Biology subjects

Surampudi, S. G.

Publications and source records attributed to Surampudi, S. G..

4 recordsLinked to original sources

bioRxiv2026

Bayesian generative modeling reveals a multi-modal hierarchical architecture in the mouse functional connectome

Understanding the principles governing large-scale functional organization of the brain remains a central challenge in systems neuroscience. Despite convergent findings, substantial variability across analytical approaches suggests that functional networks may not admit a unique partitioning. Here, we propose that this variability reflects an intrinsic property of the connectome itself: its organization may be fundamentally multi-modal rather than singular. To test this hypothesis, we employ a Bayesian generative modeling framework based on stochastic block models, enabling principled comparison of competing organizational principles and characterization of the full posterior distribution over network partitions. Applying this framework to resting-state fMRI data in mice, we find that a non-degree-corrected hierarchical architecture provides the most parsimonious description of the functional connectome. Importantly, the inferred posterior landscape is not dominated by a single configuration, but instead comprises multiple distinct and co-dominant organizational schemes. At the mesoscale, these hierarchical communities are anatomically grounded yet systematically reorganize canonical resting-state networks: primary sensory systems remain cohesive, whereas higher-order association networks are fractionated into multiple interacting sub-circuits. This global structural variation is driven by structured variability at the community level, where integrative systems exhibit variable regional affiliations while sensory systems act as structurally stable anchors. Together, these findings suggest that the resting-state connectome is best described as a distribution over alternative, yet co-dominant, organizational configurations. This perspective reconciles inconsistencies across previous studies and supports a view of brain organization as inherently degenerate, providing a latent repertoire of network configurations that may underlie adaptive information routing and dynamic functional reconfiguration.

neuroscience↗

bioRxiv2023

Threat and reward imminence processing in the human brain

In the human brain, aversive and appetitive processing have been studied with controlled stimuli in rather static settings. In addition, the extent to which aversive- and appetitive-related processing engage distinct or overlapping circuits remains poorly understood. Here, we sought to investigate the dynamics of aversive and appetitive processing while male and female participants engaged in comparable trials involving threat-avoidance or reward-seeking. A central goal was to characterize the temporal evolution of responses during periods of threat or reward imminence. For example, in the aversive domain, we predicted that the bed nucleus of the stria terminalis (BST), but not the amygdala, would exhibit anticipatory responses given the role of the former in anxious apprehension. We also predicted that the periaqueductal gray (PAG) would exhibit threat-proximity responses based on its involvement in proximal-threat processes, and that the ventral striatum would exhibit threat-imminence responses given its role in threat escape in rodents. Overall, we uncovered imminence-related temporally increasing ("ramping") responses in multiple brain regions, including the BST, PAG, and ventral striatum, subcortically, and dorsal anterior insula and anterior midcingulate, cortically. Whereas the ventral striatum generated anticipatory responses in the proximity of reward as expected, it also exhibited threat-related imminence responses. In fact, across multiple brain regions, we observed a main effect of arousal. In other words, we uncovered extensive temporally-evolving, imminence-related processing in both the aversive and appetitive domain, suggesting that distributed brain circuits are dynamically engaged during the processing of biologically relevant information irrespective of valence, findings further supported by network analysis. Significance StatementIn the human brain, aversive and appetitive processing have been studied with controlled stimuli in rather static settings. Here, we sought to investigate the dynamics of aversive/appetitive processing while participants engaged in trials involving threat-avoidance or reward-seeking. A central goal was to characterize the temporal evolution of responses during periods of threat or reward imminence. We uncovered imminence-related temporally increasing ("ramping") responses in multiple brain regions, including the bed nucleus of the stria terminalis, periaqueductal gray, and ventral striatum, subcortically, and dorsal anterior insula and anterior midcingulate, cortically. Overall, we uncovered extensive temporally-evolving, imminence-related processing in both the aversive and appetitive domain, suggesting that distributed brain circuits are dynamically engaged during the processing of biologically relevant information irrespective of valence.

