Biology subjects

Bhalla, U. S.

Publications and source records attributed to Bhalla, U. S..

At least 19 recordsLinked to original sources

bioRxiv2026

Time cells differentially populate trace and post-trace epochs, but do not remap for different trace intervals

Hippocampal time cells are a neural substrate for linking discontiguous events in time. While they have been reported both for working memory tasks ([~]15 s) and trace-eyeblink conditioning (TEC, [~]300 ms), the latter is [~]50 fold faster and requires pre-emptive and precisely timed responses. We therefore asked if novel features of time-cell rescaling, post-stimulus activity, and time-cell extinction might arise in TEC. We used 2-photon calcium-sensor imaging from mouse hippocampal CA1 pyramidal neurons during TEC tasks with varying intervals, and specifically included a protocol in which short blocks of 250 and 550ms trials were alternated to eliminate learning-dependent rescaling. We find that conditioned stimulus (CS) triggered traces extend for >5 seconds after the unconditioned stimulus (US), and the trials can be subdivided into epochs with distinct proportions of time cells and time-cell peak widths. Instead of remapping, the time-cell sequence is the same between 250 and 550ms intervals. CS-triggered time cells are present before trace learning, but are nearly lost after behavioural extinction, suggesting that their removal is an active process. We propose that in TEC, salient stimuli initiate time-cell sequences which are independent of subsequent timing or even absence of US, but can be gated off during extinction.

neuroscience↗

bioRxiv2026

Formation, stability, and competition in dendritic proto-spines

Over 80 % of excitatory synapses in the mammalian brain terminate on dendritic spines, which support key functions of memory storage and computation. Despite the neurodevelopmental and computational implications of spine formation, little is known about the interplay of physical and chemical signaling mechanisms that govern spine initiation. We developed a physics-based model for the formation, stability, and placement of membrane protrusions (proto-spines) triggered by membrane-bending proteins such as IRSp53, and activated by neurotransmitter inputs. We identify two distinct proto-spine energy manifolds depending on the concentration of activated IRSp53, causing shallow and sharp geometries respectively. The sharp proto-spines exhibit properties of wave-pinning models, including stimulus-local formation, positional stability, and merging, but unlike the wave-pinning model they also exhibit competition. We linked the physics-based model to a mass-action chemistry model of neurotransmitter-driven signaling pathway activation of IRSp53 to simulate proto-spine formation in response to neurotransmitter stimulus patterns. Multiple proto-spines compete for IRSp53, leading to inter-spine spacing of 1 - 3 {micro}m, similar to experiments. Correlated and spatially clustered synaptic input enhances proto-spine lifetime and density compared to non-correlated stimuli. Finally, we tested the effects of autism spectrum disorder-related mutations in spine-formation signaling, leading to altered RhoGEF activity. We found that elevated RhoGEF activity increased proto-spine density and reduced their lifetime, consistent with ASD phenotypes. Overall, our analysis shows that the interplay between membrane bending mechanics and reaction-diffusion chemistry creates rich proto-spine dynamics and concisely captures key features of early spine formation. Significance StatementDendritic spines mediate most of the excitatory neurotransmission in the brain, and alterations in their shape and distribution on the dendrite are correlated with neuro-developmental and neuro-degenerative conditions. We model membrane mechanics and chemical signaling to show that spine precursors arise as self-stabilizing membrane protrusions due to the binding of membrane-bending proteins. Spine precursors act as nucleation sites for spine-building proteins, and hence their stability, placement, and signaling impacts late-stage spine maturation. These proto-spines compete with each other for bending proteins, and we connect this interplay with signaling chemistry that triggers new spines immediately below sites of transmitter release. We find that aberrant signaling in neurological disease alters proto-spine competition, life-time and placement.

neuroscience↗

bioRxiv2025

Laplacian Dynamics and Kron Reduction in Species-Reaction Graphs of Chemical Reaction Networks

