Neuro-Informatics seminars

Large-scale neural recordings contain high-dimensional structure that cannot be easily captured by existing data visualization methods. We therefore developed an embedding algorithm called Rastermap, which captures highly nonlinear relationships between neurons, and provides useful visualizations by assigning each neuron to a location in the embedding space. Compared to standard algorithms such as t-SNE and UMAP, Rastermap finds finer and higher dimensional patterns of neural variability, as measured by quantitative benchmarks. We applied Rastermap to a variety of datasets, including spontaneous neural activity, neural activity during a virtual reality task, widefield neural imaging data during a 2AFC task, artificial neural activity from an agent playing atari games, and neural responses to visual textures. We found within these datasets unique subpopulations of neurons encoding abstract properties of the environment.

Where in the brain consciousness resides remains unclear. It has been suggested that the subnetworks supporting consciousness should be bidirectionally (recurrently) connected because both feed-forward and feedback processing are necessary for conscious experience. Accordingly, evaluating which subnetworks are bidirectionally connected and the strength of these connections would likely aid the identification of regions essential to consciousness. Here, we propose a method for hierarchically decomposing a network into cores with different strengths of bidirectional connection, as a means of revealing the structure of the complex brain network. We applied the method to a whole-brain mouse connectome. We found that cores with strong bidirectional connections consisted of regions presumably essential to consciousness (e.g., the isocortical and thalamic regions, and claustrum) and did not include regions presumably irrelevant to consciousness (e.g., cerebellum). Contrarily, we could not find such correspondence between cores and consciousness when we applied other simple methods which ignored bidirectionality. These findings suggest that our method provides a novel insight into the relation between bidirectional brain network structures and consciousness. Our recent preprint on this work is here: https://doi.org/10.1101/2021.07.12.452022.

Large-scale, single neuron resolution recordings are inherently high-dimensional, with as many dimensions as neurons. To make sense of them, for many the answer is: reduce the number of dimensions. In this talk I argue we can distinguish weak and strong principles of neural dimension reduction. The weak principle is that dimension reduction is a convenient tool for making sense of complex neural data. The strong principle is that dimension reduction moves us closer to how neural circuits actually operate and compute. Elucidating these principles is crucial, for which we subscribe to provides radically different interpretations of the same dimension reduction techniques applied to the same data. I outline experimental evidence for each principle, but illustrate how we could make either the weak or strong principles appear to be true based on innocuous looking analysis decisions. These insights suggest arguments over low and high-dimensional neural activity need better constraints from both experiment and theory.

Many of the collaborations we encounter in our scientific careers are centered on a common idea that can be associated with certain resources, such as a dataset, an algorithm, or a model. All partners in a collaboration need to develop a common understanding of these resources, and need to be able to access them in a simple and unambiguous manner in order to avoid incorrect conclusions especially in highly cross-disciplinary contexts. While digital computers have entered to assist scientific workflows in experiment and simulation for many decades, the high degree of heterogeneity in the field had led to a scattered landscape of highly customized, lab-internal solutions to organizing and managing the resources on a project-by-project basis. Only with the availability of modern technologies such as the semantic web, platforms for collaborative coding or the development of data standards overarching different disciplines, we have tools at our disposal to make resources increasingly more accessible, understandable, and usable. However, without overarching standardization efforts and adaptation of such technologies to the workflows and needs of individual researchers, their adoption by the neuroscience community will be impeded. From the perspective of computational neuroscience, which is inherently dependent on leveraging data and methods across the field of neuroscience for inspiration and validation, I will outline my view on past and present developments towards a more rigorous use of digital resources and how they improved collaboration, and introduce emerging initiatives to support this process in the future (e.g., EBRAINS http://ebrains.eu, NFDI-Neuro http://www.nfdi-neuro.de).

