scholarly journals Machine-learning based spectral classification for spectroscopic single-molecule localization microscopy

2019 ◽  
Vol 44 (23) ◽  
pp. 5864 ◽  
Author(s):  
Zheyuan Zhang ◽  
Yang Zhang ◽  
Leslie Ying ◽  
Cheng Sun ◽  
Hao F. Zhang
Keyword(s):  
Machine Learning ◽  
Single Molecule ◽  
Spectral Classification ◽  
Localization Microscopy ◽  
Single Molecule Localization Microscopy

scholarly journals Caveolae and scaffold detection from single molecule localization microscopy data using deep learning

2019 ◽  
Author(s):  
Ismail M. Khater ◽  
Stephane T. Aroca-Ouellette ◽  
Fanrui Meng ◽  
Ivan Robert Nabi ◽  
Ghassan Hamarneh
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Random Forest ◽  
Single Molecule ◽  
Adaptor Protein ◽  
Point Clouds ◽  
Learning Models ◽  
Localization Microscopy ◽  
Biological Structures ◽  
Single Molecule Localization Microscopy

AbstractCaveolae are plasma membrane invaginations whose formation requires caveolin-1 (Cav1), the adaptor protein polymerase I, and the transcript release factor (PTRF or CAVIN1). Caveolae have an important role in cell functioning, signaling, and disease. In the absence of CAVIN1/PTRF, Cav1 forms non-caveolar membrane domains called scaffolds. In this work, we train machine learning models to automatically distinguish between caveolae and scaffolds from single molecule localization microscopy (SMLM) data. We apply machine learning algorithms to discriminate biological structures from SMLM data. Our work is the first that is leveraging machine learning approaches (including deep learning models) to automatically identifying biological structures from SMLM data. In particular, we develop and compare three binary classification methods to identify whether or not a given 3D cluster of Cav1 proteins is a caveolae. The first uses a random forest classifier applied to 28 hand-crafted/designed features, the second uses a convolutional neural net (CNN) applied to a projection of the point clouds onto three planes, and the third uses a PointNet model, a recent development that can directly take point clouds as its input. We validate our methods on a dataset of super-resolution microscopy images of PC3 prostate cancer cells labeled for Cav1. Specifically, we have images from two cell populations: 10 PC3 and 10 CAVIN1/PTRF-transfected PC3 cells (PC3-PTRF cells) that form caveolae. We obtained a balanced set of 1714 different cellular structures. Our results show that both the random forest on hand-designed features and the deep learning approach achieve high accuracy in distinguishing the intrinsic features of the caveolae and non-caveolae biological structures. More specifically, both random forest and deep CNN classifiers achieve classification accuracy reaching 94% on our test set, while the PointNet model only reached 83% accuracy. We also discuss the pros and cons of the different approaches.

scholarly journals Identification of caveolin-1 domain signatures via machine learning and graphlet analysis of single-molecule super-resolution data

2019 ◽  
Vol 35 (18) ◽  
pp. 3468-3475 ◽  
Author(s):  
Ismail M Khater ◽  
Fanrui Meng ◽  
Ivan Robert Nabi ◽  
Ghassan Hamarneh
Keyword(s):  
Prostate Cancer ◽  
Machine Learning ◽  
Cancer Cells ◽  
Single Molecule ◽  
Classification Accuracy ◽  
Supplementary Information ◽  
Prostate Cancer Cells ◽  
Localization Microscopy ◽  
Caveolin 1 ◽  
Single Molecule Localization Microscopy

Abstract Motivation Network analysis and unsupervised machine learning processing of single-molecule localization microscopy of caveolin-1 (Cav1) antibody labeling of prostate cancer cells identified biosignatures and structures for caveolae and three distinct non-caveolar scaffolds (S1A, S1B and S2). To obtain further insight into low-level molecular interactions within these different structural domains, we now introduce graphlet decomposition over a range of proximity thresholds and show that frequency of different subgraph (k = 4 nodes) patterns for machine learning approaches (classification, identification, automatic labeling, etc.) effectively distinguishes caveolae and scaffold blobs. Results Caveolae formation requires both Cav1 and the adaptor protein CAVIN1 (also called PTRF). As a supervised learning approach, we applied a wide-field CAVIN1/PTRF mask to CAVIN1/PTRF-transfected PC3 prostate cancer cells and used the random forest classifier to classify blobs based on graphlet frequency distribution (GFD). GFD of CAVIN1/PTRF-positive (PTRF+) and -negative Cav1 clusters showed poor classification accuracy that was significantly improved by stratifying the PTRF+ clusters by either number of localizations or volume. Low classification accuracy (<50%) of large PTRF+ clusters and caveolae blobs identified by unsupervised learning suggests that their GFD is specific to caveolae. High classification accuracy for small PTRF+ clusters and caveolae blobs argues that CAVIN1/PTRF associates not only with caveolae but also non-caveolar scaffolds. At low proximity thresholds (50–100 nm), the caveolae groups showed reduced frequency of highly connected graphlets and increased frequency of completely disconnected graphlets. GFD analysis of single-molecule localization microscopy Cav1 clusters defines changes in structural organization in caveolae and scaffolds independent of association with CAVIN1/PTRF. Supplementary information Supplementary data are available at Bioinformatics online.

