CaImAn an open source tool for scalable calcium imaging data analysis

eLife, Год журнала: 2019, Номер 8

Опубликована: Янв. 14, 2019

Advances in fluorescence microscopy enable monitoring larger brain areas in-vivo with finer time resolution. The resulting data rates require reproducible analysis pipelines that are reliable, fully automated, and scalable to datasets generated over the course of months. We present CaImAn, an open-source library for calcium imaging analysis. CaImAn provides automatic methods address problems common pre-processing, including motion correction, neural activity identification, registration across different sessions collection. It does this while requiring minimal user intervention, good scalability on computers ranging from laptops high-performance computing clusters. is suitable two-photon one-photon imaging, also enables real-time streaming data. To benchmark performance we collected combined a corpus manual annotations multiple labelers nine mouse datasets. demonstrate achieves near-human detecting locations active neurons.

Язык: Английский

---

Efficient and accurate extraction of in vivo calcium signals from microendoscopic video data

eLife, Год журнала: 2018, Номер 7

Опубликована: Фев. 22, 2018

In vivo calcium imaging through microendoscopic lenses enables of previously inaccessible neuronal populations deep within the brains freely moving animals. However, it is computationally challenging to extract single-neuronal activity from data, because very large background fluctuations and high spatial overlaps intrinsic this recording modality. Here, we describe a new constrained matrix factorization approach accurately separate then demix denoise signals interest. We compared proposed method against previous independent components analysis nonnegative approaches. On both simulated experimental data recorded mice, our substantially improved quality extracted cellular detected more well-isolated neural signals, especially in noisy regimes. These advances can turn significantly enhance statistical power downstream analyses, ultimately improve scientific conclusions derived data.

Язык: Английский

---

Fast online deconvolution of calcium imaging data

PLoS Computational Biology, Год журнала: 2017, Номер 13(3), С. e1005423 - e1005423

Опубликована: Март 14, 2017

Fluorescent calcium indicators are a popular means for observing the spiking activity of large neuronal populations, but extracting each neuron from raw fluorescence imaging data is nontrivial problem. We present fast online active set method to solve this sparse non-negative deconvolution Importantly, algorithm progresses through time series sequentially beginning end, thus enabling real-time estimation neural during session. Our generalization pool adjacent violators (PAVA) isotonic regression and inherits its linear-time computational complexity. gain remarkable increases in processing speed: more than one order magnitude compared currently employed state art convex solvers relying on interior point methods. Unlike these approaches, our can exploit warm starts; therefore optimizing model hyperparameters only requires handful passes data. A minor modification further improve quality inference by imposing constraint minimum spike size. The enables simultaneous $O(10^5)$ traces whole-brain larval zebrafish laptop.

Язык: Английский

---

Fast and sensitive GCaMP calcium indicators for imaging neural populations

Nature, Год журнала: 2023, Номер 615(7954), С. 884 - 891

Опубликована: Март 15, 2023

Abstract Calcium imaging with protein-based indicators 1,2 is widely used to follow neural activity in intact nervous systems, but current protein sensors report at timescales much slower than electrical signalling and are limited by trade-offs between sensitivity kinetics. Here we large-scale screening structure-guided mutagenesis develop optimize several fast sensitive GCaMP-type 3–8 . The resulting ‘jGCaMP8’ sensors, based on the calcium-binding calmodulin a fragment of endothelial nitric oxide synthase, have ultra-fast kinetics (half-rise times 2 ms) highest for reported calcium sensor. jGCaMP8 will allow tracking large populations neurons relevant computation.

Язык: Английский

---

Ultrafast Two-Photon Imaging of a High-Gain Voltage Indicator in Awake Behaving Mice

Cell, Год журнала: 2019, Номер 179(7), С. 1590 - 1608.e23

Опубликована: Дек. 1, 2019

Язык: Английский

---

Long-Term Optical Access to an Estimated One Million Neurons in the Live Mouse Cortex

Cell Reports, Год журнала: 2016, Номер 17(12), С. 3385 - 3394

Опубликована: Дек. 1, 2016

A major technological goal in neuroscience is to enable the interrogation of individual cells across live brain. By creating a curved glass replacement dorsal cranium and surgical methods for its installation, we developed chronic mouse preparation providing optical access an estimated 800,000–1,100,000 neurons surface neocortex. Post-surgical histological studies revealed comparable glial activation as control mice. In behaving mice expressing Ca2+ indicator cortical pyramidal neurons, performed imaging neocortex using epi-fluorescence macroscope that 25,000–50,000 were accessible per multiple focal planes. Two-photon microscopy dendritic morphologies throughout neocortex, allowed time-lapse cells, yielded estimates >1 million by serial tiling. This approach supports variety techniques enables >30 neocortical areas

Язык: Английский

---

Predictive and reactive reward signals conveyed by climbing fiber inputs to cerebellar Purkinje cells

Nature Neuroscience, Год журнала: 2019, Номер 22(6), С. 950 - 962

Опубликована: Апрель 29, 2019

Язык: Английский

---

Robustness of Spike Deconvolution for Neuronal Calcium Imaging

Journal of Neuroscience, Год журнала: 2018, Номер 38(37), С. 7976 - 7985

Опубликована: Авг. 6, 2018

Calcium imaging is a powerful method to record the activity of neural populations in many species, but inferring spike times from calcium signals challenging problem. We compared multiple approaches using datasets with ground truth electrophysiology and found that simple non-negative deconvolution (NND) outperformed all other algorithms on out-of-sample test data. introduce novel benchmark applicable recordings without electrophysiological truth, based correlation responses two stimulus repeats, used this show unconstrained NND also when run “zoomed out” ∼10,000 cell visual cortex mice either sex. Finally, we NND-based methods match performance supervised convolutional networks while avoiding some biases such methods, at much faster running times. therefore recommend spikes be inferred traces because its simplicity, efficiency, accuracy. SIGNIFICANCE STATEMENT The experimental currently allows for largest numbers cells simultaneously two-photon imaging. However, use requires neuronal firing correctly large resulting datasets. Previous studies have claimed complex learning outperform task. Unfortunately, these suffered several problems biases. When repeated analysis, same data correcting problems, simpler inference perform better. Even more importantly, can artifactual structure into trains, which turn lead erroneous scientific conclusions. Of evaluated, an extremely performed best circumstances tested, was run, insensitive parameter choices, making incorrect conclusions less likely.

Язык: Английский

---

In Vivo Imaging of Dentate Gyrus Mossy Cells in Behaving Mice

Neuron, Год журнала: 2017, Номер 93(3), С. 552 - 559.e4

Опубликована: Янв. 26, 2017

Язык: Английский

---

Community-based benchmarking improves spike rate inference from two-photon calcium imaging data

PLoS Computational Biology, Год журнала: 2018, Номер 14(5), С. e1006157 - e1006157

Опубликована: Май 21, 2018

In recent years, two-photon calcium imaging has become a standard tool to probe the function of neural circuits and study computations in neuronal populations. However, acquired signal is only an indirect measurement activity due comparatively slow dynamics fluorescent indicators. Different algorithms for estimating spike rates from noisy measurements have been proposed past, but it open question how far performance can be improved. Here, we report results spikefinder challenge, launched catalyze development new rate inference through crowd-sourcing. We present ten submitted which show improved compared previously evaluated methods. Interestingly, top-performing are based on wide range principles deep networks generative models, yet provide highly correlated estimates activity. The competition shows that benchmark challenges drive algorithmic developments neuroscience.

Язык: Английский

---