Before you start
- Just curiosity. A little biology helps. (Want a refresher? Try The biology a coder needs.)
- Any term feels new? The Glossary explains everything in plain English.
Learning objectives
By the end of this lesson you will be able to: explain what RNA-seq measures and why gene expression is a snapshot of what a cell is doing, walk through the RNA-seq pipeline from samples to a count matrix to results, and describe what differential expression is asking.
The one idea to hold onto
Every cell in your body carries the same DNA, the same full instruction manual. But a brain cell and a muscle cell look and behave completely differently. Why? Because at any moment, each cell only "switches on" some of its genes, not all of them. A switched-on gene gets copied into RNA, and that RNA is the working instruction the cell actually uses.
So if you measure how much RNA there is for every gene, you get a snapshot of what a cell is actually doing, right now.
That measurement is exactly what RNA-seq gives you, for thousands of genes at once. And the moment you can measure it, you can compare it: healthy cells versus diseased, untreated versus drug-treated, before versus after. Whichever genes change between the two groups are your clue to what is going on. That comparison is the engine of modern molecular biology.
Decode the jargonGene expression
"Expression" just means how active a gene is, how much RNA the cell is making from it. A gene that is "highly expressed" is being used a lot; one that is "not expressed" is switched off. RNA-seq measures expression for every gene in one experiment.
The journey: from cells to an answer
An RNA-seq experiment is a pipeline. Here is the whole thing at a glance, and then we'll walk each step.
1. Samples and RNA (the wet lab)
It starts at the bench, not the computer. Researchers collect samples from the groups they want to compare, usually with several replicates each (more on why below), and chemically extract the RNA from the cells. This part is biology and lab work; as a bioinformatician you often receive the data after this step is done.
2. Sequencing (turning RNA into data)
A sequencing machine breaks the RNA into millions of tiny fragments and "reads" the sequence of each one. Each fragment becomes a short string of letters called a read. The raw output is a giant text file of these reads (the FASTQ format). This is the hand-off point from the lab to the computer.
3. From reads to a count matrix
On the computer, software figures out which gene each read came from (this is alignment) and then tallies them up: how many reads landed on gene 1, gene 2, and so on, for every sample. The result is the count matrix, a big table of genes by samples. A bigger number means the gene was more active in that sample.
Decode the jargonCount matrix
One row per gene, one column per sample, and each cell holds a whole number: how many reads were counted for that gene in that sample. This table is where the data analysis really begins, and it is exactly where our hands-on lessons start.
4. Finding the genes that changed
Now the statistics. For every gene, we ask: is the difference between the two groups bigger than the natural noise between replicates? Genes that pass become your differentially expressed hits. This is where tools like DESeq2 come in, and it is the heart of the analysis.
Decode the jargonReplicates, and why they matter
A replicate is a repeat of the same condition (for example, three separate treated samples instead of one). They matter because biology is noisy: without repeats you cannot tell a real change from a random fluke. More replicates means more confidence that a change is real.
5. Turning hits into biology
A list of changed genes is not yet an answer. The final step is interpretation: drawing the plots that reveal the patterns (volcano plots, heatmaps) and asking which biological functions and pathways the changed genes belong to (enrichment). That is how "200 genes changed" becomes "the immune response was activated".
Where you begin in these lessons
The wet lab and the read-to-count steps are their own deep topics. In this track, we start from the count matrix, which is where most real analysis begins, and take it all the way to biology. Here is the path:
- Your First RNA-seq Analysis, the complete project from start to finish.
- From Raw Counts to Differential Expression, going deeper on DESeq2.
- Reading Volcano and MA Plots, then Heatmaps and Clustering.
- Functional Enrichment, turning your gene list into biology.
Check your understanding
Sources & further reading
- Stark R, Grzelak M, Hadfield J. RNA sequencing: the teenage years. Nature Reviews Genetics 20:631–656, 2019. doi:10.1038/s41576-019-0150-2
- Conesa A, et al. A survey of best practices for RNA-seq data analysis. Genome Biology 17:13, 2016. doi:10.1186/s13059-016-0881-8
Last reviewed: June 2026.