scholarly journals X-Ray Scattering Image Classification Using Deep Learning

Author(s):  
Boyu Wang ◽  
Kevin Yager ◽  
Dantong Yu ◽  
Minh Hoai
2019 ◽  
Author(s):  
Hao He ◽  
Can Liu ◽  
Haiguang Liu

AbstractWe present an algorithm based on a deep learning method for model reconstruction from small angle X-ray scattering (SAXS) data. An auto-encoder for protein 3D models was trained to compress 3D shape information into vectors of a 200-dimensional latent space, and the vectors are optimized using genetic algorithms to build 3D models that are consistent with the scattering data. The algorithm was implemented using Python with the TensorFlow framework and tested with experimental data, demonstrating capacity and robustness of accurate model reconstruction even without using prior model size information.SynopsisA deep learning method based on the auto-encoder framework for model reconstruction from small angle scattering data


Author(s):  
Eva-Maria Mandelkow ◽  
Eckhard Mandelkow ◽  
Joan Bordas

When a solution of microtubule protein is changed from non-polymerising to polymerising conditions (e.g. by temperature jump or mixing with GTP) there is a series of structural transitions preceding microtubule growth. These have been detected by time-resolved X-ray scattering using synchrotron radiation, and they may be classified into pre-nucleation and nucleation events. X-ray patterns are good indicators for the average behavior of the particles in solution, but they are difficult to interpret unless additional information on their structure is available. We therefore studied the assembly process by electron microscopy under conditions approaching those of the X-ray experiment. There are two difficulties in the EM approach: One is that the particles important for assembly are usually small and not very regular and therefore tend to be overlooked. Secondly EM specimens require low concentrations which favor disassembly of the particles one wants to observe since there is a dynamic equilibrium between polymers and subunits.


Author(s):  
Eva-Maria Mandelkow ◽  
Ron Milligan

Microtubules form part of the cytoskeleton of eukaryotic cells. They are hollow libers of about 25 nm diameter made up of 13 protofilaments, each of which consists of a chain of heterodimers of α-and β-tubulin. Microtubules can be assembled in vitro at 37°C in the presence of GTP which is hydrolyzed during the reaction, and they are disassembled at 4°C. In contrast to most other polymers microtubules show the behavior of “dynamic instability”, i.e. they can switch between phases of growth and phases of shrinkage, even at an overall steady state [1]. In certain conditions an entire solution can be synchronized, leading to autonomous oscillations in the degree of assembly which can be observed by X-ray scattering (Fig. 1), light scattering, or electron microscopy [2-5]. In addition such solutions are capable of generating spontaneous spatial patterns [6].In an earlier study we have analyzed the structure of microtubules and their cold-induced disassembly by cryo-EM [7]. One result was that disassembly takes place by loss of protofilament fragments (tubulin oligomers) which fray apart at the microtubule ends. We also looked at microtubule oscillations by time-resolved X-ray scattering and proposed a reaction scheme [4] which involves a cyclic interconversion of tubulin, microtubules, and oligomers (Fig. 2). The present study was undertaken to answer two questions: (a) What is the nature of the oscillations as seen by time-resolved cryo-EM? (b) Do microtubules disassemble by fraying protofilament fragments during oscillations at 37°C?


1992 ◽  
Vol 2 (6) ◽  
pp. 899-913 ◽  
Author(s):  
Patrick Davidson ◽  
Elisabeth Dubois-Violette ◽  
Anne-Marie Levelut ◽  
Brigitte Pansu

1996 ◽  
Vol 6 (8) ◽  
pp. 1085-1094 ◽  
Author(s):  
A. Gibaud ◽  
J. Wang ◽  
M. Tolan ◽  
G. Vignaud ◽  
S. K. Sinha

2002 ◽  
Vol 12 (6) ◽  
pp. 385-390 ◽  
Author(s):  
J.-F. Bérar ◽  
L. Blanquart ◽  
N. Boudet ◽  
P. Breugnon ◽  
B. Caillot ◽  
...  

2007 ◽  
Vol 2007 (suppl_26) ◽  
pp. 247-252
Author(s):  
R. Kužel ◽  
L. Nichtová ◽  
Z. Matěj ◽  
D. Heřman ◽  
J. Šicha ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document