We present a new method for deriving the dynamics of chemical reaction networks using the Laplacian matrix of the corresponding species-reaction graph, in contrast to previous works that use the Laplacian of the graph of complexes. Species-reaction graphs are bipartite graphs that contain two sets of vertices, one representing species and the other representing reactions, connected by directed edges that indicate relationships between them. Our approach starts by assigning appropriate edge weights to this bipartite graph, which are then used to compute the weighted graph Laplacian. This Laplacian reformulation of the system of differential equations governing the network dynamics emphasizes the flow of information throughout the chemical reaction network considered as causal network. As an application of this framework, we introduce a novel model reduction technique based on the Kron reduction of the weighted Laplacian matrix associated with the species-reaction graphs. Our systematic approach involves identifying nodes for deletion while preserving the bipartite structure, followed by constructing the Kron-reduced model. To demonstrate the effectiveness of our method, we apply it to a complex biochemical network, showing how model simplification facilitates analysis and interpretation of these systems.

systems biology↗

bioRxiv2025

Heterogeneity and variability in short term synaptic plasticity and implications for signal transformation

There are many sources of heterogeneity in the CA1 network, including plasticity, connectivity and cell properties, yet the extent and functional consequences of this diversity remains poorly understood. We used patterned optogenetic stimulation of CA3 pyramidal neurons and whole-cell patch clamp recordings from CA1 pyramidal neurons in acute mouse hippocampal slices, to characterize the contributions of different forms of heterogeneity to information flow. We found pronounced heterogeneity in synaptic responses and short-term plasticity (STP), influenced by the neurotransmitter identity, input pattern, and size of the activated presynaptic ensemble. Inhibitory synapses exhibited greater diversity in both response variability and depression profiles than excitatory synapses. We incorporated these readings in a molecule-to-network multiscale model of the CA3->CA1 circuit. The reference model shows strong decorrelation of autocorrelated input, but removal of STP makes the decorrelation frequency dependent. Removal of stochasticity and heterogeneity in connections makes the output periodic. Thus heterogeneity, short-term plasticity, and stochasticity each have distinct effects on cellular information transmission.

neuroscience↗

bioRxiv2025

Automated hierarchical block decomposition of biochemical networks

Biochemical networks are models of biological functions and processes in biomedicine. Hierarchical decomposition simplifies complex biochemical networks by partitioning them into smaller blocks (modules), facilitating computationally intensive analyses and providing deeper insights into cellular processes and regulatory mechanisms. We introduce a novel algorithm for the hierarchical decomposition of large-scale biochemical systems. By using causality and information flow as organizing principles, our approach combines strongly connected components with r-causality to identify and structure manageable network blocks. Benchmarking against a comprehensive database of biochemical reaction networks demonstrates the computational efficiency and scalability of our algorithm. To ensure broad applicability, we integrate our algorithm into tools that support standardized Systems Biology Markup Language (SBML) formats, facilitating its use in biochemical modeling workflows.

systems biology↗

bioRxiv2024

A subset of mouse hippocampus CA1 pyramidal neurons learns sparse synaptic input patterns.

Synaptic plasticity in the hippocampus is fundamental to learning and memory, yet few studies examine how pattern learning occurs across multiple synapses. Such cross-synapse learning is fundamental to emergent properties of pattern discrimination and generalisation, which depend on assumptions about independence of plasticity and linearity of summation. We used sparse optogenetic spatio-temporal pattern stimulation in the CA3 coupled with postsynaptic depolarization to elicit plasticity on CA1 pyramidal neurons, and found that trained patterns were selectively strengthened, but only in a subset of postsynaptic cells. Increased resting membrane potential and background mini-EPSP rates were predictive of learner cells. Summation following plasticity became more linear in learners compared to non-learners, consistent with the observed elevated post-stimulus hyperpolarization on non-learner cells. Thus our exploration of biologically plausible sparse activity supports pattern-selective learning, but in a heterogeneous manner modulated by both cell-intrinsic and network features.

neuroscience↗

bioRxiv2024

Short-term plasticity of EI balance at single neurons can detect pattern transitions

