[3dem] Add random noise to reconstruction

[Steinar Halldorsson]

Hi Vladan


I have removed overlapping particles but it didn't change the FSC behaviour.


Steinar

________________________________
From: 3dem <3dem-bounces at ncmir.ucsd.edu> on behalf of Vladan Lucic <vladan at biochem.mpg.de>
Sent: 02 November 2018 17:21:04
To: 3dem at ncmir.ucsd.edu
Subject: Re: [3dem] Add random noise to reconstruction


Hi Stelnar,


Perhaps you have double (overlapping) particles, as a result of double picking or translations during refinement for example. A quick way to check for this is to make particle half-sets from different tomograms (split tomograms in two groups and make each half particle set from one group). In this way, double particles will always be in the same particle half-set. If the FSC between these halves goes to 0, it's double particles.


Just a guess, hope it helps


Best,

Vladan


On 11/2/18 5:44 PM, Steinar Halldorsson wrote:

Hi all


I have something which looks like an artefact in my FSC curves from tomography data calculated in Relion. Basically the curves hover around 0.1 instead of reaching 0... And this is not coming from my mask btw. There is quite a lot of information (protein) outside of the mask, you can see clear but slightly smeary density from neighbours in the final maps, and I wonder if this is causing these correlations? One way I have thought to test this is to take my two half maps, somehow add random noise to parts outside of my region of interest and then recalculate my FSC.

Firstly, does this sound like a reasonable thing to try?
Secondly, does anyone have a suggestion of how to actually do the random noise addition outside of my region of interest?

Cheers,

Steinar


---

Steinar Halldorsson

Postdoctorate Fellow

Rosenthal lab

The Francis Crick Institute
London, United Kingdom

The Francis Crick Institute Limited is a registered charity in England and Wales no. 1140062 and a company registered in England and Wales no. 06885462, with its registered office at 1 Midland Road London NW1 1AT



_______________________________________________
3dem mailing list
3dem at ncmir.ucsd.edu<mailto:3dem at ncmir.ucsd.edu>
https://mail.ncmir.ucsd.edu/mailman/listinfo/3dem<https://emea01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.ncmir.ucsd.edu%2Fmailman%2Flistinfo%2F3dem&data=02%7C01%7C%7Ccb937a620d89499489db08d640e79c9e%7C4eed7807ebad415aa7a99170947f4eae%7C0%7C0%7C636767760866341286&sdata=ZGt4aA6fFvTtxlf2mbVxg5eVJzyuhS3LgKiiCH0shTc%3D&reserved=0>





The Francis Crick Institute Limited is a registered charity in England and Wales no. 1140062 and a company registered in England and Wales no. 06885462, with its registered office at 1 Midland Road London NW1 1AT
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.ncmir.ucsd.edu/pipermail/3dem/attachments/20181102/7363db70/attachment.html>