Main Menu (Mobile)- Block

Main Menu - Block

janelia7_blocks-janelia7_secondary_menu | block
More in this page
janelia7_blocks-janelia7_fake_breadcrumb | block
Turaga Lab / Publications
general_search_page-panel_pane_1 | views_panes

21 Publications

Showing 1-10 of 21 results
05/28/18 | Discrete flow posteriors for variational inference in discrete dynamical systems.
Aitchison L, Adam V, Turaga SC
arXiv. 2018 May 28:1805.10958

Each training step for a variational autoencoder (VAE) requires us to sample from the approximate posterior, so we usually choose simple (e.g. factorised) approximate posteriors in which sampling is an efficient computation that fully exploits GPU parallelism. However, such simple approximate posteriors are often insufficient, as they eliminate statistical dependencies in the posterior. While it is possible to use normalizing flow approximate posteriors for continuous latents, some problems have discrete latents and strong statistical dependencies. The most natural approach to model these dependencies is an autoregressive distribution, but sampling from such distributions is inherently sequential and thus slow. We develop a fast, parallel sampling procedure for autoregressive distributions based on fixed-point iterations which enables efficient and accurate variational inference in discrete state-space latent variable dynamical systems. To optimize the variational bound, we considered two ways to evaluate probabilities: inserting the relaxed samples directly into the pmf for the discrete distribution, or converting to continuous logistic latent variables and interpreting the K-step fixed-point iterations as a normalizing flow. We found that converting to continuous latent variables gave considerable additional scope for mismatch between the true and approximate posteriors, which resulted in biased inferences, we thus used the former approach. Using our fast sampling procedure, we were able to realize the benefits of correlated posteriors, including accurate uncertainty estimates for one cell, and accurate connectivity estimates for multiple cells, in an order of magnitude less time.

View Publication Page
05/21/18 | Community-based benchmarking improves spike inference from two-photon calcium imaging data.
Berens P, Freeman J, Deneux T, Chenkov N, McColgan T, Speiser A, Macke JH, Turaga SC, Mineault P, Rupprecht P, Gerhard S, Friedrich RW, Friedrich J, Paninski L, Pachitariu M, Harris KD, Bolte B, Machado TA, Ringach D, etal
PLoS Computational Biology. 2018 May 21;14(5):e1006157. doi: 10.1371/journal.pcbi.1006157

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

View Publication Page
04/04/18 | Opportunities and obstacles for deep learning in biology and medicine.
Ching T, Himmelstein DS, Beaulieu-Jones BK, Kalinin AA, Do BT, Way GP, Ferrero E, Agapow P, Zietz M, Hoffman MM, Xie W, Rosen GL, Lengerich BJ, Israeli J, Lanchantin J, Woloszynek S, Carpenter AE, Shrikumar A, Xu J, Cofer EM
Journal of The Royal Society Interface. 2018 Apr 4:. doi: 10.1098/rsif.2017.0387

Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems—patient classification, fundamental biological processes and treatment of patients—and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

View Publication Page
12/04/17 | Extracting low-dimensional dynamics from multiple large-scale neural population recordings by learning to predict correlations.
Nonnenmacher M, Turaga SC, Macke JH
31st Conference on Neural Information Processing Systems (NIPS 2017). 2017 Dec 04:

A powerful approach for understanding neural population dynamics is to extract low-dimensional trajectories from population recordings using dimensionality reduction methods. Current approaches for dimensionality reduction on neural data are limited to single population recordings, and can not identify dynamics embedded across multiple measurements. We propose an approach for extracting low-dimensional dynamics from multiple, sequential recordings. Our algorithm scales to data comprising millions of observed dimensions, making it possible to access dynamics distributed across large populations or multiple brain areas. Building on subspace-identification approaches for dynamical systems, we perform parameter estimation by minimizing a moment-matching objective using a scalable stochastic gradient descent algorithm: The model is optimized to predict temporal covariations across neurons and across time. We show how this approach naturally handles missing data and multiple partial recordings, and can identify dynamics and predict correlations even in the presence of severe subsampling and small overlap between recordings. We demonstrate the effectiveness of the approach both on simulated data and a whole-brain larval zebrafish imaging dataset. 