Sensory input and internal context converge onto the hippocampus as spatio-temporal activity patterns. Transitions in these input patterns are frequently salient. We demonstrate that short-term potentiation (STP) mediates escape from EI balance to implement mismatch detection in spatiotemporally patterned activity sequences. We characterized STP in the mouse hippocampus CA3-CA1 network using optogenetic patterned stimuli in CA3 while recording from CA1 pyramidal neurons. STP modulates EI summation across patterns, first amplifying, then reducing responses. We parameterized a multiscale model of network projections onto hundreds of E and I boutons on a CA1 neuron, each including stochastic signaling to mediate STP. The model detected mismatches in trains of input patterns, which we experimentally confirmed. Mismatch selectivity depends on stimulus overlap, network weights, and connectivity. It is robust over a wide range of model parameters and assumptions about input spike timing jitter, postsynaptic spiking and stochasticity. Finally, we predict that optimal mismatch selectivity can be tuned over low to high gamma frequencies by modulating network parameters, and show that there is strong mismatch detection for gamma-frequency bursts between theta cycles, consistent with theta-tuned snapshots of novel input.

neuroscience↗

bioRxiv2024

Hierarchical optimization of biochemical networks

Biological signalling systems are complex, and efforts to build mechanistic models must confront a huge parameter space, indirect and incomplete data, and frequently encounter multiscale and multiphysics phenomena. We present HOSS, a framework for Hierarchical Optimization of Systems Simulations, to address such problems. HOSS operates by breaking down extensive systems models into individual pathway blocks organized in a nested hierarchy. At the first level, dependencies are solely on signalling inputs, and subsequent levels rely only on the preceding ones. We demonstrate that each independent pathway in every level can be efficiently optimized. Once optimized, its parameters are held constant while the pathway serves as input for succeeding levels. We develop an algorithmic approach to identify the necessary nested hierarchies for the application of HOSS in any given biochemical network. Furthermore, we devise two parallelizable variants that generate numerous model instances using stochastic scrambling of parameters during initial and intermediate stages of optimization. Our results indicate that these variants produce superior models and offer an estimate of solution degeneracy. Additionally, we showcase the effectiveness of the optimization methods for both abstracted, event-based simulations and ODE-based models. Author summaryBiochemical pathway models integrate quantitative and qualitative data to understand cell functioning, disease effects, and to test treatments in silico. Constructing and optimizing these models is challenging due to the complexity and multitude of variables and parameters involved. Although hundreds of biochemical models have been developed and are available in repositories, they are rarely reused. To enhance the utilization of these models in biomedicine, we propose HOSS, an innovative hierarchical model optimization method. HOSS takes advantage of the modular structure of pathway models by breaking down large mechanistic computational models into smaller modules. These modules are then optimized progressively, starting with input modules and following causality paths. This method significantly reduces the computational burden as each step involves solving a simpler problem. By making the optimization process more manageable, HOSS accelerates the lifecycle of biochemical models and promotes their broader use in biomedical research and applications.

systems biology↗

bioRxiv2024

Learning and modality independent time cells tile stimulus and post-stimulus period.

Hippocampal neurons represent multiple dimensions of stimulus and behavioural context, including time. It is unclear how soon time-encoding cells emerge during learning, and if they additionally encode context in a manner specific to behavioural paradigm. We investigated simultaneous time and context encoding using 2-photon calcium imaging of mouse hippocampus during a trace eyeblink conditioning (TEC) task with sound and light stimuli. We find that the fraction of time cells is independent of learning and of stimulus modality. Only 13% of cells retain time-encoding on successive days, but persisters remain active within their original stimulus and post-stimulus epochs. Finally, we show that 60% of modality-specific time-encoding cells are active after the stimulus period, but modality-agnostic time cells are rare post-stimulus. Thus, compared to other paradigms, time cells in TEC have distinct learning and turnover properties, and exhibit sustained coding of stimulus modality and time which may subserve associations with subsequent events.

neuroscience↗

bioRxiv2024

Discriminating neural ensemble patterns through dendritic computations in randomly connected feedforward networks

