The Unreasonable Effectiveness of Cell Types in Describing Neuronal Physiological Features
- 1Allen Institute
Abstract
Single-cell RNA sequencing (scRNA-seq) captures detailed gene expression profiles at scale, while patch-clamp recordings measure intrinsic neuronal electrophysiological properties. Modeling the relations between these two modalities remains a challenge. Here, we compare how well electrophysiological features can be predicted by traditional transcriptomic cell type classification, representations derived from a foundational model (scGPT) pretrained on large-scale scRNA-seq datasets, ion channel-coding genes, and highly variable genes. Using paired transcriptomic and electrophysiological patch-sequencing data from 495 human neurons from neurosurgical tissue, we find that cluster-level cell type representations generally outperform highly variable gene selection, ion channel gene selection, and context-enriched scGPT embeddings. Notably, performance varies across model architectures and initializations, and the best results are obtained by combining the outputs of separate cell type and scGPT-based models. Together, these findings suggest that traditional discrete cellular classification is highly effective in predicting physiological features. For maximum performance it can be complemented by pretrained transformer models.
Related
- Physiological brain clearance architecture revealed by neuronal protein tracing
- Disease-associated variants alter a novel physiological role of hnRNP H2 in neuronal development
- Concurrent cognitive, behavioral and systemic physiological features mark preclinical Alzheimer's disease
- The unreasonable likelihood of being: origin of life, terraforming, and AI
- Describing the emotional change experienced by patients with chronic somatic diseases who took part in a supportive program with complementary therapies: a qualitative study
BibTeX
@misc{sharma2026unreasonable,
title = {{The Unreasonable Effectiveness of Cell Types in Describing Neuronal Physiological Features}},
author = {H. Sharma and X.-P. Liu and T. Chartrand and B. Kalmbach and E. Lein and S. Mihalas and Z. Lu},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.27.744753},
url = {https://www.biorxiv.org/content/10.64898/2026.08.27.744753v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - The Unreasonable Effectiveness of Cell Types in Describing Neuronal Physiological Features
AU - H. Sharma
AU - X.-P. Liu
AU - T. Chartrand
AU - B. Kalmbach
AU - E. Lein
AU - S. Mihalas
AU - Z. Lu
PY - 2026
DA - 2026/08/27
PB - bioRxiv
DO - 10.64898/2026.08.27.744753
UR - https://www.biorxiv.org/content/10.64898/2026.08.27.744753v1
AB - Single-cell RNA sequencing (scRNA-seq) captures detailed gene expression profiles at scale, while patch-clamp recordings measure intrinsic neuronal electrophysiological properties. Modeling the relations between these two modalities remains a challenge. Here, we compare how well electrophysiological features can be predicted by traditional transcriptomic cell type classification, representations derived from a foundational model (scGPT) pretrained on large-scale scRNA-seq datasets, ion channel-coding genes, and highly variable genes. Using paired transcriptomic and electrophysiological patch-sequencing data from 495 human neurons from neurosurgical tissue, we find that cluster-level cell type representations generally outperform highly variable gene selection, ion channel gene selection, and context-enriched scGPT embeddings. Notably, performance varies across model architectures and initializations, and the best results are obtained by combining the outputs of separate cell type and scGPT-based models. Together, these findings suggest that traditional discrete cellular classification is highly effective in predicting physiological features. For maximum performance it can be complemented by pretrained transformer models.
ER -