View Publication Page
12/04/17 | Fast amortized inference of neural activity from calcium imaging data with variational autoencoders.
Speiser A, Yan J, Archer E, Buesing L, Turaga SC, Macke JH
Neural Information Processing Systems (NIPS 2017). 2017 Dec 04:

Calcium imaging permits optical measurement of neural activity. Since intracellular calcium concentration is an indirect measurement of neural activity, computational tools are necessary to infer the true underlying spiking activity from fluorescence measurements. Bayesian model inversion can be used to solve this problem, but typically requires either computationally expensive MCMC sampling, or faster but approximate maximum-a-posteriori optimization. Here, we introduce a flexible algorithmic framework for fast, efficient and accurate extraction of neural spikes from imaging data. Using the framework of variational autoencoders, we propose to amortize inference by training a deep neural network to perform model inversion efficiently. The recognition network is trained to produce samples from the posterior distribution over spike trains. Once trained, performing inference amounts to a fast single forward pass through the network, without the need for iterative optimization or sampling. We show that amortization can be applied flexibly to a wide range of nonlinear generative models and significantly improves upon the state of the art in computation time, while achieving competitive accuracy. Our framework is also able to represent posterior distributions over spike-trains. We demonstrate the generality of our method by proposing the first probabilistic approach for separating backpropagating action potentials from putative synaptic inputs in calcium imaging of dendritic spines. 

View Publication Page
12/04/17 | Model-based Bayesian inference of neural activity and connectivity from all-optical interrogation of a neural circuit.
Aitchison L, Russell L, Packer AM, Yan J, Castonguaye P, Häusser M, Turaga SC
31st Conference on Neural Information Processing Systems (NIPS 2017). 2017 Dec 04:

Population activity measurement by calcium imaging can be combined with cellular resolution optogenetic activity perturbations to enable the mapping of neural connectivity in vivo. This requires accurate inference of perturbed and unperturbed neural activity from calcium imaging measurements, which are noisy and indirect, and can also be contaminated by photostimulation artifacts. We have developed a new fully Bayesian approach to jointly inferring spiking activity and neural connectivity from in vivo all-optical perturbation experiments. In contrast to standard approaches that perform spike inference and analysis in two separate maximum-likelihood phases, our joint model is able to propagate uncertainty in spike inference to the inference of connectivity and vice versa. We use the framework of variational autoencoders to model spiking activity using discrete latent variables, low-dimensional latent common input, and sparse spike-and-slab generalized linear coupling between neurons. Additionally, we model two properties of the optogenetic perturbation: off-target photostimulation and photostimulation transients. Our joint model includes at least two sets of discrete random variables; to avoid the dramatic slowdown typically caused by being unable to differentiate such variables, we introduce two strategies that have not, to our knowledge, been used with variational autoencoders. Using this model, we were able to fit models on 30 minutes of data in just 10 minutes. We performed an all-optical circuit mapping experiment in primary visual cortex of the awake mouse, and use our approach to predict neural connectivity between excitatory neurons in layer 2/3. Predicted connectivity is sparse and consistent with known correlations with stimulus tuning, spontaneous correlation and distance.

View Publication Page
09/09/17 | A deep structured learning approach towards automating connectome reconstruction from 3D electron micrographs.
Funke J, Tschopp FD, Grisaitis W, Singh C, Saalfeld S, Turaga SC
arXiv. 2017 Sep 09:

We present a deep learning method for neuron segmentation from 3D electron microscopy (EM), which improves significantly upon state of the art in terms of accuracy and scalability. Our method consists of a fully 3D extension of the U-NET architecture, trained to predict affinity graphs on voxels, followed by a simple and efficient iterative region agglomeration. We train the U-NET using a structured loss function based on MALIS that encourages topological correctness. The resulting affinity predictions are accurate enough that we obtain state-of-the-art results by a simple new learning-free percentile-based iterative agglomeration algorithm. We demonstrate the accuracy of our method on three different and diverse EM datasets where we significantly improve over the current state of the art. We also show for the first time that a common 3D segmentation strategy can be applied to both well-aligned nearly isotropic block-face EM data, and poorly aligned anisotropic serial sectioned EM data. The runtime of our method scales with O(n) in the size of the volume and is thus ready to be applied to very large datasets.

