Dataset Information

NanoTRON: a Picasso module for MLP-based classification of super-resolution data.

ABSTRACT: MOTIVATION:Classification of images is an essential task in higher-level analysis of biological data. By bypassing the diffraction limit of light, super-resolution microscopy opened up a new way to look at molecular details using light microscopy, producing large amounts of data with exquisite spatial detail. Statistical exploration of data usually needs initial classification, which is up to now often performed manually. RESULTS:We introduce nanoTRON, an interactive open-source tool, which allows super-resolution data classification based on image recognition. It extends the software package Picasso with the first deep learning tool with a graphic user interface. AVAILABILITY AND IMPLEMENTATION:nanoTRON is written in Python and freely available under the MIT license as a part of the software collection Picasso on GitHub (http://www.github.com/jungmannlab/picasso). All raw data can be obtained from the authors upon reasonable request. CONTACT:jungmann@biochem.mpg.de. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.

SUBMITTER: Auer A

PROVIDER: S-EPMC7267816 | biostudies-literature | 2020 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

nanoTRON: a Picasso module for MLP-based classification of super-resolution data.

Auer Alexander A Strauss Maximilian T MT Strauss Sebastian S Jungmann Ralf R

Bioinformatics (Oxford, England) 20200601 11

<h4>Motivation</h4>Classification of images is an essential task in higher-level analysis of biological data. By bypassing the diffraction limit of light, super-resolution microscopy opened up a new way to look at molecular details using light microscopy, producing large amounts of data with exquisite spatial detail. Statistical exploration of data usually needs initial classification, which is up to now often performed manually.<h4>Results</h4>We introduce nanoTRON, an interactive open-source t ...[more]

PMID: 32145010

Similar Datasets

Project description:X-ray diffraction plays a pivotal role in the understanding of biological systems by revealing atomic structures of proteins, nucleic acids and their complexes, with much recent interest in very large assemblies like the ribosome. As crystals of such large assemblies often diffract weakly (resolution worse than 4 A), we need methods that work at such low resolution. In macromolecular assemblies, some of the components may be known at high resolution, whereas others are unknown: current refinement methods fail as they require a high-resolution starting structure for the entire complex. Determining the structure of such complexes, which are often of key biological importance, should be possible in principle as the number of independent diffraction intensities at a resolution better than 5 A generally exceeds the number of degrees of freedom. Here we introduce a method that adds specific information from known homologous structures but allows global and local deformations of these homology models. Our approach uses the observation that local protein structure tends to be conserved as sequence and function evolve. Cross-validation with R(free) (the free R-factor) determines the optimum deformation and influence of the homology model. For test cases at 3.5-5 A resolution with known structures at high resolution, our method gives significant improvements over conventional refinement in the model as monitored by coordinate accuracy, the definition of secondary structure and the quality of electron density maps. For re-refinements of a representative set of 19 low-resolution crystal structures from the Protein Data Bank, we find similar improvements. Thus, a structure derived from low-resolution diffraction data can have quality similar to a high-resolution structure. Our method is applicable to the study of weakly diffracting crystals using X-ray micro-diffraction as well as data from new X-ray light sources. Use of homology information is not restricted to X-ray crystallography and cryo-electron microscopy: as optical imaging advances to subnanometre resolution, it can use similar tools.

Project description:DNA double strand breaks (DSB) are the most severe damages in chromatin induced by ionizing radiation. In response to such environmentally determined stress situations, cells have developed repair mechanisms. Although many investigations have contributed to a detailed understanding of repair processes, e.g., homologous recombination repair or non-homologous end-joining, the question is not sufficiently answered, how a cell decides to apply a certain repair process at a certain damage site, since all different repair pathways could simultaneously occur in the same cell nucleus. One of the first processes after DSB induction is phosphorylation of the histone variant H2AX to ?H2AX in the given surroundings of the damaged locus. Since the spatial organization of chromatin is not random, it may be conclusive that the spatial organization of ?H2AX foci is also not random, and rather, contributes to accessibility of special repair proteins to the damaged site, and thus, to the following repair pathway at this given site. The aim of this article is to demonstrate a new approach to analyze repair foci by their topology in order to obtain a cell independent method of categorization. During the last decade, novel super-resolution fluorescence light microscopic techniques have enabled new insights into genome structure and spatial organization on the nano-scale in the order of 10 nm. One of these techniques is single molecule localization microscopy (SMLM) with which the spatial coordinates of single fluorescence molecules can precisely be determined and density and distance distributions can be calculated. This method is an appropriate tool to quantify complex changes of chromatin and to describe repair foci on the single molecule level. Based on the pointillist information obtained by SMLM from specifically labeled heterochromatin and ?H2AX foci reflecting the chromatin morphology and repair foci topology, we have developed a new analytical methodology of foci or foci cluster characterization, respectively, by means of persistence homology. This method allows, for the first time, a cell independent comparison of two point distributions (here the point distributions of two ?H2AX clusters) with each other of a selected ensample and to give a mathematical measure of their similarity. In order to demonstrate the feasibility of this approach, cells were irradiated by low LET (linear energy transfer) radiation with different doses and the heterochromatin and ?H2AX foci were fluorescently labeled by antibodies for SMLM. By means of our new analysis method, we were able to show that the topology of clusters of ?H2AX foci can be categorized depending on the distance to heterochromatin. This method opens up new possibilities to categorize spatial organization of point patterns by parameterization of topological similarity.

Dataset Information

NanoTRON: a Picasso module for MLP-based classification of super-resolution data.

Publications

nanoTRON: a Picasso module for MLP-based classification of super-resolution data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets