1
|
Sgouralis I, Xu LWQ, Jalihal AP, Kilic Z, Walter NG, Pressé S. BNP-Track: a framework for superresolved tracking. Nat Methods 2024:10.1038/s41592-024-02349-9. [PMID: 39039336 DOI: 10.1038/s41592-024-02349-9] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/06/2023] [Accepted: 06/03/2024] [Indexed: 07/24/2024]
Abstract
Superresolution tools, such as PALM and STORM, provide nanoscale localization accuracy by relying on rare photophysical events, limiting these methods to static samples. By contrast, here, we extend superresolution to dynamics without relying on photodynamics by simultaneously determining emitter numbers and their tracks (localization and linking) with the same localization accuracy per frame as widefield superresolution on immobilized emitters under similar imaging conditions (≈50 nm). We demonstrate our Bayesian nonparametric track (BNP-Track) framework on both in cellulo and synthetic data. BNP-Track develops a joint (posterior) distribution that learns and quantifies uncertainty over emitter numbers and their associated tracks propagated from shot noise, camera artifacts, pixelation, background and out-of-focus motion. In doing so, we integrate spatiotemporal information into our distribution, which is otherwise compromised by modularly determining emitter numbers and localizing and linking emitter positions across frames. For this reason, BNP-Track remains accurate in crowding regimens beyond those accessible to other single-particle tracking tools.
Collapse
Affiliation(s)
- Ioannis Sgouralis
- Department of Mathematics, University of Tennessee, Knoxville, TN, USA
| | - Lance W Q Xu
- Center for Biological Physics, Arizona State University, Tempe, AZ, USA
- Department of Physics, Arizona State University, Tempe, AZ, USA
| | - Ameya P Jalihal
- Department of Cell Biology, Duke University, Durham, NC, USA
| | - Zeliha Kilic
- Department of Structural Biology, St. Jude Children's Research Hospital, Memphis, TN, USA
| | - Nils G Walter
- Single Molecule Analysis Group and Center for RNA Biomedicine, Department of Chemistry, University of Michigan, Ann Arbor, MI, USA
| | - Steve Pressé
- Center for Biological Physics, Arizona State University, Tempe, AZ, USA.
- Department of Physics, Arizona State University, Tempe, AZ, USA.
- School of Molecular Sciences, Arizona State University, Tempe, AZ, USA.
| |
Collapse
|
2
|
Xu LW, Sgouralis I, Kilic Z, Pressé S. BNP-Track: A framework for multi-particle superresolved tracking. BIORXIV : THE PREPRINT SERVER FOR BIOLOGY 2023:2023.04.03.535440. [PMID: 37066179 PMCID: PMC10104013 DOI: 10.1101/2023.04.03.535440] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 04/18/2023]
Abstract
When tracking fluorescently labeled molecules (termed "emitters") under widefield microscopes, point spread function overlap of neighboring molecules is inevitable in both dilute and especially crowded environments. In such cases, superresolution methods leveraging rare photophysical events to distinguish static targets nearby in space introduce temporal delays that compromise tracking. As we have shown in a companion manuscript, for dynamic targets, information on neighboring fluorescent molecules is encoded as spatial intensity correlations across pixels and temporal correlations in intensity patterns across time frames. We then demonstrated how we used all spatiotemporal correlations encoded in the data to achieve superresolved tracking. That is, we showed the results of full posterior inference over both the number of emitters and their associated tracks simultaneously and self-consistently through Bayesian nonparametrics. In this companion manuscript we focus on testing the robustness of our tracking tool, BNP-Track, across sets of parameter regimes and compare BNP-Track to competing tracking methods in the spirit of a prior Nature Methods tracking competition. We explore additional features of BNP-Track including how a stochastic treatment of background yields greater accuracy in emitter number determination and how BNP-Track corrects for point spread function blur (or "aliasing") introduced by intraframe motion in addition to propagating error originating from myriad sources (such as criss-crossing tracks, out-of-focus particles, pixelation, shot and camera artefact, stochastic background) in posterior inference over emitter numbers and their associated tracks. While head-to-head comparison with other tracking methods is not possible (as competitors cannot simultaneously learn molecule numbers and associated tracks), we can give competing methods some advantages in order to perform approximate head-to-head comparison. We show that even under such optimistic scenarios, BNP-Track is capable of tracking multiple diffraction-limited point emitters conventional tracking methods cannot resolve thereby extending the superresolution paradigm to dynamical targets.
