CITEViz: Replicating the Interactive Flow Cytometry Workflow in CITE-Seq

Garth Kong

Knight Cancer Institute, Oregon Health and Science University

Thai Nguyen

Bioinformatics and Genomics Masters Program, University of Oregon

Wesley Rosales

Bioinformatics and Genomics Masters Program, University of Oregon

Anjali Panikar

Bioinformatics and Genomics Masters Program, University of Oregon

John Cheney

Bioinformatics and Genomics Masters Program, University of Oregon

Brittany Curtiss

Knight Cancer Institute, Oregon Health and Science University

Sarah Carratt

Knight Cancer Institute, Oregon Health and Science University

Theodore Braun

Knight Cancer Institute, Oregon Health and Science University

Julia Maxson

Knight Cancer Institute, Oregon Health and Science University
Source: vignettes/CITEVizWorkshop.Rmd

About CITEViz

CITEViz is a single-cell visualization platform with a custom module that replicates the interactive flow-cytometry gating workflow. With CITE-Viz, users can investigate CITE-Seq specific quality control (QC) metrics, view multi-omic co-expression feature plots, and classify cell clusters by iteratively gating on the abundance of cell surface markers. CITE-Viz was developed to make multi-modal single-cell analysis accessible to a wide variety of biologists, with the aim to discover new insights into their data and to facilitate novel hypothesis generation. If you use our package, please cite our paper: CITE-Viz: Replicating the Interactive Flow Cytometry Workflow in CITE-Seq..

CITEViz Workshop Curriculum

  • Gating

    • 1-layer gates

    • 2-layer gates

    • Back-gate

    • Differential expression

  • Feature Co-Expression Plots

    • ADT-vs-ADT

    • ADT-vs-RNA

  • Quality Control

  • Cluster Visualization

Gating - Differential Expression

When two cell populations of interest are isolated, the gating data can be read back into Seurat to facilitate differential expression as follows:

library(Seurat)

# import original data
pbmc = readRDS("inst/extdata/small_pbmc_2.rds")

# import gate information
b_gate = readRDS("inst/extdata/B-Cells-2.rds")
nk_gate = readRDS("inst/extdata/NK-Cells-2.rds")

# extract cell barcodes in each gate
b_barcodes <- b_gate$gate_1@subset_cells[[1]]
nk_barcodes <- nk_gate$gate_1@subset_cells[[1]]
# make sure no overlapping cells between gates
b_barcodes <- setdiff(b_barcodes, nk_barcodes)

# differential expression
diff_exp <- FindMarkers(pbmc, ident.1 = b_barcodes, ident.2 = nk_barcodes)

View(diff_exp)
p_val avg_log2FC pct.1 pct.2 p_val_adj
MS4A1 3.07e-279 3.17 0.983 0.03 6.38e-275
CD79A 1.99e-278 2.96 0.978 0.02 4.14e-274
IGHD 1.29e-272 3.22 0.957 0.01 2.67e-268
IGHM 9.57e-266 2.50 0.985 0.08 1.98e-261
BANK1 1.21e-258 2.29 0.93 0.01 2.50e-254
RALGPS2 2.69e-244 2.34 0.942 0.07 5.59e-240
CD79B 1.50e-222 2.17 0.949 0.16 3.12e-218
TYROBP 8.83e-219 2.51 0.28 0.98 1.83e-214
LINC00926 1.21e-216 1.65 0.828 0.02 2.51e-212

Supplementary Information

CITEViz comes with accompanying slides located here.