Extract tube particles

[Carsten Sachse]

Hi Kaiming,

In principle, you don't need to extract them because in other operations such as SEGMENTEXAM or SEGMENTREFINE3D you can simply choose your desired subset and the programs will only regard those. Have a look at a previous question in our forum, too: https://groups.google.com/forum/#!topic/emspring/N9uNiPVwpBE

Best wishes,

Carsten

On 14 Feb 2014, at 17:55, kmzhang wrote:

Hi, Carsten,

      I have a question about using Spring.

      When I get a class averages, and I found one of them is quite better, how can I know the average from which "tube particles"  and how to  extract the  " tube particles" ?

     Regards, 

     Kaiming

[Carsten Sachse]

Hi Kaiming,

Sorry, I don't understand your question. You want to trace the best diffracting tube to do an individual FFT, is that right? For this, you actually need to trace the segments of your best classes. You can do this by plotting your selected classes on micrographs using SEGMENTPLOT.

Best wishes,

Carsten 

On 15 Feb 2014, at 17:19, kmzhang  wrote:

Hi, Carsten,

      

     Actually I don't know which the layer lines to be used, so I want to find the best tube to do the FFT to make sure  about that.  Some guys didn't believe the FFT from the averages. So I couldn't go to the next step to 3drefine??Any suggestions?

      By the way, thanks to invite me to join the spring group. 

     Kaiming 

[Carsten Sachse]

Hi Kaiming,

Your class average looks good but we are aware that classification can introduce artefacts into helical class averages. So be careful. The best test to validate your layer lines in the classes is to generate the sum of power spectra using SEGMENTEXAM with the same subset of class members. The outcome must be identical from the power of the class (obtained from SEGCLASSEXAM) vs the sum of the power spectra from the class members (obtained from SEGMENTEXAM). If this is not the case the classification results are not reliable.

Best wishes,

Carsten

PS Please use post reply interface in your browser. Otherwise your messages are not added to the emspring list.

On 16 Feb 2014, at 21:50, kmzhang wrote:

Hi, Carsten,

      Ok,I'll try it.

      Some guys told me maybe the power.hdf(FFT) is not real from the class average. I attached the good_average and enhanced_power.  I can't make sure which layer line to be used?