# Week 4 Resources: Bulk RNA-seq And Differential Expression

These files support Week 4: "Bulk RNA-seq Field Guide: From Reads To Rigorous Differential Expression."

## Files

| File | Purpose |
|---|---|
| `nfcore_samplesheet.csv` | Minimal nf-core/rnaseq-style samplesheet example |
| `metadata_design_examples.tsv` | Toy metadata for simple, batch-adjusted, paired, and interaction designs |
| `deseq2_design_cheatsheet.md` | Saveable DESeq2 design formula reference |
| `deseq2_design_matrix_demo.R` | Runnable R script showing what common design formulas create |
| `expression_units_demo.R` | Runnable R script showing raw counts, FPKM, TPM, normalized counts, log counts, and z-scores |

## Quick Start

```bash
cd content/resources/week-04
Rscript deseq2_design_matrix_demo.R
Rscript expression_units_demo.R
```

These scripts do not require DESeq2. They use base R so you can inspect design formulas and expression units before running a differential expression analysis.

## Key Lesson

The RNA-seq processing steps are broadly similar across many count-based assays:

```text
raw reads -> QC -> trimming/filtering -> alignment or quantification -> feature counts -> metadata-aware statistical model
```

The statistics are not interchangeable. The model must match the experimental design.