Collapse
Affiliation(s)
- Lance W.Q. Xu
- Center for Biological Physics, Department of Physics, Arizona State University, Tempe, AZ 85287, USA
- Department of Physics, Arizona State University, Tempe, AZ 85287, USA
| | - Ioannis Sgouralis
- Department of Mathematics, University of Tennessee, Knoxville, TN 37996, USA
| | - Zeliha Kilic
- Single-Molecule Imaging Center, Department of Structural Biology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA
| | - Steve Pressé
- Center for Biological Physics, Department of Physics, Arizona State University, Tempe, AZ 85287, USA
- Department of Physics, Arizona State University, Tempe, AZ 85287, USA
- School of Molecular Science, Arizona State University, Tempe, AZ 85287, USA
| |
Collapse
|
3
|
Zhu Y, Wang M, Yin X, Zhang J, Meijering E, Hu J. Deep Learning in Diverse Intelligent Sensor Based Systems. SENSORS (BASEL, SWITZERLAND) 2022; 23:s23010062. [PMID: 36616657 PMCID: PMC9823653 DOI: 10.3390/s23010062] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/02/2022] [Revised: 12/06/2022] [Accepted: 12/14/2022] [Indexed: 05/27/2023]
Abstract
Deep learning has become a predominant method for solving data analysis problems in virtually all fields of science and engineering. The increasing complexity and the large volume of data collected by diverse sensor systems have spurred the development of deep learning methods and have fundamentally transformed the way the data are acquired, processed, analyzed, and interpreted. With the rapid development of deep learning technology and its ever-increasing range of successful applications across diverse sensor systems, there is an urgent need to provide a comprehensive investigation of deep learning in this domain from a holistic view. This survey paper aims to contribute to this by systematically investigating deep learning models/methods and their applications across diverse sensor systems. It also provides a comprehensive summary of deep learning implementation tips and links to tutorials, open-source codes, and pretrained models, which can serve as an excellent self-contained reference for deep learning practitioners and those seeking to innovate deep learning in this space. In addition, this paper provides insights into research topics in diverse sensor systems where deep learning has not yet been well-developed, and highlights challenges and future opportunities. This survey serves as a catalyst to accelerate the application and transformation of deep learning in diverse sensor systems.
Collapse
Affiliation(s)
- Yanming Zhu
- School of Computer Science and Engineering, University of New South Wales, Sydney, NSW 2052, Australia
| | - Min Wang
- School of Engineering and Information Technology, University of New South Wales, Canberra, ACT 2612, Australia
| | - Xuefei Yin
- School of Engineering and Information Technology, University of New South Wales, Canberra, ACT 2612, Australia
| | - Jue Zhang
- School of Engineering and Information Technology, University of New South Wales, Canberra, ACT 2612, Australia
| | - Erik Meijering
- School of Computer Science and Engineering, University of New South Wales, Sydney, NSW 2052, Australia
| | - Jiankun Hu
- School of Engineering and Information Technology, University of New South Wales, Canberra, ACT 2612, Australia
| |
Collapse
|
4
|
Abstract
In this issue of Biomedical Journal we encounter the chemokine superfamily and its clinical potential. The time course from 56 days zero COVID-19 to a resurgence in cases is presented, as well as a possible solution to overcome rejection in vascularized composite allotransplantation. We are shown the opportunity deep learning (DL) offers in the case of tracking single cells and particles, and also use of DL to bring all hands on deck to counter the current challenge of the COVID-19 pandemic. This issue contains articles about the effect of low energy shock waves in cystitis; the negative effect of high fructose on aortic valve stenosis; a study about the outcome of fecal microbiota transplantation in case of refractory Clostridioides difficile infection; a novel long non-coding RNA that could serve in treating triple-negative breast cancer; the benefits of acupressure in patients with restless leg syndrome; and Filamin A mutations in abnormal neuronal migration development. Finally, a link between jaw surgery and the psychological impact on the patient is explored; a method presented that allows identification of cervical characteristics associated with difficult embryo transfer; and a letter suggesting new parameters to evaluate the use of bone-substitute augmentation in the treatment of osteoporotic intertrochanteric fractures.
Collapse
|
5
|
Del Giudice F, Barnes C. Rapid Temperature-Dependent Rheological Measurements of Non-Newtonian Solutions Using a Machine-Learning Aided Microfluidic Rheometer. Anal Chem 2022; 94:3617-3628. [PMID: 35167252 DOI: 10.1021/acs.analchem.1c05208] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/29/2022]
Abstract
Biofluids such as synovial fluid, blood plasma, and saliva contain several proteins which impart non-Newtonian properties to the biofluids. The concentration of such protein macromolecules in biofluids is regarded as an important biomarker for the diagnosis of several health conditions, including cardiovascular disorders, joint quality, and Alzheimer's. Existing technologies for the measurements of macromolecules in biofluids are limited; they require a long turnaround time, or require complex protocols, thus calling for alternative, more suitable, methodologies aimed at such measurements. According to the well-established relations for polymer solutions, the concentration of macromolecules in solutions can also be derived via measurement of rheological properties such as shear-viscosity and the longest relaxation time. We here introduce a microfluidic rheometer for rapid simultaneous measurement of shear viscosity and longest relaxation time of non-Newtonian solutions at different temperatures. At variance with previous technologies, our microfluidic rheometer provides a very short turnaround time of around 2 min or less thanks to the implementation of a machine-learning algorithm. We validated our platform on several aqueous solutions of poly(ethylene oxide). We also performed measurements on hyaluronic acid solutions in the clinical range for joint grade assessment. We observed monotonic behavior with the concentration for both rheological properties, thus speculating on their use as potential rheo-markers, i.e., rheological biomarkers, across several disease states.
Collapse
Affiliation(s)
- Francesco Del Giudice
- Department of Chemical Engineering, Faculty of Science and Engineering, School of Engineering and Applied Science, Swansea University Fabian Way, Swansea, SA1 8EN, United Kingdom
| | - Claire Barnes
- Department of Biomedical Engineering, Faculty of Science and Engineering, School of Engineering and Applied Science, Swansea University Fabian Way, Swansea, SA1 8EN, United Kingdom
| |
Collapse
|