Identification of Neuronal Polarity by Node-Based Machine Learning

Chen-Zhi Su,Chiau-Jou Li,Kuan-Ting Chou,Chung-Chuan Lo,Hsuan-Pei Huang,Daw-Wei Wang,Ching-Che Charng

doi:10.1007/s12021-021-09513-y

Chen-Zhi Su, Chiau-Jou Li + Show 5 more

Open Access

https://doi.org/10.1007/s12021-021-09513-y

Copy DOI

Abstract

Identifying the direction of signal flows in neural networks is important for understanding the intricate information dynamics of a living brain. Using a dataset of 213 projection neurons distributed in more than 15 neuropils of a Drosophila brain, we develop a powerful machine learning algorithm: node-based polarity identifier of neurons (NPIN). The proposed model is trained only by information specific to nodes, the branch points on the skeleton, and includes both Soma Features (which contain spatial information from a given node to a soma) and Local Features (which contain morphological information of a given node). After including the spatial correlations between nodal polarities, our NPIN provided extremely high accuracy (>96.0%) for the classification of neuronal polarity, even for complex neurons with more than two dendrite/axon clusters. Finally, we further apply NPIN to classify the neuronal polarity of neurons in other species (Blowfly and Moth), which have much less neuronal data available. Our results demonstrate the potential of NPIN as a powerful tool to identify the neuronal polarity of insects and to map out the signal flows in the brain’s neural networks if more training data become available in the future.

Highlights

Rapid technology advances in recent years have led to the development of several connectomic projects and large-scale databases for cellular-level neural images (Chiang et al 2011; Chen-Zhi Su and Kuan-Ting Chou contributed to this work.Kuan et al 2015; Milyaev et al 2012; Parekh and Ascoli 2013; Peng et al 2015; Shinomiya et al 2011; Xu et al, 2013; Xu et al 2020)
From results shown in Figs. 5(a3) and 5(b3), we discover that node-based polarity identifier of neurons (NPIN) is a very powerful classifier with an overall accuracy of 96%
These two values become 2.2% and 5.7%, respectively, for deep neural networks (DNN). This clearly implies that Local Features which are included in Model I are more important for complex neurons compared to simple neurons

Summary

Introduction

Rapid technology advances in recent years have led to the development of several connectomic projects and large-scale databases for cellular-level neural images (Chiang et al 2011; Chen-Zhi Su and Kuan-Ting Chou contributed to this work.Kuan et al 2015; Milyaev et al 2012; Parekh and Ascoli 2013; Peng et al 2015; Shinomiya et al 2011; Xu et al, 2013; Xu et al 2020). How to integrate and transform the data to address scientific questions (Lo and Chiang 2016) remains a central challenge. Overall, these projects aim to provide sufficient information for the analysis of information flows in the brain. The axon-dendrite polarity of a neuron can be identified by experimental methods (Craig and Banker 1994; Matus et al 1981; Wang et al 2004) These methods are not practical for large-scale neural image projects and for the image datasets that were already acquired. Morphology-based polarity identification at the post-imaging stage is possible, but this is challenging for insects because of their highly diverse neuronal morphology (Cuntz et al 2008; Lee et al 2014)

Methods

Results

Discussion

Conclusion