A Benchmark Analysis of Graph and Non-Graph Methods for Caenorhabditis Elegans Neuron Classification

Researchers conducted a benchmark study comparing graph neural networks (GNNs) against traditional methods for classifying neurons in C. elegans worms. The study found that attention-based GNNs significantly outperformed baseline methods when using spatial and connection features, validating the effectiveness of graph-based approaches for biological neural network analysis.

• →Attention-based graph neural networks (GCN, GraphSAGE, GAT, GraphTransformer) significantly outperformed traditional machine learning methods for neuron classification.
• →Spatial and connection features proved to be key predictors for classifying sensory, interneuron, and motor neurons in C. elegans.
• →Neuronal activity features yielded poor classification performance due to low temporal resolution limitations in the underlying data.
• →The benchmark establishes a standardized framework for comparing different approaches to biological neural network classification.
• →Code and methodology are publicly available, enabling reproducible research in computational neuroscience.