Co-active or temporally ordered neural ensembles are a signature of salient sensory, motor, and cognitive events. Local convergence of such patterned activity as synaptic clusters on dendrites could help single neurons harness the potential of dendritic nonlinearities to decode neural activity patterns. We combined theory and simulations to assess the likelihood of whether projections from neural ensembles could converge onto synaptic clusters even in networks with random connectivity. Using rat hippocampal and cortical network statistics, we show that clustered convergence of axons from 3-4 different co-active ensembles is likely even in randomly connected networks, leading to representation of arbitrary input combinations in at least ten target neurons in a 100,000 population. In the presence of larger ensembles, spatiotemporally ordered convergence of 3-5 axons from temporally ordered ensembles is also likely. These active clusters result in higher neuronal activation in the presence of strong dendritic nonlinearities and low background activity. We mathematically and computationally demonstrate a tight interplay between network connectivity, spatiotemporal scales of subcellular electrical and chemical mechanisms, dendritic nonlinearities, and uncorrelated background activity. We suggest that dendritic clustered and sequence computation is pervasive, but its expression as somatic selectivity requires confluence of physiology, background activity, and connectomics.

neuroscience↗

bioRxiv2024

Mechanisms and implications of high depolarization baseline offsets in conductance-based neuronal models

Somatic step-current injection is commonly used to characterize the electrophysiological properties of neurons. Many neuronal types show a large depolarization baseline offset (DBLO), which is defined as the positive difference between the minimum membrane potential during action potential trains and resting. We used stochastic parameter search in experimentally constrained conductance-based models to show that four key factors together account for high DBLO: Liquid Junction Potential correction, high backpropagating passive charges during the repolarization phase of an action potential, fast potassium delayed rectifier kinetics, and appropriate transient sodium current kinetics. Several plausible mechanisms for DBLO, such as Ohmic depolarization due to current input or low-pass filtering by the membrane, fail to explain the effect, and many published conductance-based models do not correctly manifest high DBLO. Finally, physiological levels of DBLO constrain ion channel levels and kinetics, and are linked to cellular processes such as bistable firing, spikelets, and calcium influx. O_FIG O_LINKSMALLFIG WIDTH=195 HEIGHT=200 SRC="FIGDIR/small/575308v2_ufig1.gif" ALT="Figure 1"> View larger version (35K): org.highwire.dtl.DTLVardef@127a9aorg.highwire.dtl.DTLVardef@12f3f95org.highwire.dtl.DTLVardef@f753e8org.highwire.dtl.DTLVardef@1a5e470_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗

bioRxiv2022

Synthetic Data Resource and Benchmarks for Time Cell Analysis and Detection Algorithms

Hippocampal CA1 cells take part in reliable, time-locked activity sequences in tasks that involve an association between temporally separated stimuli, in a manner that tiles the interval between the stimuli. Such cells have been termed time cells. Here we adopt a first-principles approach to comparing diverse analysis and detection algorithms for identifying time cells. We generated synthetic activity datasets using calcium signals recorded in vivo from the mouse hippocampus using 2-Photon imaging, as template response waveforms. We assigned known, ground truth values to perturbations applied to perfect activity signals, including noise, calcium event width, timing imprecision, hit-trial ratio and background (untuned) activity. We tested a range of published and new algorithms and their variants on this dataset. We find that most algorithms correctly classify over 80% of cells, but have different balances between true and false positives, and different sensitivity to the five categories of perturbation. Reassuringly, most methods are reasonably robust to perturbations, including background activity, and show good concordance in classification of time cells. The same algorithms were also used to analyse and identify time cells in experimental physiology datasets recorded in vivo and most show good concordance. Significance StatementNumerous approaches have been developed to analyze time cells and neuronal activity sequences, but it is not clear if their classifications match, nor how sensitive they are to various sources of data variability. We provide two main contributions to address this: 1) A resource to generate ground truth labelled synthetic 2-P Calcium activity data with defined distributions for confounds such as noise and background activity, and 2) a survey of several methods for analyzing time-cell data using our synthetic data as ground truth. As a further resource, we provide a library of efficient C++ implementations of several algorithms with a Python interface. The synthetic dataset and its generation code are useful for profiling future methods, testing analysis toolchains, and as input to computational and experimental models of sequence detection.

neuroscience↗

bioRxiv2020

An early cortical progenitor-specific mechanism regulates thalamocortical innervation

