Question about corrected half-maps

77 views
Skip to first unread message

Vitor Hugo Balasco Serrão

unread,
May 14, 2024, 2:15:31 PM5/14/24
to spis...@googlegroups.com
Hi there,

My apologies for the naive question, but I've been playing with spisonet in some datasets, and what would be the most appropriate way to proceed after it? 
I mean, we have the corrected half-maps, would the only option refine it in Relion or can we use other approaches (loading into cryosparc, trying deep-enhancer, etc)?
I'd like to hear your thoughts and suggestions what is the best way to proceed after using spisonet.

Thank you!

Vitor Hugo Balasco Serrão, Ph.D.
Research Specialist
Biomolecular Cryo-Electron Microscopy Facility
Department of Chemistry and Biochemistry
University California - Santa Cruz
1156 High St - Santa Cruz, CA - 95060

YUNTAO LIU

unread,
May 14, 2024, 2:57:55 PM5/14/24
to Vitor Hugo Balasco Serrão, spis...@googlegroups.com

Hi Vitor Hugo Balasco,

First, the misalignment correction with relion_wrapper is similar to a regular RELION refinement. You can perform any post-processing or continue with anisotropy correction using spisonet.py reconstruct.

Second, if you are using spisonet.py reconstruct, the results are corrected half-maps. You can perform post-processing on these maps using either RELION or cryoSPARC, and you may also manually test different B-factors. Based on my knowledge, sharpening results from RELION and cryoSPARC are similar. Automatic modeling tools like ModelAngelo should work well with spIsoNet-generated maps. Other tools, such as deep enhancers, are also applicable.

The local resolution estimation tools should provide accurate results, but be aware that the FSC for spIsoNet-generated maps is not gold-standard unless you use the --independent argument with spisonet.py reconstruct. Finally, if you plan to deposit spIsoNet-generated half-maps, you may need to include the pre-correction maps as well, even though spIsoNet is "conservative" due to its self-supervised nature.


--
You received this message because you are subscribed to the Google Groups "spIsoNet" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spisonet+u...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/spisonet/CANhJ31DcWtjhgCrow%2BNjYCovnB1vJbT955uxzGXk2edtJfoHwA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.


--
Best Regards,
Yuntao Liu,  Postdoc.

California NanoSystem Institute
University of California Los Angeles

Vitor Hugo Balasco Serrão

unread,
May 14, 2024, 6:25:26 PM5/14/24
to YUNTAO LIU, spis...@googlegroups.com
Thank you, Yuntao. This is very useful and I'll try to follow these steps.

Best,

Vitor Hugo Balasco Serrão, Ph.D.
Research Specialist
Biomolecular Cryo-Electron Microscopy Facility
Department of Chemistry and Biochemistry
University California - Santa Cruz
1156 High St - Santa Cruz, CA - 95060

Reply all
Reply to author
Forward
0 new messages