The analysis of functional magnetic resonance imaging (fMRI) data can greatly benefit from flexible analytic approaches. In particular, the advent of data-driven approaches to identify whole-brain time-varying connectivity and activity has revealed a number of interesting relevant variation in the data which, when ignored, can provide misleading information. In this lecture I will provide a comparative introduction of a range of data-driven approaches to estimating time-varying connectivity. I will also present detailed examples where studies of both brain health and disorder have been advanced by approaches designed to capture and estimate time-varying information in resting fMRI data. I will review several exemplar data sets analyzed in different ways to demonstrate the complementarity as well as trade-offs of various modeling approaches to answer questions about brain function. Finally, I will review and provide examples of strategies for validating time-varying connectivity including simulations, multimodal imaging, and comparative prediction within clinical populations, among others. As part of the interactive aspect I will provide a hands-on guide to the dynamic functional network connectivity toolbox within the GIFT software, including an online didactic analytic decision tree to introduce the various concepts and decisions that need to be made when using such tools

The Group Iterative Multiple Model Estimation (GIMME) framework has emerged as a promising method for characterizing connections between brain regions in functional neuroimaging data. Two of the most appealing features of this framework are its ability to estimate the directionality of connections between network nodes and its ability to determine whether those connections apply to everyone in a sample (group-level) or just to one person (individual-level). However, there are outstanding questions about the validity and stability of these estimates, including: 1) how recovery of connection directionality is affected by features of data sets such as scan length and autoregressive effects, which may be strong in some imaging modalities (resting state fMRI, fNIRS) but weaker in others (task fMRI); and 2) whether inferences about directionality at the group and individual levels are stable across time. This talk will provide an overview of the GIMME framework and describe relevant results from a large-scale simulation study that assesses directionality recovery under various conditions and a separate project that investigates the temporal stability of GIMME’s inferences in the Human Connectome Project data set. Analyses from these projects demonstrate that estimates of directionality are most precise when autoregressive and cross-lagged relations in the data are relatively strong, and that inferences about the directionality of group-level connections, specifically, appear to be stable across time. Implications of these findings for the interpretation of directional connectivity estimates in different types of neuroimaging data will be discussed.

Technological advances have increased the availability of recordings from large populations of neurons across multiple brain areas. Coupling these recordings with dimensionality reduction techniques, recent work has led to new proposals for how populations of neurons can send and receive signals selectively and flexibly. Advancement of these proposals depends, however, on untangling the bidirectional, parallel communication between neuronal populations. Because our current data analytic tools struggle to achieve this task, we have recently validated and presented a novel dimensionality reduction framework: DLAG, or Delayed Latents Across Groups. DLAG decomposes the time-varying activity in each area into within- and across-area latent variables. Across-area variables can be decomposed further into feedforward and feedback components using automatically estimated time delays. In this talk, I will review the DLAG framework. Then I will discuss new insights into the moment-by-moment nature of feedforward and feedback communication between visual cortical areas V1 and V2 of macaque monkeys. Overall, this work lays the foundation for dissecting the dynamic flow of signals across populations of neurons, and how it might change across brain areas and behavioral contexts.

Behavior arises from the coordinated activity of numerous distinct brain regions. Modern experimental tools allow access to neural populations brain-wide, yet understanding such large-scale datasets necessitates scalable computational models to extract meaningful features of inter-region communication. In this talk, I will introduce Current-Based Decomposition (CURBD), an approach for inferring multi-region interactions using data-constrained recurrent neural network models. I will first show that CURBD accurately isolates inter-region currents in simulated networks with known dynamics. I will then apply CURBD to understand the brain-wide flow of information leading to behavioral state transitions in larval zebrafish. These examples will establish CURBD as a flexible, scalable framework to infer brain-wide interactions that are inaccessible from experimental measurements alone.

