Clustering method and system for unnormalized single-cell transcriptome sequencing data

Abstract

The present disclosure provides a clustering method and system for non-standardized single-cell transcriptome sequencing data, which acquires single-cell transcriptome sequencing data; preprocesses the acquired sequencing data; performs dimensionality reduction and clustering on the preprocessed sequencing data Class processing to obtain clustering results; arrange the clustering results according to Spearman correlation from small to large or from large to small, and delete the cluster results with small Spearman correlation from the gap where the Spearman correlation changes the most; Perform hierarchical clustering on the average value of the equivalence relationship matrix of each clustering result to obtain the final clustering result; the present disclosure makes the clustering result participating in the clustering fusion and other clustering results far behind before the clustering fusion stage. The anomalous clustering results are eliminated, thereby improving the performance of cluster fusion.

Description

technical field [0001] The present disclosure relates to the technical field of biological cell processing, and in particular, to a clustering method and system for non-standardized single-cell transcriptome sequencing data. Background technique [0002] The statements in this section merely provide background related to the present disclosure and do not necessarily constitute prior art. [0003] Single-cell sequencing technologies are widely used in practical studies such as discovering differentiation relationships between cells and differences in gene expression among different types of cells. Downstream analysis of these single-cell sequencing technologies is often based on unsupervised clustering of cells. [0004] The inventors found that the traditional cluster fusion-based single-cell transcriptome sequencing data clustering method introduces many poor clustering results when generating a clustering result set participating in clustering fusion, making the final clus...

AI Technical Summary

Problems solved by technology

[0004] The inventors found that the traditional clustering fusion-based single-cell transcriptome sequencing data clustering method introduces more poor clustering results when generating clustering result sets participating in clustering fusion, making the final clustering results inaccurate

Images