From: Jose M. de la R. T. <del...@gm...> - 2017-02-03 10:32:19
|
Dear Luigi, Thanks for providing feedback to us. Regarding your specific issues with picking, what do you mean by "after providing the references" in Xmipp and Eman pickers? To the best of my knowledge, both programs generate the references on-the-fly, after the selection of some particles manually. So, there is not need to provide pre-computed references such as 2D class averages. In the case of EMAN2-Box, both the manual/supervised steps are interwinned in a very interactive way. On the other hand, in Xmipp3 picker, those steps are separated in two phases. You first need to picking manually some micrographs entirely (or a well-defined box in one micrograph is its contains many particles)...and then switch to "Supervised" mode, where the algorithm will propose you some particle candidates and you can correct it (either removing bad proposed particles or adding missing one). The algorithm will "learn" from that and then in another protocol you can launch the automatic picking based on that training. Anyway, if you provide a few micrographs we could try to help about the parameters selection. Hope this helps, Jose Miguel Scipion Team ----------------------------------------- My name is Luigi De Colibus and I’m Post Doc at the Division of Structural Biology of Oxford University. I’m writing you because I’m trying to run automated particle picking programs on my micrographs trough SCIPION. Surprisingly, despite the fact my virus particle is very big and evident on the micrographs, neither Xmipp particle picking program nor Boxer in EMAN2 is able to pick correctly the particles after providing the reference. I guess my settings could be wrong and trying different values has not helped so far. It would be great if you could point me in the right direction. I can send you one of my micrographs if you need to do a test run. Thanks a lot in advance for your help! Best Regards Reply to: luigi at strubi.ox.ac.uk |