Artificial-cell-type aware cell-type classification in CITE-seq

Abstract

Motivation Cellular Indexing of Transcriptomes and Epitopes by sequencing (CITE-seq), couples the measurement of surface marker proteins with simultaneous sequencing of mRNA at single cell level, which brings accurate cell surface phenotyping to single-cell transcriptomics. Unfortunately, multiplets in CITE-seq datasets create artificial cell types (ACT) and complicate the automation of cell surface phenotyping. Results We propose CITE-sort, an artificial-cell-type aware surface marker clustering method for CITE-seq. CITE-sort is aware of and is robust to multiplet-induced ACT. We benchmarked CITE-sort with real and simulated CITE-seq datasets and compared CITE-sort against canonical clustering methods. We show that CITE-sort produces the best clustering performance across the board. CITE-sort not only accurately identifies real biological cell types (BCT) but also consistently and reliably separates multiplet-induced artificial-cell-type droplet clusters from real BCT droplet clusters. In addition, CITE-sort organizes its clustering process with a binary tree, which facilitates easy interpretation and verification of its clustering result and simplifies cell-type annotation with domain knowledge in CITE-seq. Availability and implementation http://github.com/QiuyuLian/CITE-sort. Supplementary information Supplementary data is available at Bioinformatics online.

SECANT: a biology-guided semi-supervised method for clustering, classification, and annotation of single-cell multi-omics

“…To overcome this challenge, our method is designed to utilize cells with uncertain cell type label. It is worth noting that tools other than manual gating can also be used for identifying confident cell types with ADT data ( 24 , 25 , 27 ).…”

“…However, by doing so, the important underlying biological information from surface protein marker could be undermined. On the contrary, biological researchers often consider cell surface markers as the gold standard to define cell types in molecular biology, where researchers identify distinguished cell types through cell gating such as flow cytometry with a list of classic differentiation (CD) markers, such as CD3, CD4, CD8, and CD19 ( 24 , 25 , 26 , 27 , 28 ). Thus, a more biological knowledge-driven approach should consider putting more weight on ADT data for the purpose of cell clustering and cell type identification.…”

“…Our method utilizes a model-based approach motivated by classic statistical models in semi-supervised learning ( 33 ). As a biological knowledge-driven approach, the input of our method from ADT data is the confident cell type label, which can be obtained through cell gating or other existing methods ( 24 , 25 , 26 , 27 , 28 ). To overcome a common issue in cell gating that there are always cells distributed near or on the gating boundaries (e.g.…”

Normalizing and denoising protein expression data from droplet-based single cell profiling

“…In addition, the specific noise sources revealed by our analyses and approaches to estimate them could lead to more informative prior distributions used by Bayesian probabilistic modeling approaches such as TotalVI. As demonstrated here, the denoised and normalized data from dsb can be used in any downstream analysis application to potentially enhance the results of higher level single cell data analysis methods, such as joint protein–mRNA clustering 31 , 43 – 45 .…”

Acquisition of murine splenic myeloid cells for protein and gene expression profiling by advanced flow cytometry and CITE-seq

“…CITE-seq data can be analyzed using standard single cell bioinformatic workflows, such as Seurat ( Butler et al., 2018 ) or Scanpy ( Wolf et al., 2018 ). Examples of integrated analysis platforms for multi-omic data are weighted-nearest neighbor analysis ( Hao et al., 2021 ), One-SENSE ( Cheng et al., 2016 ; Mair et al., 2020 ), CiteFuse ( Kim et al., 2020 ) and CITE-sort ( Lian et al., 2020 ).…”