MulCNN: An efficient and accurate deep learning method based on gene embedding for cell type identification in single-cell RNA-seq data

Abstract

Advancements in single-cell sequencing research have revolutionized our understanding of cellular heterogeneity and functional diversity through the analysis of single-cell transcriptomes and genomes. A crucial step in single-cell RNA sequencing (scRNA-seq) analysis is identifying cell types. However, scRNA-seq data are often high dimensional and sparse, and manual cell type identification can be time-consuming, subjective, and lack reproducibility. Consequently, analyzing scRNA-seq data remains a computational challenge. With the increasing availability of well-annotated scRNA-seq datasets, advanced methods are emerging to aid in cell type identification by leveraging this information. Deep learning neural networks have great potential for analyzing single-cell data. This paper proposes MulCNN, a multi-level convolutional neural network that uses a unique cell type-specific gene expression feature extraction method. This method extracts critical features through multi-scale convolution while filtering noise. Extensive testing using datasets from various species and comparisons with popular classification methods show that MulCNN has outstanding performance and offers a new and scalable direction for scRNA-seq analysis.

Artificial Intelligence-Assisted Diagnostic Cytology and Genomic Testing for Hematologic Disorders

“…There are emerging AI-assisted single-cell sequencing platforms that will be furthered via validation using clinical samples from patients with hematologic disorders. A multi-level convolutional neural network (MulCNN) was developed by Jiao and his colleagues to provide a unique, single-cell gene expression profile by extracting critical features through multi-scale convolution while filtering noise [ 118 ]. BERMUDA (batch effect removal using deep autoencoders), another novel DL-based method, provides a higher-resolution cellular subtype.…”

scSwinFormer: A Transformer-Based Cell-Type Annotation Method for scRNA-Seq Data Using Smooth Gene Embedding and Global Features

“…29−31 Shao et al 32 applied a weighted graph neural network to scRNA-seq, proposed the scDeepSort, which improved the accuracy of the model. Jiao et al 33 proposed a deep learning cell classification method, MulCNN, based on the multilayer convolutional neural network, using multiscale convolutional pooling operation combined with principal component analysis to extract multidimensional features and trained the model to predict the cell type. Zhou et al 34 took into account the dependency of the feature genes and proposed the scDNN based on the LSTM structure of the scDLC model.…”

“…In recent years, with the development of deep learning and the great progress of neural networks in image recognition, natural language processing, and so forth, as well as the large volume of scRNA-seq data, a number of deep neural network-based methods have been developed for identifying cell types. − Shao et al applied a weighted graph neural network to scRNA-seq, proposed the scDeepSort, which improved the accuracy of the model. Jiao et al . proposed a deep learning cell classification method, MulCNN, based on the multilayer convolutional neural network, using multiscale convolutional pooling operation combined with principal component analysis to extract multidimensional features and trained the model to predict the cell type.…”