From raw reads to biological inference

Reference guides for analysing RNA-seq, ATAC-seq, ChIP-seq, Hi-C and single-cell RNA-seq data. Each one documents one way of taking the data from raw reads to interpretation, with a ready-to-run Snakemake pipeline, the reasoning behind each choice, and the main alternatives.

NoteStatus

All R code in the guides has been run and tested. The RNA-seq and ChIP/ATAC pipelines have been run end to end on simulated data; none has yet been run on a full-size real dataset. See About for what was checked and how to report problems.

What these guides are

A record of the author’s workflow for each assay, written up so others can follow, compare and adapt it. They are a reference, not an authoritative standard: other valid approaches exist, and the right choices depend on your data. Where a common alternative exists, the guides name it. Check each tool’s own documentation before relying on a setting.

They are written for researchers who have (or are planning) a sequencing experiment. You don’t have to be a bioinformatician. If you can use a terminal and have met R before, you can follow them.

About the author

These guides were created by Debadeep Chaudhury. They are based on five years of experience as a bioinformatician during a PhD, and on hands-on research experience.

ORCID 0000-0002-9089-732X · Google Scholar · LinkedIn

What each guide covers

  • Experimental design: replicates, depth, controls and batches.
  • Turning raw FASTQ files into analysis-ready data with a reproducible pipeline that runs on a workstation or an HPC cluster.
  • Data quality at every step, using named metrics with thresholds from published standards.
  • A commonly used statistical analysis for the assay, and when an alternative may suit better.
  • The figures that test a conclusion, and how to read them.
  • Comparing across species (liftOver, consensus genomes, outgroup logic).

The guides

Start here

Getting started

Software, reference genomes, chromosome naming, reproducibility checklist, example data.

Bulk transcriptomics

RNA-seq

Gene expression: trimming, STAR, featureCounts, DESeq2, functional enrichment.

Chromatin

ATAC-seq

Accessibility: peak calling, differential accessibility, enhancers, motifs.

Chromatin

ChIP-seq

Protein–DNA binding and histone marks: peaks, QC, differential binding.

3D genome

Hi-C

Contact maps, normalisation, TADs, loops and their differences.

Single cell

scRNA-seq

Cell Ranger, QC, clustering, annotation, pseudobulk and abundance testing.

Across species

Comparative genomics

liftOver, orthologs, consensus genomes and outgroup logic, shared by all guides.

How every guide is organised

Step What it covers Question it addresses
1. Design Replicates, depth, controls, confounders Can this experiment answer my question?
2. Raw-read QC FastQC / MultiQC Is the sequencing usable?
3. Processing Run pipelines/<assay> What did each step do to my data?
4. Processed-data QC Assay-specific metrics Did the assay work?
5. Sample-level QC PCA, correlation, outliers Do replicates agree? Is there batch?
6. Statistics One commonly used test, plus alternatives What changed, and how confident am I?
7. Visualisation The plots that test the result Does the result look right?
8. Interpretation Annotation, enrichment What does it mean biologically?
9. Inference Pros/cons, pitfalls What can I claim, and what can’t I?
10. Cross-species extension liftOver, consensus genome How do I compare species?

Before you start

  1. Software: follow Getting started to install conda and Snakemake. The pipelines create their own tool environments.
  2. Hardware: a 16-core / 64 GB workstation is enough for most assays. HPC users can run the same pipelines through a Snakemake executor.
  3. Data: use your own, or the open example data below.

Example data

The R code in each guide runs on public example datasets that Bioconductor downloads for you. For trying the processing pipelines, an openly available comparative study of primate astrocytes (Ciuba et al. 2025, Cell Stem Cell, doi:10.1016/j.stem.2024.12.011) provides RNA-seq, ATAC-seq, ChIP-seq and Hi-C from human, chimpanzee and macaque cells (ArrayExpress E-MTAB-13252 to 13259). Its results appear in the guides as examples of what each analysis produces.

Acknowledgements

The guides were developed at:

Nencki Institute of Experimental Biology, Warsaw, Poland
Dioscuri Centre for Chromatin Biology and Epigenomics, Nencki Institute of Experimental Biology, Polish Academy of Sciences
3 Pasteur Street, 02-093 Warsaw, Poland

The author thanks Aleksandra Pękowska, Katarzyna Ciuba, Aleksandra Piotrowska and Eryk Duński for their collaboration and support.

The page structure is modelled on the excellent, openly licensed training materials of the Harvard Chan Bioinformatics Core (HBC).

Back to top