omics.co.in logo omics.co.in

Chapter 17: scRNA-Seq Differential Expression, Marker Prioritization & Label Assignment

1 min read • Verified October 2026 • Peer-Reviewed Learning Resource
Tested Stacks: GATK 4.6.1 Samtools 1.21 Nextflow 24.10 Seurat v5.1 Python 3.12
Prerequisites: Basic Linux CLI & Sequencing Formats
📚 Chapter 17 of 18 0% Course Completed

1. Marker Discovery in Single-Cell Transcriptomics

Once single cells are clustered into discrete UMAP communities, researchers must identify the key genetic markers distinguishing each cluster from all other cells. Because single-cell count matrices contain excessive zero counts (structural zeros and technical dropouts), statistical marker discovery requires specialized tests:

  • Wilcoxon Rank-Sum Test: Default non-parametric test. Extremely fast, robust, and highly recommended for general cluster marker prioritization.
  • MAST (Model-based Analysis of Single-cell Transcriptomics): Hurdle GLM model adjusting for cellular detection rate (proportion of genes detected per cell).
  • LR (Logistic Regression): Models cell cluster membership as a function of gene expression.

2. Multi-Feature Visualization Strategies

Visualization MethodInformation EncodedPrimary Research Value
DotPlotDot size = % of cells expressing marker; Dot color = Mean expression intensityCompact multi-cluster marker panel validation in publications
FeaturePlotColor gradient projected directly onto 2D UMAP coordinatesSpatial confirmation of marker restriction to distinct UMAP islands
VlnPlot (Violin Plot)Kernel density estimate of expression distribution across clustersAssessing expression homogeneity and detecting bimodal distributions

3. Comprehensive Single-Cell Marker Prioritization Script

library(Seurat)
library(dplyr)
library(ggplot2)

# 1. Identify Marker Genes for Every Cluster Against All Other Cells
# Pre-filtering: minimum 25% cells expressing gene in either group, logfc >= 0.25
pbmc.markers <- FindAllMarkers(
    pbmc,
    only.pos = TRUE,
    min.pct = 0.25,
    logfc.threshold = 0.25,
    test.use = "wilcox"
)

# 2. Extract Top 3 Diagnostic Markers Per Cluster
top3_markers <- pbmc.markers %>%
    group_by(cluster) %>%
    slice_max(n = 3, order_by = avg_log2FC)

cat("=== Top 3 Marker Genes Per Single-Cell Cluster ===\n")
print(top3_markers[, c("cluster", "gene", "avg_log2FC", "p_val_adj")])

# 3. Multi-Cluster Diagnostic DotPlot
canonical_markers <- c("CD3D", "CD4", "CD8A", "MS4A1", "CD14", "FCGR3A", "NKG7", "PPBP")
DotPlot(pbmc, features = canonical_markers) +
    RotatedAxis() +
    scale_color_gradient2(low = "#2563eb", mid = "#ffffff", high = "#dc2626") +
    labs(title = "Expression of Canonical Lineage Markers Across Single-Cell Clusters")

# 4. FeaturePlot Projection of Key Biomarkers
FeaturePlot(pbmc, features = c("CD3D", "MS4A1", "CD14", "NKG7"), ncol = 2)
⚠️

Common Pitfalls & Troubleshooting Checklist

  • Java Heap Exhaustion (GATK / Picard): If processes terminate with java.lang.OutOfMemoryError, set your environment variable explicitly: export _JAVA_OPTIONS="-Xmx16g -Djava.io.tmpdir=/tmp".
  • Index Timestamp Mismatch: If tools complain that a BAM index is missing or out of date, regenerate with samtools index -@ 4 sample.bam.
  • Conda Channel Conflicts: Avoid ABI incompatibilities between bioconda and conda-forge by setting strict priority: conda config --set channel_priority strict.
  • Chromosome Naming Contamination: Always verify whether reference FASTA headers use chr1 or 1; divergent naming causes silent alignment drops.
Was this guide helpful to your research?
Help us maintain 100% accurate, peer-reviewed bioinformatics standards.
📝 Propose Edit on GitHub ↗