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

TL;DR

This paper benchmarks various graph and non-graph machine learning methods for classifying neurons in Caenorhabditis elegans, demonstrating the superiority of attention-based GNNs on certain features and highlighting key predictors for neuron classification.

Contribution

It provides the first comprehensive benchmark comparing graph and non-graph methods for C. elegans neuron classification using multiple feature types.

Findings

Attention-based GNNs outperform baselines on Spatial and Connection features.

Neuronal Activity features perform poorly due to low temporal resolution.

Spatial and Connection features are key predictors for neuron classes.

Abstract

This study establishes a benchmark for Caenorhabditis elegans neuron classification, comparing four graph methods (GCN, GraphSAGE, GAT, GraphTransformer) against four non-graph methods (Logistic Regression, MLP, LOLCAT, NeuPRINT). Using the functional connectome, we classified Sensory, Interneuron, and Motor neurons based on Spatial, Connection, and Neuronal Activity features. Results show that attention-based GNNs significantly outperform baselines on the Spatial and Connection features. The Neuronal Activity features yielded poor performance, likely due to the low temporal resolution of the underlying neuronal activity data. Our benchmark validates the use of GNNs and highlights that Spatial and Connection features are key predictors for Caenorhabditis elegans neuron classes. Code is available at: https://github.com/JingqiLuu/neuronclf-gnn-benchmark.