The cortical subplate is critical in regulating the entry of thalamocortical sensory afferents into the cortex. These afferents reach the subplate at embryonic day (E)15.5 in the mouse, but "wait" for several days, entering the cortical plate postnatally. We report that when transcription factor Lhx2 is lost in E11.5 cortical progenitors, which give rise to subplate neurons, thalamocortical afferents display premature, exuberant innervation of the E15.5 cortex. Embryonic mutant subplate neurons are correctly positioned below the cortical plate, but they display an altered transcriptome and immature electrophysiological properties during the waiting period. The sensory thalamus in these cortex-specific Lhx2 mutants displays atrophy, eventually leading to severe deficits in thalamocortical innervation. Strikingly, these phenotypes do not manifest if Lhx2 is lost in postmitotic subplate neurons. These results demonstrate a mechanism operating in subplate progenitors that has profound consequences on the growth of thalamocortical axons into the cortex.

neuroscience↗

bioRxiv2020

HillTau: A fast, compact abstraction for model reduction in biochemical signaling networks

Signaling networks mediate many aspects of cellular function. The conventional, mechanistically motivated approach to modeling such networks is through mass-action chemistry, which maps directly to biological entities and facilitates experimental tests and predictions. However such models are complex, need many parameters, and are computationally costly. Here we introduce the HillTau form for signaling models. HillTau retains the direct mapping to biological observables, but it uses far fewer parameters, and is 100 to over 1000 times faster than ODE-based methods. In the HillTau formalism, the steady-state concentration of signaling molecules is approximated by the Hill equation, and the dynamics by a time-course tau. We demonstrate its use in implementing several biochemical motifs, including association, inhibition, feedforward and feedback inhibition, bistability, oscillations, and a synaptic switch obeying the BCM rule. The major use-cases for HillTau are system abstraction, model reduction, scaffolds for data-driven optimization, and fast approximations to complex cellular signaling. Author summaryChemical signals mediate many computations in cells, from housekeeping functions in all cells to memory and pattern selectivity in neurons. These signals form complex networks of interactions. Computer models are a powerful way to study how such networks behave, but it is hard to get all the chemical details for typical models, and it is slow to run them with standard numerical approaches to chemical kinetics. We introduce HillTau as a simplified way to model complex chemical networks. HillTau models condense multiple reaction steps into single steps defined by a small number of parameters for activation and settling time. As a result the models are simple, easy to find values for, and they run quickly. Remarkably, they fit the full chemical formulations rather well. We illustrate the utility of HillTau for modeling several signaling network functions, and for fitting complicated signaling networks.

systems biology↗

bioRxiv2020

SWITCHES: Searchable Web Interface for Topologies of CHEmical Switches.

Bistable biochemical switches are key motifs in cellular state decisions and long-term storage of cellular memory. There are a few known biological switches that have been well characterized, however these examples are insufficient for systematic surveys of properties of these important systems. Here we present a resource of all possible bistable biochemical reaction networks with up to 6 reactions between 3 molecules, and 3 reactions between 4 molecules. Over 35,000 reaction topologies were constructed by identifying unique combinations of reactions between a fixed number of molecules. Then, these topologies were populated with rates within a biologically realistic range. The Searchable Web Interface for Topologies of CHEmical Switches (SWITCHES, https://switches.ncbs.res.in) provides a bistability and parameter analysis of over 7 million models from this systematic survey of chemical reaction space. This database will be useful for theoreticians interested in analyzing stability in chemical systems and also experimentalists for creating robust synthetic biological switches. Availability and ImplementationFreely available on the web at https://switches.ncbs.res.in. Website implemented in PHP, MariaDB, Graphviz, and Apache, with all major browsers supported. Supplementary InformationNot applicable.

bioinformatics↗

bioRxiv2020

Different dimensions of robustness- noise, topology and rates - are nearly independent in chemical switches.

