English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Single-cell RNA sequencing (scRNA-seq) technology provides high-throughputgene expression data to study the cellular heterogeneity and dynamics ofcomplex organisms. Graph neural networks (GNNs) have been widely used forautomatic cell type classification, which is a fundamental problem to solve inscRNA-seq analysis. However, existing methods do not sufficiently exploit bothgene-gene and cell-cell relationships, and thus the true potential of GNNs isnot realized. In this work, we propose a bilevel graph representation learningmethod, named scBiGNN, to simultaneously mine the relationships at both geneand cell levels for more accurate single-cell classification. Specifically,scBiGNN comprises two GNN modules to identify cell types. A gene-level GNN isestablished to adaptively learn gene-gene interactions and cell representationsvia the self-attention mechanism, and a cell-level GNN builds on the cell-cellgraph that is constructed from the cell representations generated by thegene-level GNN. To tackle the scalability issue for processing a large numberof cells, scBiGNN adopts an Expectation Maximization (EM) framework in whichthe two modules are alternately trained via the E-step and M-step to learn fromeach other. Through this interaction, the gene- and cell-level structuralinformation is integrated to gradually enhance the classification performanceof both GNN modules. Experiments on benchmark datasets demonstrate that ourscBiGNN outperforms a variety of existing methods for cell type classificationfrom scRNA-seq data.

scBiGNN: Bilevel Graph Representation Learning for Cell Type Classification from Single-cell RNA Sequencing Data