Hi Roberto,

you have an example where d99 does indeed depend on masking: there’s a lot of high-resolution noise around the molecule.

The fact that d_fsc(0.143) is better (2.83 Å) than d99 (4.07 Å) likely indicates that the map terms in Fourier space within the resolution range of 2.8–4.0 Å are similar between the half-maps but do not contain structural information.

We hadn’t seen such an example when we were writing that paper, which led us to the conclusion you’re referring to. This might be because, back in 2018 -- not too far from the "resolution revolution" and almost "prehistoric" today given the rapid growth of cryo-EM -- our relatively small set of test cases from the database did not include such an example.

Would it be possible for you to share the map and model so we can take a closer look?

Thanks!
Pavel

Dear Phenix people,

In "New tools for the analysis and validation of cryo-EM maps and atomic models” (ActaD 2018), it is reported that "map sharpening can affect d99 (Fig. 9g) but it is rather insensitive to masking (Fig. 9h).”

I have a case where d99 increases a lot with masking, as shown in the screenshot below (from Mtriage). On the other hand, FSC decreases.

What does this mean? Can it be attributed to the fact that there is a strong scatterer in the molecule (a heavy atom) that strongly influences the intensity of the map?

Are there other possible explanations (preferential orientation for example)?

What is the best way to handle such maps?

Thank you for the attention.

Best,

Roberto. 

Roberto Battistutta
Associate Professor
Department of Chemical Sciences
University of Padua
via Marzolo 1, 35131 Padova - ITALY
tel. +39.049.827.5262
fax. +39.049.827.5829
[email protected]

_______________________________________________
phenixbb mailing list -- [email protected]
To unsubscribe send an email to [email protected]
Unsubscribe: phenixbb-leave@%(host_name)s