Life prospers despite adverse conditions in many unpredictable dimensions. This requires that cellular processes work reliably, that is they are robust against many kinds of perturbations. For example, a cellular decision to differentiate should be stable despite changes in metabolic conditions and stochasticity due to thermal noise. For evolutionary stability, the same differentiation switch should function despite mutations or the evolution of further regulatory inputs. We asked how cellular decision making responds to these three forms of perturbation, expressed in chemical terms as rate parameters, stochasticity, and reaction topology. Remarkably, we found that there was no correlation between noise robustness and either of the others and only a weak correlation between robustness to parameters and topology. Thus, a given chemical switch could be robust to noise yet sensitive to parametric or topological changes. However, we found families of reaction topologies derived from a common core bistable with symmetric feedback loops, which retained bistability despite the removal of reactions or substantially changing parameters. We propose that evolution involving chemical switches must navigate a complex landscape involving multiple forms of robustness, and the only way for a given switch to have a systematic advantage in robustness is to come from a good family of mirrorsymmetric topologies. Significance StatementLife endures despite metabolic fluctuations and environmental assaults. For the thousands of cellular decisions to continue to work, they must be robust to these perturbations. Many cellular decisions are made and stored by chemical switches, which like light switches retain their state - on or off - even after the trigger is gone. We computationally explored what makes chemical switches robust. It turns out that some are robust to thermal noise, others to mutations that disable part of the switch, or to changes in chemical conditions. Surprisingly, these different forms of robustness are mostly independent. However, chemical switches come in families built around core reactions, and these families tend to score high or low on several measures of robustness.

systems biology↗

bioRxiv2018

Subunit exchange enhances information retention by CaMKII in dendritic spines

Molecular bistables are strong candidates for long-term information storage, for example, in synaptic plasticity. CaMKII is a highly expressed synaptic protein which has been proposed to form a molecular bistable switch capable of maintaining its state for years despite protein turnover and stochastic noise. It has recently been shown that CaMKII holoenzymes exchange subunits among themselves. Here we used computational methods to analyze the effect of subunit exchange on the CaMKII pathway in the presence of diffusion in two different microenvironments, the Post Synaptic Density (PSD) and spine cytosol. We show that in the PSD, subunit exchange leads to coordinated switching and prolongs state stability of the fraction of CaMKII that is present in clusters; and underlies spreading of activation among the remaining CaMKII that is uniformly distributed. Subunit exchange increases the robustness of the CaMKII switch measured as range of bistability both with respect to protein phosphatase 1 (PP1) levels and protein turnover rates. In the phosphatase-rich spine cytosol, subunit exchange leads to slower decay of activity following calcium stimuli. We find that subunit exchange can explain two time-courses of CaMKII activity decay observed in recent experiments monitoring endogenous activity of CaMKII in the spine. Overall, CaMKII exhibits multiple timescales of activity in the synapse and subunit exchange enhances the information retention ability of CaMKII by improving the stability of its switching in the PSD, and by slowing the decay of its activity in the spine cytosol. The existence of diverse timescales in the synapse has important theoretical implications for memory storage in networks.\n\nSignificance StatementDespite everyday forgetfulness, we can recall some memories years after they were formed. How are we able to protect some memories for so long? Previous work has shown that the abundant brain protein Calcium/calmodulin dependent protein Kinase II (CaMKII) can form a very stable binary switch which can store information for years. Building on this work, we analyzed the implications of a recently discovered phenomenon of subunit exchange on the state switching properties of CaMKII. In subunit exchange fragments of one CaMKII molecule detatch and exchange with another. We discovered that this improves the information retention ability of CaMKII both in the context where it stores information for long times, and also where it integrates information over the timescale of minutes.

neuroscience↗

bioRxiv2017

Precise excitation-inhibition balance controls gain and timing in hippocampus

Excitation-inhibition (EI) balance controls excitability, dynamic range, and input gating in many brain circuits. Subsets of synaptic input can be selected or gated by precise modulation of finely tuned EI balance, but assessing the granularity of EI balance requires combinatorial analysis of excitatory and inhibitory inputs. Using patterned optogenetic stimulation of hippocampal CA3 neurons, we show that hundreds of unique CA3 input combinations recruit excitation and inhibition with a nearly identical ratio, demonstrating precise EI balance at the hippocampus. Crucially, the delay between excitation and inhibition decreases as excitatory input increases from a few synapses to tens of synapses. This creates a dynamic millisecond-range window for postsynaptic excitation, controlling membrane depolarization amplitude and timing via subthreshold divisive normalization. We suggest that this combination of precise EI balance and dynamic EI delays forms a general mechanism for millisecond-range input gating and subthreshold gain control in feedforward networks.

neuroscience↗