Inferring Functional Circuitry from Neural Time Series Based on Markovian Structural Causal Model
Poster No:
1266
Submission Type:
Late-Breaking Abstract Submission
Authors:
Rahul Biswas1, SuryaNarayana Sripada2, Somabha Mukherjee3, Reza Abbasi-Asl1
Institutions:
1Department of Neurology, University of California, San Francisco, CA, 2Center for Science and Consciousness, Redmond, WA, 3Department of Statistics & Data Science, National University of Singapore, Singapore
First Author:
Co-Author(s):
Somabha Mukherjee, PhD
Department of Statistics & Data Science, National University of Singapore
Singapore
Department of Statistics & Data Science, National University of Singapore
Singapore
Late Breaking Reviewer(s):
Introduction:
Mapping functional circuitry is essential for understanding brain function and dysfunction. Unlike traditional functional connectivity, which captures statistical associations, functional circuitry represents causal signal propagation in brain networks. For causal inference of functional circuitry, many methods rely on parametric models, such as vector autoregressive models in Granger Causality or additive non-linear models, assuming specific dynamical equations. This limits generalizability across non-linear, non-Gaussian neural interactions and recording paradigms. We introduce Causal Inference in Time Series (CITS), a non-parametric method for inferring causal functional circuitry from neural time series based on a Markovian Structural Causal Model (SCM) of arbitrary order. CITS does not assume a specific functional form for interactions or noise distribution, enabling its application across diverse neural datasets. We establish mathematical guarantees, validate on simulated data, and apply CITS to Neuropixels data, extracting directed functional pathways in mouse sensory processing.
Methods:
CITS infers functional circuitry from neural signals based on a Markovian Structural Causal Model (SCM) of arbitrary order τ, performing causal inference via conditional dependence tests in time series data. Specifically, CITS tests whether neuron u at time s remains conditionally dependent on neuron v at time t (for s≤t) after conditioning on subsets of neurons in the preceding 2τ window. For the conditional dependence test, we use Fisher's Z-test for Gaussian settings and the Hilbert-Schmidt independence criterion for non-Gaussian settings. We establish mathematical guarantees that CITS recovers the true causal structure under broad dependence conditions, ensuring reliable inference. CITS was validated on simulated data from linear Gaussian, non-linear non-Gaussian, and Continuous-Time Recurrent Neural Network (CTRNN) models, evaluated using true/false positive rates and combined score (true positive rate minus false positive rate). We applied CITS to Neuropixels data from the Allen Visual Coding Neuropixels database, analyzing a 116-day-old male mouse (Session ID 791319847) with 555 neurons recorded at 1 kHz across six probes. Spike trains were analyzed across visual stimuli: natural scenes, static gratings, Gabor patches, and full-field flashes to infer directed functional pathways.
Results:
We evaluated CITS against Pairwise and Multivariate Granger Causality, naive Peter-Clark (PC), and Time-Aware PC on simulated and real neural datasets. In simulations, CITS outperformed other methods across linear and nonlinear models in combined score (see Figure 1). Even in the presence of a common cause with Gaussian noise, CITS had a combined score of 99% while other methods failed to distinguish direct from indirect effects leading to their score of below 45%. Across all trials, edge weights reflected excitatory or inhibitory connections in the ground truth. Applied to Neuropixels recordings from the Allen Brain Observatory, CITS extracted distinct causal functional connectivity (CFC) patterns across stimuli. Natural scenes evoked stronger connectivity in Primary Visual Cortex, static gratings in Posteromedial and Anteromedial Visual Cortex, and full-field flashes in Anterolateral Visual Cortex and Thalamus. All stimuli exhibited distinct connectivity in the Cornu Ammonis regions of the hippocampus, with natural scenes and static gratings enhancing Subiculum connectivity.
Conclusions:
CITS provides a non-parametric approach for inferring neural functional circuitry, adaptable to various neural data scenarios. Beyond mapping functional circuitry in the healthy brain, CITS can infer circuit disruptions in neuropsychiatric disorders from non-invasive (fMRI) and minimally invasive (sEEG) recordings, enabling network-based biomarkers without interventions. Its broad applicability makes CITS a powerful tool for basic and clinical neuroscience. Code: github.com/biswasr/cits.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 1
Methods Development 2
Multivariate Approaches
Other Methods
Keywords:
Computational Neuroscience
ELECTROPHYSIOLOGY
Modeling
Multivariate
Neuron
Open-Source Code
Statistical Methods
Other - Structural causal model; Functional connectivity; Visual cortex
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Are you Internal Review Board (IRB) certified? Please note: Failure to have IRB, if applicable will lead to automatic rejection of abstract.
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
2. Biswas, R., & Shlizerman, E. (2022). Statistical perspective on functional and causal neural connectomics: The time-aware PC algorithm. PLoS Computational Biology, 18(10), e1010653. https://doi.org/10.1371/journal.pcbi.1010653
3. Reid, A. T., et al. (2019). Advancing functional connectivity research from association to causation. Nature Neuroscience, 22(11), 1751–1760.
4. Barnett, L., Barrett, A. B., & Seth, A. K. (2009). Granger causality and transfer entropy are equivalent for Gaussian variables. Physical Review Letters, 103(23), 238701.
5. Spirtes, P., Glymour, C. N., Scheines, R., & Heckerman, D. (2000). Causation, prediction, and search. MIT Press.
6. Eichler, M. (2013). Causal inference with multiple time series: Principles and problems. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 371(1997), 20110613.
7. Holland, P. W. (1986). Statistics and causal inference. Journal of the American Statistical Association, 81(396), 945–960.
8. Nakamura, A., et al. (2017). Early functional network alterations in asymptomatic elders at risk for Alzheimer’s disease. Scientific Reports, 7, 6517.
9. Stokes, P. A., & Purdon, P. L. (2017). A study of problems encountered in Granger causality analysis from a neuroscience perspective. Proceedings of the National Academy of Sciences, 114(34), E7063–E7072.
10. Finn, E. S., et al. (2015). Functional connectome fingerprinting: Identifying individuals using patterns of brain connectivity. Nature Neuroscience, 18(11), 1664–1671.