Trajectory analysis of cancer subpopulationsidentified differentially expressed genes for single-cellRNA seq data of breast cancer

Background: Breast cancer is the most common cancer amongwomen, with an estimated ۲.۳ million new cases diagnosed in۲۰۲۰ [۱]. Single-cell RNA sequencing (scRNA-seq) generatedgene expression profiles at the resolution of an individual cell[۲]. This method has been used to uncover the heterogeneityamong different types of cancers. [۳,۸,۹] Trajectory analysis isa technique used in scRNA-seq research to study changes ingene expression over time. Trajectory inference ordered cellsalong a path and assigned pseudo-time values to represent theirposition. This method helped identify genes associated with lineages.[۷] In this study, we exploit the trajectory method in single-cell RNA seq to investigate differentially expressed genesin cancer subpopulations and dynamic changes.Materials and Methods: The dataset with ID numberGSE۱۵۸۳۹۹ was accessed to select the luminal B breast cancersample for further analysis. The sample had the T۳N۲M۰ stageand had not received treatment. Data were filtered for qualitycontrol using the Seurat package. UMAP method was selectedfor dimension reduction and data visualization. CellMarker۲and SingleR package were applied to label cellular clusters inUMAP. Cancer cells were selected to subcluster, and the Monocle۳package was utilized to perform trajectory analysis oncancer cells.Results: The dataset included ۱۱۴۶۴ cells. After filtering outlow-quality cells, ۱۱۲۸۱ cells remained. Cancer cells with thenumber ۳۸۱۱ were picked for further analysis. Subsetted cellsunderwent normalization and scaling and were then subclusteredinto ۹ clusters. Pseudo times for subclusters were determinedutilizing the Monocle۳ package, and cells of eachsubcluster were ordered based on pseudo times. Differentiallyexpressed genes based on pseudo times were investigated, andCAVIN۱, CCND۱, CPB۱, and NUCKS۱ presented the lowestq.value, and their expressions were dependent on pseudo times.UMAP based on pseudo times plotted for cancer subclusters.Conclusion: Differentially expressed genes based on pseudotimes were detected using trajectory analysis for luminal Bbreast cancer scRNA-seq data.