I had to somehow sort it out outside cryoSPARC.

André

1 Like
6

Thanks @olibclarke for bringing up this issue. We made a note.

1 Like