Single Cell Omics - iffatAGheyas/bioinformatics-tutorial-wiki GitHub Wiki

6.1 Single-Cell Omics

Single-cell technologies let us profile gene expression (and other modalities) at cellular resolution. In this module we’ll cover a typical scRNA-seq workflow plus key analysis steps and tools.

What We’ll Cover

Data Types & Formats

• 10x Genomics Chromium (Cell Ranger matrices)
• Smart-seq2 (full-length)
• UMI vs. non-UMI protocols

Preprocessing & QC

• Demultiplexing & alignment (Cell Ranger, STARsolo)
• Generating gene × cell count matrices

Quality Filtering

• Remove low-quality cells (min genes, UMI counts)
• Filter out genes expressed in few cells
• Mitochondrial / ribosomal content checks

Normalization & Feature Selection

• Library-size normalization (CPM, log1p)
• SCTransform (Seurat) or scran’s deconvolution
• Identify highly variable genes (HVGs)

Dimensionality Reduction & Clustering

• PCA on HVGs
• Neighborhood graph (k-NN)
• UMAP / t-SNE visualization
• Leiden / Louvain clustering

Marker Detection & Differential Expression

• Find markers per cluster
• Wilcoxon, MAST tests

Trajectory & Pseudotime Analysis

• Monocle 3, Slingshot, scVelo (RNA velocity)

Multi-Omics & Batch Integration

• CITE-seq (protein + RNA)
• Batch correction (Harmony, Seurat IntegrateData)
• Multi-ome integration (MOFA2, Seurat WNN)

Visualization

• Feature plots, dot plots, heatmaps
• Trajectory / velocity streamlines

Simulation & Benchmarking (Optional)

• Splatter (R)
• dyngen (Python) (Optional)
• Splatter (R)
• dyngen (Python)