Deep learning enables fast and dense single-molecule localization with high accuracy

Speiser, Artur; Müller, Lucas-Raphael; Matti, Ulf; Obara, Christopher J.; Legant, Wesley R.; Kreshuk, Anna; Macke, Jakob H.; Ries, Jonas; Turaga, Srinivas C.

doi:10.1101/2020.10.26.355164

Cited by 34 publications

(58 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Next, we demonstrated analysis of extended structures with LocMoFit on the example of immunolabeled microtubules (Fig. 2g-j, original data from Speiser et al 28 ). The apparent diameter of the ring is of interest, because it directly informs on the linkage error induced by the primary and secondary antibodies 29 in indirect immunolabeling.…”

Section: Resultsmentioning

confidence: 99%

Maximum-likelihood model fitting for quantitative analysis of SMLM data

Hoess

Tschanz

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Quantitative analysis is an important part of any single-molecule localization microscopy (SMLM) data analysis workflow to extract biological insights from the coordinates of the single fluorophores, but current approaches are restricted to simple geometries or do not work on heterogenous structures. Here, we present LocMoFit (Localization Model Fit), an open-source framework to fit an arbitrary model directly to the localization coordinates in SMLM data. Using maximum likelihood estimation, this tool extracts the most likely parameters for a given model that best describe the data, and can select the most likely model from alternative models. We demonstrate the versatility of LocMoFit by measuring precise dimensions of the nuclear pore complex and microtubules. We also use LocMoFit to assemble static and dynamic multi-color protein density maps from thousands of snapshots. In case an underlying geometry cannot be postulated, LocMoFit can perform single-particle averaging of super-resolution structures without any assumption about geometry or symmetry. We provide extensive simulation and visualization routines to validate the robustness of LocMoFit and tutorials based on example data to enable any user to increase the information content they can extract from their SMLM data.

show abstract

Section: Resultsmentioning

confidence: 99%

Maximum-likelihood model fitting for quantitative analysis of SMLM data

Hoess

Tschanz

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, CARE has been trained to resolve sub-diffraction structures in low-SNR brightfield microscopy images using synthetically generated super-resolution data ( Weigert et al, 2018 ). More recently, the DECODE method ( Speiser et al, 2020 preprint) uses a U-net architecture to address the related challenge of computationally increasing resolution in the context of single-molecule localization microscopy. The U-net model takes into account multiple image frames, as well as their temporal context.…”

Section: Deep Learning For Bioimage Analysismentioning

confidence: 99%

Deep learning for bioimage analysis in developmental biology

et al. 2021

View full text Add to dashboard Cite

Deep learning has transformed the way large and complex image datasets can be processed, reshaping what is possible in bioimage analysis. As the complexity and size of bioimage data continues to grow, this new analysis paradigm is becoming increasingly ubiquitous. In this Review, we begin by introducing the concepts needed for beginners to understand deep learning. We then review how deep learning has impacted bioimage analysis and explore the open-source resources available to integrate it into a research project. Finally, we discuss the future of deep learning applied to cell and developmental biology. We analyze how state-of-the-art methodologies have the potential to transform our understanding of biological systems through new image-based analysis and modelling that integrate multimodal inputs in space and time.

show abstract

“…In recent years a number of open source deep learning platforms have been made available [13][14][15][16][17][18][19][20][21][22] that require minimal know-how on the user side. At the same time, methods have been developed to reduce background 23,24 and to accelerate image processing in superresolution microscopy [25][26][27] .…”

Section: Introductionmentioning

confidence: 99%

Precise measurement of nanoscopic septin ring structures in deep learning-assisted quantitative superresolution microscopy

Zehtabian

Müller

Goisser

et al. 2021

Preprint

View full text Add to dashboard Cite

The combination of image analysis and fluorescence superresolution microscopy methods allows for unprecedented insight into the organization of macromolecular assemblies in cells. Advances in deep learning-based object recognition enables the automated processing of large amounts of data, resulting in high accuracy through averaging. However, while the analysis of highly symmetric structures of constant size allows for a resolution approaching the dimensions of structural biology, deep learning methods are prone to different forms of bias. A biased recognition of structures may prohibit the development of readouts for processes that involve significant changes in size or shape of amorphous macromolecular complexes. What is required to overcome this problem is a detailed investigation of potential sources of bias and the rigorous testing of trained models using real or simulated data covering a wide dynamic range of possible results. Here we combine single molecule localization-based superresolution microscopy of septin ring structures with the training of several different deep learning models for a quantitative investigation of bias resulting from different training approaches and finally quantitative changes in septin ring structures. We find that trade-off exists between measurement accuracy and the dynamic range of recognized phenotypes. Using our trained models, we furthermore find that septin ring size can be explained by the number of subunits they are assembled from alone. Our work provides a new experimental system for the investigation of septin polymerization.

show abstract

Deep learning enables fast and dense single-molecule localization with high accuracy

Cited by 34 publications

References 38 publications

Maximum-likelihood model fitting for quantitative analysis of SMLM data

Maximum-likelihood model fitting for quantitative analysis of SMLM data

Deep learning for bioimage analysis in developmental biology

Precise measurement of nanoscopic septin ring structures in deep learning-assisted quantitative superresolution microscopy

Contact Info

Product

Resources

About