Identifying cell types from single-cell data based on similarities and dissimilarities between cells

Abstract

Background With the development of the technology of single-cell sequence, revealing homogeneity and heterogeneity between cells has become a new area of computational systems biology research. However, the clustering of cell types becomes more complex with the mutual penetration between different types of cells and the instability of gene expression. One way of overcoming this problem is to group similar, related single cells together by the means of various clustering analysis methods. Although some methods such as spectral clustering can do well in the identification of cell types, they only consider the similarities between cells and ignore the influence of dissimilarities on clustering results. This methodology may limit the performance of most of the conventional clustering algorithms for the identification of clusters, it needs to develop special methods for high-dimensional sparse categorical data. Results Inspired by the phenomenon that same type cells have similar gene expression patterns, but different types of cells evoke dissimilar gene expression patterns, we improve the existing spectral clustering method for clustering single-cell data that is based on both similarities and dissimilarities between cells. The method first measures the similarity/dissimilarity among cells, then constructs the incidence matrix by fusing similarity matrix with dissimilarity matrix, and, finally, uses the eigenvalues of the incidence matrix to perform dimensionality reduction and employs the K-means algorithm in the low dimensional space to achieve clustering. The proposed improved spectral clustering method is compared with the conventional spectral clustering method in recognizing cell types on several real single-cell RNA-seq datasets. Conclusions In summary, we show that adding intercellular dissimilarity can effectively improve accuracy and achieve robustness and that improved spectral clustering method outperforms the traditional spectral clustering method in grouping cells.

Deconer: A comprehensive and systematic evaluation toolkit for reference-based cell type deconvolution algorithms using gene expression data

“…In the fine_bulk dataset, these three cell types are further subdivided into memory B cells, naï ve B cells, memory CD4 T cells, naï ve CD4 T cells, memory CD8 T cells, naï ve CD8 T cells, and regulatory T cells. The functionally similar cell types share similar gene expressions, which can lead to increased cell-to-cell confusion in the deconvolution process [53,55,56]. Simultaneously, unlike the mouse_tissue dataset where each cell type has 1500 high-quality sequenced cells, the cell number of each type in the human_PBMC dataset varies significantly.…”

Single-Cell DNA Methylation Analysis in Cancer

“…Furthermore, pairwise dissimilarity analyses can be performed, which highlight the heterogeneity of the sequenced tumour. Pairwise dissimilarity matrices differ from other clustering methods, as it appreciates the influence of dissimilarities between cells, as well as the similarities [ 134 ]. Appreciating both positive and negative correlations between cells helps infer the epigenetic entropy of different cells within the tumour, showing how individual cells’ epigenetic states differ.…”

Equipment Pattern Recognition of Unbalanced Fuel Consumption Data Based on Grouping Multi-BP Neural Network

“…To assess the performance of GMBPN algorithm and further find the indicator of significant effect, the real HFC dataset on some large equipment are adopted. The original HFC data of equipment in six patterns is shown in Supplementary Table . To evaluate the performance of clustering algorithm, the adjusted Rand index (ARI) and Normalized Mutual Information (NMI) are widely used to measure accuracy and similarity between the inferred labels and reference labels [21], [22].…”