arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.12482stat.MLcs.LGq-bio.NC
收藏

神经元形态的自监督图表示学习

Self-Supervised Graph Representation Learning for Neuronal Morphologies

University of Göttingen(哥廷根大学);University of Tübingen(图宾根大学);Max Planck Institute for Dynamics and Self-Organization(马克斯·普朗克动力学与自组织研究所)

Marissa A. Weis, Laura Hansel, Timo Lüddecke, Alexander S. Ecker

AI总结:

本文提出GraphDINO,一种基于Transformer的自监督图表示学习方法,通过空间图增强、位置编码调整和AC-Attention机制学习三维神经元形态低维表示,在无先验知识下实现与专家分类相当的细胞类型聚类,并可用于一般图级嵌入任务。

摘要:

无监督图表示学习近年来在神经科学等多个应用领域引起了关注,在这些领域中,对大脑中细胞类型的多样形态进行建模是关键挑战之一。目前尚不清楚存在多少种兴奋性皮层细胞类型,以及它们的定义性形态特征是什么。在此,我们提出GraphDINO,这是一种纯数据驱动的方法,用于从未标记的大规模数据集中学习三维神经元形态的低维表示。GraphDINO是一种新颖的基于Transformer的表示学习方法,适用于空间嵌入图。为了实现Transformer上的自监督学习,我们(1)开发了针对空间嵌入图的数据增强策略,(2)调整了位置编码,并(3)引入了一种新颖的注意力机制AC-Attention,该机制结合了基于注意力的节点间全局交互与经典图卷积处理。我们在两个不同物种和多个脑区中表明,该方法产生的形态学细胞类型聚类与专家基于手工特征的分类相当,但未使用关于神经元结构特征的先验知识。此外,在预测专家标签的定量基准上,它优于先前的方法。我们的方法有可能在大规模数据集中实现数据驱动的新形态特征和细胞类型发现。它可应用于神经科学以外的场景,在这些场景中,数据集中的样本是图,并且需要图级嵌入。

英文摘要:

Unsupervised graph representation learning has recently gained interest in several application domains such as neuroscience, where modeling the diverse morphology of cell types in the brain is one of the key challenges. It is currently unknown how many excitatory cortical cell types exist and what their defining morphological features are. Here we present GraphDINO, a purely data-driven approach to learn low-dimensional representations of 3D neuronal morphologies from unlabeled large-scale datasets. GraphDINO is a novel transformer-based representation learning method for spatially-embedded graphs. To enable self-supervised learning on transformers, we (1) developed data augmentation strategies for spatially-embedded graphs, (2) adapted the positional encoding and (3) introduced a novel attention mechanism, AC-Attention, which combines attention-based global interaction between nodes and classic graph convolutional processing. We show, in two different species and across multiple brain areas, that this method yields morphological cell type clusterings that are on par with manual feature-based classification by experts, but without using prior knowledge about the structural features of neurons. Moreover, it outperforms previous approaches on quantitative benchmarks predicting expert labels. Our method could potentially enable data-driven discovery of novel morphological features and cell types in large-scale datasets. It is applicable beyond neuroscience in settings where samples in a dataset are graphs and graph-level embeddings are desired.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