Pure noise result after relax symmetry of a virus

jmq...@tamu.edu

unread,

Mar 25, 2019, 6:05:03 PM3/25/19

to jspr

Hi all,

I'm using jspr to do asymmetric reconstruction to see the portal of the virus. The icosahedral refinement reached to 7A using RELION. Then I took the final data file to extract the euler angles and output as a lst file. Then I followed the command:

jspr.py test.lst --sym=c1 --diameter=1024 --apix=3.7 --cpus=10 --iters=50 --iter0 40 --aligner=jalign --alignerOption " --superaligner symmetryRelax:startsym=icos:endsym=c1 --aligner dummy:maskradius=140" --reconstructor j3dr --batchSize 300

test.lst is the file that contains euler angles of each particles. But I ended up with nothing but pure noise. I have a picture showing the result. Is this because I did something wrong or I just failed to align the portal properly? I also tried using different initial models or change the mask diameter. None of them worked. Any suggestion will be appreicated.

Best wishes,

Mengqiu

image.png

Wen Jiang

unread,

Mar 26, 2019, 6:19:50 PM3/26/19

to jspr

Please try icos reconstruction first to make sure the map is correct. I suspect that the problem was caused by different icos view conventions used by Relion and jspr (same as EMAN2). You should use I4 sym in Relion.

jmq...@tamu.edu

unread,

Mar 26, 2019, 6:56:08 PM3/26/19

to jspr

Hi Dr. Jiang,

Thank you for your reply! I did an icosahedral using RELION with I4 symmetry and it gave me a pretty good density map. I have the map picture attached below. But using orientation of that map didn't help. Is there any other possibilities? Thanks.