View Publication Page
06/15/16 | Efficient convolutional neural networks for pixelwise classification on heterogeneous hardware systems.
Tschopp F, Martel JN, Turaga SC, Cook M, Funke J
IEEE 13th International Symposium on Biomedical Imaging: From Nano to Macro. 2016 Jun 15:. doi: 10.1109/ISBI.2016.7493487

With recent advances in high-throughput Electron Microscopy (EM) imaging it is now possible to image an entire nervous system of organisms like Drosophila melanogaster. One of the bottlenecks to reconstruct a connectome from these large volumes (œ 100 TiB) is the pixel-wise prediction of membranes. The time it would typically take to process such a volume using a convolutional neural network (CNN) with a sliding window approach is in the order of years on a current GPU. With sliding windows, however, a lot of redundant computations are carried out. In this paper, we present an extension to the Caffe library to increase throughput by predicting many pixels at once. On a sliding window network successfully used for membrane classification, we show that our method achieves a speedup of up to 57×, maintaining identical prediction results.

View Publication Page
11/05/15 | Crowdsourcing the creation of image segmentation algorithms for connectomics.
Arganda-Carreras I, Turaga SC, Berger DR, Ciresan D, Giusti A, Gambardella LM, Schmidhuber J, Laptev D, Dwivedi S, Buhmann JM
Frontiers in Neuroanatomy. 2015 Nov 05;9:142. doi: 10.3389/fnana.2015.00142

To stimulate progress in automating the reconstruction of neural circuits, we organized the first international challenge on 2D segmentation of electron microscopic (EM) images of the brain. Participants submitted boundary maps predicted for a test set of images, and were scored based on their agreement with a consensus of human expert annotations. The winning team had no prior experience with EM images, and employed a convolutional network. This “deep learning” approach has since become accepted as a standard for segmentation of EM images. The challenge has continued to accept submissions, and the best so far has resulted from cooperation between two teams. The challenge has probably saturated, as algorithms cannot progress beyond limits set by ambiguities inherent in 2D scoring and the size of the test dataset. Retrospective evaluation of the challenge scoring system reveals that it was not sufficiently robust to variations in the widths of neurite borders. We propose a solution to this problem, which should be useful for a future 3D segmentation challenge.

View Publication Page
01/13/15 | Mapping social behavior-induced brain activation at cellular resolution in the mouse.
Kim Y, Venkataraju KU, Pradhan K, Mende C, Taranda J, Turaga SC, Arganda-Carreras I, Ng L, Hawrylycz MJ, Rockland KS, Seung HS, Osten P
Cell Reports. 2015 Jan 13;10(2):292-305. doi: 10.1016/j.celrep.2014.12.014

Understanding how brain activation mediates behaviors is a central goal of systems neuroscience. Here, we apply an automated method for mapping brain activation in the mouse in order to probe how sex-specific social behaviors are represented in the male brain. Our method uses the immediate-early-gene c-fos, a marker of neuronal activation, visualized by serial two-photon tomography: the c-fos-GFP+ neurons are computationally detected, their distribution is registered to a reference brain and a brain atlas, and their numbers are analyzed by statistical tests. Our results reveal distinct and shared female and male interaction-evoked patterns of male brain activation representing sex discrimination and social recognition. We also identify brain regions whose degree of activity correlates to specific features of social behaviors and estimate the total numbers and the densities of activated neurons per brain areas. Our study opens the door to automated screening of behavior-evoked brain activation in the mouse.

View Publication Page