BunDLe-Net: Neuronal Manifold Learning Meets Behaviour

Abstract

Neuronal manifold learning techniques represent high-dimensional neuronal dynamics in low-dimensional embeddings to reveal the intrinsic structure of neuronal manifolds. Common to these techniques is their goal to learn low-dimensional embeddings that preserve all dynamic information in the high-dimensional neuronal data, i.e., embeddings that allow for reconstructing the original data. We introduce a novel neuronal manifold learning technique, BunDLe-Net, that learns a low-dimensional Markovian embedding of the neuronal dynamics which preserves only those aspects of the neuronal dynamics that are relevant for a given behavioural context. In this way, BunDLe-Net eliminates neuronal dynamics that are irrelevant to decoding behaviour, effectively de-noising the data to reveal better the intricate relationships between neuronal dynamics and behaviour. We demonstrate the quantitative superiority of BunDLe-Net over commonly used and state-of-the-art neuronal manifold learning techniques in terms of dynamic and behavioural information in the learned manifold on calcium imaging data recorded in the nematode C. elegans. Qualitatively, we show that BunDLe-Net learns highly consistent manifolds across multiple worms that reveal the neuronal and behavioural motifs that form the building blocks of the neuronal manifold.

Neuro-Cognitive Multilevel Causal Modeling: A Framework that Bridges the Explanatory Gap between Neuronal Activity and Cognition

“…The goal of neuronal manifold learning techniques is to find low-dimensional representations of neuronal data that enable insights into the structure of neuronal dynamics and their relation to behavior (Mitchell-Heggs et al, 2023). In neuroscience, classic dimensionality reduction techniques, such as principal component analysis (PCA), Laplacian eigenmaps (LEM), and t-SNE, are complemented by modern deep learning techniques such as CEBRA (Schneider et al, 2023) and BundDLe-Net (Kumar et al, 2023). In the following, we show that the cognitive-behavioral state diagrams introduced in the previous section can be interpreted as a neuronal manifold learning technique that represents the essential aspects of the manifold as a directed graph.…”

“…Figure 7 shows a side-by-side comparison of the neuronal manifold of the third worm as inferred by BundDLe-Net (Kumar et al, 2023) (left column) and the cognitive-behavioral state diagram of the NC-MCM framework (right column)8. Looking at the left column, it is immediately apparent that the neuronal manifold exhibits two main cycles, a revsus (gray) → vt (turquoise) → slow (red) → rev1 (green) → revsus (gray) and a revsus (gray) → dt (orange) → slow (red) → fwd (yellow) → rev2 (blue) → revsus (gray) cycle, that correspond to the behavioral motifs revealed by the cognitive-behavioral state diagram in the right column of Figure 7 and discussed in the previous section.…”

“…Due to increasing awareness of the importance of (potentially widely distributed) neuronal activity patterns for behavior, and the difficulty in scaling up hand-crafted models to large-scale neuronal recordings, machine learning methods (also referred to as decoding- or multivariate pattern analysis (MVPA) models) have been developed to uncover relations between complex neuronal activity patterns, cognitvie states, and behaviors (Norman et al, 2006; Mitchell et al, 2008; Pereira et al, 2009). More recently, these models have been complemented by algorithms for learning neuronal manifolds that enable the visualization of high-dimensional neuronal dynamics (Mitchell-Heggs et al, 2023) and their relation to behavior (Schneider et al, 2023; Kumar et al, 2023). However, the ability to visualize and decode behavior from complex neuronal activity patterns does not imply that we have revealed their representational contents (Ritchie et al, 2020).…”

“… 8 An in-depth comparison of BundDLe-Net with other neuronal manifold learning algorithms on the present data set is available in Kumar et al (2023).…”