[3dem] troubleshooting help in relion?

[Takanori Nakane]

Hi,

With "skip padding: Yes", you can refine a 1000 px box on 1080 Ti
(I don't remember if 1024 works).

That being said, there is no point using such a large box for initial
model generation. As it performs many iterations, M steps will take
huge amount of time with a large box. For initial model generation,
I recommend down-sampling to 4 - 7 A/px. Low resolution also means that
you have less signal delocalisation due to CTF, so you can use a tighter
box.

Best regards,

Takanori Nakane

On 2019/05/07 20:15, Hamidreza Rahmani wrote:
> We have experience working with relion up too 768^2 box size. I think it 
> will do 1024 too if you skip padding (I might be wrong on this!)
> 
> Best,
> 
> Hamid
> 
> Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986> for 
> Windows 10
> 
> ------------------------------------------------------------------------
> *From:* 3dem <3dem-bounces at ncmir.ucsd.edu> on behalf of Sigworth, 
> Frederick <fred.sigworth at yale.edu>
> *Sent:* Tuesday, May 7, 2019 1:45:54 PM
> *To:* Parent, Kristin
> *Cc:* 3dem at ncmir.ucsd.edu
> *Subject:* Re: [3dem] troubleshooting help in relion?
> The box size awfully big, and it’s an odd number, which Relion won’t 
> allow. At the typical 1A pixel size, your box would be 1099 angstroms in 
> size. Is your particle that huge? Unless you know you will reach sub-2A 
> resolution, your pixel size should be about 1A at the smallest. And many 
> people use the Rescale Particles option in Relion’s particle extraction 
> to make the pixel size larger (and the box size, i.e. number of pixels 
> on a side lower) for initial modeling.  Even for something as large as a 
> ribosome I think resampling to 128 might be good for starters.
> 
> Fred
> 
>> On May 7, 2019, at 12:52 PM, Parent, Kristin <kparent at msu.edu 
>> <mailto:kparent at msu.edu>> wrote:
>>
>> Hello fellow cryo-EM-ers,
>> Has anyone seen this error message before? We are new to relion and 
>> can’t figure out the problem. Box size is 1099. Is that to big?
>> THANKS!
>> Kristin
>> <image001.png>
>> --
>> Kristin N. Parent, PhD
>> JK Billman, Jr., MD Endowed Research Professor
>> Acting Director, MSU Cryo-EM
>> Assistant Professor of Biochemistry andMicrobiology
>> Michigan State University
>> 603 Wilson Rd
>> East Lansing, MI 48824
>> Phone: 517-432-8434
>> https://kparentlab.natsci.msu.edu/ 
>> <https://urldefense.proofpoint.com/v2/url?u=https-3A__nam05.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Fkparentlab.natsci.msu.edu-252F-26data-3D02-257C01-257Cfred.sigworth-2540yale.edu-257C1513c89d7cd240010b0808d6d30c734a-257Cdd8cbebb21394df8b4114e3e87abeb5c-257C0-257C0-257C636928447795109387-26sdata-3D2oIvgjHapT2E5Y-252FgdCIpEMQ4bA60et1SmK-252FtcKsVnmE-253D-26reserved-3D0&d=DwMGaQ&c=MNHwOqQ8N1u91SoMLfIblwuGXKgp50OPUXjl8uRAbak&r=RsBMFMFp5qaziXZ3VLyvhg&m=3Or5mkzVDCEvFQgsl8XdBZ3YQGQU5vdd0SrE_7FRwgU&s=JMcLUeOB9syDWqkx0UsAoqLGHtlvjM5vJIAtssNLSCU&e=>
>> Twitter: @Phage4Lyfe
>> _______________________________________________
>> 3dem mailing list
>> 3dem at ncmir.ucsd.edu <mailto:3dem at ncmir.ucsd.edu>
>> https://nam05.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.ncmir.ucsd.edu%2Fmailman%2Flistinfo%2F3dem&data=02%7C01%7Cfred.sigworth%40yale.edu%7C1513c89d7cd240010b0808d6d30c734a%7Cdd8cbebb21394df8b4114e3e87abeb5c%7C0%7C0%7C636928447795129401&sdata=CmSbZogcNzlhaogvMRUTp7YOU5F4ypSOTA5OH6uOuJU%3D&reserved=0 
>> <https://urldefense.proofpoint.com/v2/url?u=https-3A__nam05.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Fmail.ncmir.ucsd.edu-252Fmailman-252Flistinfo-252F3dem-26amp-3Bdata-3D02-257C01-257Cfred.sigworth-2540yale.edu-257C1513c89d7cd240010b0808d6d30c734a-257Cdd8cbebb21394df8b4114e3e87abeb5c-257C0-257C0-257C636928447795129401-26amp-3Bsdata-3DCmSbZogcNzlhaogvMRUTp7YOU5F4ypSOTA5OH6uOuJU-253D-26amp-3Breserved-3D0&d=DwMGaQ&c=MNHwOqQ8N1u91SoMLfIblwuGXKgp50OPUXjl8uRAbak&r=RsBMFMFp5qaziXZ3VLyvhg&m=3Or5mkzVDCEvFQgsl8XdBZ3YQGQU5vdd0SrE_7FRwgU&s=W7bBuMWkeYwJWQ0IATpM1cJA3LUAO3hX007svaamdmg&e=>
> 
> 
> _______________________________________________
> 3dem mailing list
> 3dem at ncmir.ucsd.edu
> https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem
>