Rajanlab designs neural network models constrained by experimental data, and reverse engineers them to figure out how brain circuits function in health and disease. Recently, we have been developing a powerful new theory-based framework for “in-vivo tract tracing” from multi-regional neural activity collected experimentally. We call this framework CURrent-Based Decomposition (CURBD). CURBD employs recurrent neural networks (RNNs) directly constrained, from the outset, by time series measurements acquired experimentally, such as Ca2+ imaging or electrophysiological data. Once trained, these data-constrained RNNs let us infer matrices quantifying the interactions between all pairs of modeled units. Such model-derived “directed interaction matrices” can then be used to separately compute excitatory and inhibitory input currents that drive a given neuron from all other neurons. Therefore different current sources can be de-mixed – either within the same region or from other regions, potentially brain-wide – which collectively give rise to the population dynamics observed experimentally. Source de-mixed currents obtained through CURBD allow an unprecedented view into multi-region mechanisms inaccessible from measurements alone. We have applied this method successfully to several types of neural data from our experimental collaborators, e.g., zebrafish (Deisseroth lab, Stanford), mice (Harvey lab, Harvard), monkeys (Rudebeck lab, Sinai), and humans (Rutishauser lab, Cedars Sinai), where we have discovered both directed interactions brain wide and inter-area currents during different types of behaviors. With this framework based on data-constrained multi-region RNNs and CURrent Based Decomposition (CURBD), we can ask if there are conserved multi-region mechanisms across different species, as well as identify key divergences.

I will review advances in understanding the cortical encoding of speech-related oral movements. These discoveries are being translated to develop algorithms to decode speech from population neural activity.

Many networks in the brain are sparsely connected, and the brain eliminates synapses during development and learning. How could the brain decide which synapses to prune? In a recurrent network, determining the importance of a synapse between two neurons is a difficult computational problem, depending on the role that both neurons play and on all possible pathways of information flow between them. Noise is ubiquitous in neural systems, and often considered an irritant to be overcome. In the first part of this talk, I will suggest that noise could play a functional role in synaptic pruning, allowing the brain to probe network structure and determine which synapses are redundant. I will introduce a simple, local, unsupervised plasticity rule that either strengthens or prunes synapses using only synaptic weight and the noise-driven covariance of the neighboring neurons. For a subset of linear and rectified-linear networks, this rule provably preserves the spectrum of the original matrix and hence preserves network dynamics even when the fraction of pruned synapses asymptotically approaches 1. The plasticity rule is biologically-plausible and may suggest a new role for noise in neural computation. Time permitting, I will then turn to the problem of extracting structure from neural population data sets using dimensionality reduction methods. I will argue that nonlinear structures naturally arise in neural data and show how these nonlinearities cause linear methods of dimensionality reduction, such as Principal Components Analysis, to fail dramatically in identifying low-dimensional structure.

Information that is shared across brains is encoded in idiosyncratic fine-scale functional topographies. Hyperalignment jointly models shared information and idiosyncratic topographies. Pattern vectors for neural responses and connectivities are projected into a common, high-dimensional information space, rather than being aligned in a canonical anatomical space. Hyperalignment calculates individual transformation matrices that preserve the geometry of pairwise dissimilarities between pattern vectors. Individual cortical topographies are modeled as mixtures of overlapping, individual-specific topographic basis functions, rather than as contiguous functional areas. The fundamental property of brain function that is preserved across brains is information content, rather than the functional properties of local features that support that content.

Information is represented in the brain by the coordinated activity of populations of neurons. Recent large-scale neural recording methods in combination with machine learning algorithms are helping understand how sensory processing and cognition emerge from neural population activity. This talk will explore the most popular machine learning methods used to gather meaningful low-dimensional representations from higher-dimensional neural recordings. To illustrate the potential of these approaches, Pedro will present his research in which chemical information is decoded from the olfactory system of the mouse for technological applications. Pedro and co-researchers have successfully extracted odor identity and concentration from olfactory receptor neuron low-dimensional activity trajectories. They have further developed a novel method to identify a shared latent space that allowed decoding of odor information across animals.