Single-Molecule Imaging of Active Mitochondrial Nitroreductases using a Photo-Crosslinking Fluorescent Sensor

2019 ◽  
Author(s):  
Zacharias Thiel ◽  
Pablo Rivera-Fuentes
Keyword(s):  
Subcellular Localization ◽  
Fluorescent Probe ◽  
Single Molecule ◽  
Mammalian Cells ◽  
Fluorescent Sensor ◽  
Biological Functions ◽  
Localization Microscopy ◽  
Drug Activation ◽  
Nitroreductase Activity ◽  
Single Molecule Localization Microscopy

Many biomacromolecules are known to cluster in microdomains with specific subcellular localization. In the case of enzymes, this clustering greatly defines their biological functions. Nitroreductases are enzymes capable of reducing nitro groups to amines and play a role in detoxification and pro-drug activation. Although nitroreductase activity has been detected in mammalian cells, the subcellular localization of this activity remains incompletely characterized. Here, we report a fluorescent probe that enables super-resolved imaging of pools of nitroreductase activity within mitochondria. This probe is activated sequentially by nitroreductases and light to give a photo-crosslinked adduct of active enzymes. In combination with a general photoactivatable marker of mitochondria, we performed two-color, threedimensional, single-molecule localization microscopy. These experiments allowed us to image the sub-mitochondrial organization of microdomains of nitroreductase activity.<br>

Single-Molecule Imaging of Active Mitochondrial Nitroreductases using a Photo-Crosslinking Fluorescent Sensor

2019 ◽  
Author(s):  
Zacharias Thiel ◽  
Pablo Rivera-Fuentes
Keyword(s):  
Subcellular Localization ◽  
Fluorescent Probe ◽  
Single Molecule ◽  
Mammalian Cells ◽  
Fluorescent Sensor ◽  
Biological Functions ◽  
Localization Microscopy ◽  
Drug Activation ◽  
Nitroreductase Activity ◽  
Single Molecule Localization Microscopy

Many biomacromolecules are known to cluster in microdomains with specific subcellular localization. In the case of enzymes, this clustering greatly defines their biological functions. Nitroreductases are enzymes capable of reducing nitro groups to amines and play a role in detoxification and pro-drug activation. Although nitroreductase activity has been detected in mammalian cells, the subcellular localization of this activity remains incompletely characterized. Here, we report a fluorescent probe that enables super-resolved imaging of pools of nitroreductase activity within mitochondria. This probe is activated sequentially by nitroreductases and light to give a photo-crosslinked adduct of active enzymes. In combination with a general photoactivatable marker of mitochondria, we performed two-color, threedimensional, single-molecule localization microscopy. These experiments allowed us to image the sub-mitochondrial organization of microdomains of nitroreductase activity.<br>

scholarly journals Single-molecule localization microscopy

2021 ◽  
Vol 1 (1) ◽  
Author(s):  
Mickaël Lelek ◽  
Melina T. Gyparaki ◽  
Gerti Beliu ◽  
Florian Schueder ◽  
Juliette Griffié ◽  
...  
Keyword(s):  
Single Molecule ◽  
Localization Microscopy ◽  
Single Molecule Localization Microscopy

scholarly journals Three-dimensional total-internal reflection fluorescence nanoscopy with nanometric axial resolution by photometric localization of single molecules

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
Alan M. Szalai ◽  
Bruno Siarry ◽  
Jerónimo Lukin ◽  
David J. Williamson ◽  
Nicolás Unsain ◽  
...  
Keyword(s):  
Single Molecule ◽  
Total Internal Reflection ◽  
Single Molecules ◽  
Fluorescence Microscope ◽  
Internal Reflection ◽  
Total Internal Reflection Fluorescence ◽  
Axial Position ◽  
Localization Microscopy ◽  
Localization Precision ◽  
Single Molecule Localization Microscopy

AbstractSingle-molecule localization microscopy enables far-field imaging with lateral resolution in the range of 10 to 20 nanometres, exploiting the fact that the centre position of a single-molecule’s image can be determined with much higher accuracy than the size of that image itself. However, attaining the same level of resolution in the axial (third) dimension remains challenging. Here, we present Supercritical Illumination Microscopy Photometric z-Localization with Enhanced Resolution (SIMPLER), a photometric method to decode the axial position of single molecules in a total internal reflection fluorescence microscope. SIMPLER requires no hardware modification whatsoever to a conventional total internal reflection fluorescence microscope and complements any 2D single-molecule localization microscopy method to deliver 3D images with nearly isotropic nanometric resolution. Performance examples include SIMPLER-direct stochastic optical reconstruction microscopy images of the nuclear pore complex with sub-20 nm axial localization precision and visualization of microtubule cross-sections through SIMPLER-DNA points accumulation for imaging in nanoscale topography with sub-10 nm axial localization precision.

scholarly journals Phasor based single-molecule localization microscopy in 3D (pSMLM-3D): An algorithm for MHz localization rates using standard CPUs

2018 ◽  
Vol 148 (12) ◽  
pp. 123311 ◽  
Author(s):  
Koen J. A. Martens ◽  
Arjen N. Bader ◽  
Sander Baas ◽  
Bernd Rieger ◽  
Johannes Hohlbein
Keyword(s):  
Single Molecule ◽  
Localization Microscopy ◽  
Single Molecule Localization Microscopy
Sign in / Sign up

Export Citation Format

Share Document