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

Jingqi Lu; Keqi Han; Yun Wang; Lu Mi; Carl Yang

q-bio.NCFeb 19, 2026

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

Authors: Jingqi Lu, Keqi Han, Yun Wang, Lu Mi, Carl Yang

Organizations: Department of Computer Science, Emory University, USA · Department of Biomedical Informatics, Emory University, USA · College of AI, Tsinghua University, China

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.

Figures & tables

Type	Model	Spatial Feature		Neuronal Activity Feature		Connection Feature
Type	Model	Accuracy	F1	Accuracy	F1	Accuracy	F1
Non-graph method	Logistic Regression	$56.33_{\pm 4.52}$	$54.17_{\pm 4.38}$	$36.59_{\pm 6.20}$	$17.86_{\pm 6.11}$	$54.67_{\pm 5.62}$	$52.49_{\pm 6.56}$
	MLP	$56.67_{\pm 2.98}$	$55.52_{\pm 3.69}$	$40.00_{\pm 5.25}$	$31.81_{\pm 4.93}$	$54.67_{\pm 1.94}$	$53.00_{\pm 1.75}$
	LOLCAT	$57.00_{\pm 2.67}$	$55.86_{\pm 2.31}$	$39.00_{\pm 5.83}$	$25.11_{\pm 8.84}$	$54.00_{\pm 2.26}$	$52.78_{\pm 2.28}$
	NeuPRINT	$55.33_{\pm 9.33}$	$50.87_{\pm 10.61}$	$45.57_{\pm 8.69}$	$37.19_{\pm 14.18}$	$50.67_{\pm 4.90}$	$42.51_{\pm 13.08}$
Graph method	GCN	$60.00_{\pm 6.65}$	$54.38_{\pm 8.16}$	$45.00_{\pm 8.50}$	$40.53_{\pm 9.83}$	$58.05_{\pm 7.14}$	$44.87_{\pm 12.29}$

Table 1: Performance Comparison of Graph and Non-graph Methods for Neuron Classification Across Three Feature Sets. For each metric (column), the highest score is in bold and the second-highest score is underlined .

Panel of Figure 1 — Figure 1: Visualization of C. elegans neuron data features and class distributions. (a) Example neuronal activity data for representative Sensory, Interneuron, and Motor neurons. (b) Normalized probability density distributions of cosine similarity, comparing the Spatial and Connection features across all neuron pairs. (c) Magnified view of the spatial distribution of neuron classes in the head region. (d) Spatial distribution of the three neuron classes along the entire body.

Explore similar work

DevoTG: Temporal Graph Neural Networks for Modeling C. elegans Developmental Connectomics
Jun 20, 2026Jayadratha Gayen, Bradly AliceaTemporal GNNs Dynamic Graph Learning
1IIIT Hyderabad · 2OpenWorm Foundation · 3Orthogonal Research and Education Laboratory
Benchmarking the Connectomes of Caenorhabditis elegans within the Reservoir Computing Framework
Sep 24, 2026Felix S. Reimers, Ola Huse Ramstad, Aliaksandr Hubin +1Brain Network Analysis Reservoir Computing
Department of Computer Science and Communication, Østfold University College, Halden, 1757, Østfold, Norway. · Faculty of Chemistry, Biotechnology and Food Science, Norwegian University of Life Sciences, ˚As, 1432, Akershus, Norway. · Department of Mathematics, University of Oslo, Oslo, 0316, Oslo, Norway.
A Temporally Augmented Graph Attention Network for Affordance Classification
Apr 11, 2026Ami Chopra, Supriya Bordoloi, Shyamanta M. HazarikaGraph Attention Networks Temporal GNNs
Indian Institute of Technology Guwahati, India

Jun 20, 2026J/K move · Enter open · S save

DevoTG: Temporal Graph Neural Networks for Modeling C. elegans Developmental Connectomics

Jayadratha Gayen, Bradly Alicea

1IIIT Hyderabad · 2OpenWorm Foundation · 3Orthogonal Research and Education Laboratory

Temporal GNNs Dynamic Graph Learning

Open paper AskPDF arXiv

Understanding how a nervous system wires itself from birth to adulthood is a fundamental challenge in developmental neuroscience. We present DevoTG, a temporal graph framework that applies Temporal Graph Neural Networks (TGNs) to two complementary representations of C. elegans neural development: a Continuous-Time Dynamic Graph (CTDG) of cell division events derived from cell lineage data, and a Discrete-Time Dynamic Graph (DTDG) of the developing synaptic connectome spanning eight reconstructed electron-microscopy datasets. On the lineage prediction task, our TGN achieves a mean test AUC of 0.839 +/- 0.007 (5 seeds; validation AUC 0.937 +/- 0.001), outperforming a static GNN with the identical architecture by 26 AUC points (0.577 +/- 0.080), demonstrating that temporal memory is the decisive factor. Applied to the connectome DTDG, DevoTG identifies three connection stability classes (stable, developmental, and variable) across 225 neurons and 858 to 2,496 connections over development (L1 birth to adult), providing a temporal-graph-theoretic complement to the individual-variability classification of Witvliet et al. Analysis of hub command interneurons AVA, AVB, and AVE reveals their persistent centrality and how their integration roles are progressively reinforced across larval stages. Accompanying interactive visualizations (3D animated networks, centrality heatmaps, and a spatiotemporal lineage graph) make developmental dynamics accessible for biological hypothesis generation. DevoTG is open-source and designed for extension to other developing nervous systems. Code is publicly available at https://github.com/DevoLearn/DevoGraph/tree/main/DevoTG.

Abstract

Explore similar work

DevoTG: Temporal Graph Neural Networks for Modeling C. elegans Developmental Connectomics

Benchmarking the Connectomes of Caenorhabditis elegans within the Reservoir Computing Framework

A Temporally Augmented Graph Attention Network for Affordance Classification