PLX131283

GSE135992: Identification and elimination of the platelet-derived transcripts contamination in circulating tumor cell single-cell RNA-seq study

• Organsim human
• Type RNASEQ
• Target gene
• Project ARCHS4

Single-cell RNA sequencing (scRNA-seq) holds great potential to revolutionize our understanding of cancer biology by discovering biomarkers and biological functions from circulating tumor cells (CTCs). However, the contamination from blood cells is one of the main challenges in CTC studies. Through investigation of recently published studies, we found that several previously reported CTC biomarkers were apparently contaminated by blood-derived transcripts such as HBB, HBA1/2, ITGA2B, and LCP1. Through re-analysis of several published CTC and cell line datasets, we revealed that CTC biomarkers were not only highly expressed in blood, but showed a high correlation between CTC and blood. We sequenced transcriptome of seven blood components from a healthy donor, finding that the platelet exhibited the highest probability to confound CTC data in terms of expression profiling and correlation analysis. Using CTC spike-in and recovery assay and we showed that the platelet induced artifacts could be reproduced in vitro. Surprisingly, the reproduced artifacts were also highly expressed in the platelet data but not in the other six blood components. We performed SNP calling on recovered CTC simulators, finding cross-contamination of exogenous transcripts. Furthermore, we applied a multiple linear regression model to estimate the platelet confounding index, showing that previously published CTC datasets were confounded with 13%~26% of platelet fractions and 7.9%~17% platelet cloaking judged by fluorescence flow cytometry. We devised a red blood cell lysis based protocol that better removed platelet compared to several platelet inhibitory reagents and kept transcriptomic profile undisturbed in CTC scRNA-seq. This work not only demonstrated the potential blood cell confounding factors in CTC studies, validated them in the past published datasets, and reproduced the artifacts in CTC simulation assay, but also devised an easily applicable method to remove the platelet cloaking, guaranteeing the reliability of CTC scRNA-seq research. SOURCE: Sibo Zhu (sibozhu@fudan.edu.cn) - Fudan University