neuroscience↗

bioRxiv2021

Learning brain dynamics for decoding and predicting individual differences

Insights from functional Magnetic Resonance Imaging (fMRI), as well as recordings of large numbers of neurons, reveal that many cognitive, emotional, and motor functions depend on the multivariate interactions of brain signals. To decode brain dynamics, we propose an architecture based on recurrent neural networks to uncover distributed spatiotemporal signatures. We demonstrate the potential of the approach using human fMRI data during movie-watching data and a continuous experimental paradigm. The model was able to learn spatiotemporal patterns that supported 15-way movie-clip classification (~90%) at the level of brain regions, and binary classification of experimental conditions (~60%) at the level of voxels. The model was also able to learn individual differences in measures of fluid intelligence and verbal IQ at levels comparable or better than existing techniques. We propose a dimensionality reduction approach that uncovers low-dimensional trajectories and captures essential informational (i.e., classification related) properties of brain dynamics. Finally, saliency maps were employed to characterize brain-region/voxel importance, and uncovered how dynamic but consistent changes in fMRI activation influenced decoding performance. When applied at the level of voxels, our framework implements a dynamic version of multivariate pattern analysis. We believe our approach provides a powerful framework for visualizing, analyzing, and discovering dynamic spatially distributed brain representations during naturalistic conditions.1 Author summaryBrain signals are inherently dynamic and evolve in both space and time as a function of cognitive or emotional task condition or mental state. To characterize brain dynamics, we employed an architecture based on recurrent neural networks, and applied it to functional magnetic resonance imaging data from humans watching movies or during continuous experimental conditions. The model learned spatiotemporal patterns that allowed it to correctly classify which clip a participant was watching based entirely on data from other participants; the model also learned a binary classification of experimental conditions at the level of voxels. We developed a dimensionality reduction approach that uncovered low-dimensional "trajectories" and captured essential information properties of brain dynamics. When applied at the level of voxels, our framework implements a dynamic version of multivariate pattern analysis. We believe our approach provides a powerful framework for visualizing, analyzing, and discovering dynamic spatially distributed brain representations during naturalistic conditions.

neuroscience↗

bioRxiv2018

Temporal Multiple Kernel Learning (tMKL) model for predicting resting state FC via characterizing fMRI connectivity dynamics

Over the last decade there has been growing interest in understanding the brain activity in the absence of any task or stimulus captured by the resting-state functional magnetic resonance imaging (rsfMRI). These resting state patterns are not static, but exhibit complex spatio-temporal dynamics. In the recent years substantial effort has been put to characterize different FC configurations while brain states makes transitions over time. The dynamics governing this transitions and their relationship with stationary functional connectivity remains elusive. Over the last years a multitude of methods has been proposed to discover and characterize FC dynamics and one of the most accepted method is sliding window approach. Moreover, as these FC configurations are observed to be cyclically repeating in time there was further motivation to use of a generic clustering scheme to identify latent states of dynamics. We discover the underlying lower-dimensional manifold of the temporal structure which is further parameterized as a set of local density distributions, or latent transient states. We propose an innovative method that learns parameters specific to these latent states using a graph-theoretic model (temporal Multiple Kernel Learning, tMKL) and finally predicts the grand average functional connectivity (FC) of the unseen subjects by leveraging a state transition Markov model. tMKL thus learns a mapping between the underlying anatomical network and the temporal structure. Training and testing were done using the rs-fMRI data of 46 healthy participants and the results establish the viability of the proposed solution. Parameters of the model are learned via state-specific optimization formulations and yet the model performs at par or better than state-of-the-art models for predicting the grand average FC. Moreover, the model shows sensitivity towards subject-specific anatomy. The proposed model performs significantly better than the established models of predicting resting state functional connectivity based on whole-brain dynamic mean-field model, single diffusion kernel model and another version of multiple kernel learning model. In summary, We provide a novel solution that does not make strong assumption about underlying data and is generally applicable to resting or task data to learn subject specific state transitions and successful characterization of SC-dFC-FC relationship through an unifying framework.

neuroscience↗