Single-Cell RNA Sequencing: From Droplets to a Count Matrix
A single-cell protocol captures one cell per droplet, tags its transcripts with a cell barcode and a molecular identifier, and returns a sparse matrix of counts. This stop follows the sample through the bench steps that decide what that matrix can support.
What a learner can do afterwards
- Trace a transcript from lysis through barcoding, amplification and sequencing to one entry in the count matrix
- Explain what unique molecular identifiers correct for and what they leave uncorrected
- Name the artefacts the protocol introduces: doublets, empty droplets, ambient RNA and dropout
1 · Read
A single-cell protocol captures one cell per droplet, then breaks it open and catches its messenger RNA. Each transcript gets two tags before amplification and sequencing: a cell barcode saying which droplet it came from, and a molecular identifier marking the original molecule. Reads sharing both tags collapse into one molecule count. All counts assemble into a matrix with genes as rows and cells as columns, and one entry counts transcripts of one gene in one cell.
Suppose one original molecule is copied a hundred times by amplification. Without identifiers you would count a hundred molecules. With identifiers all hundred reads carry the same barcode and tag, so they collapse back to one. That is what molecular identifiers correct for: amplification bias. They cannot fix capture loss, since molecules never caught leave no tag at all.
The bench steps decide what the matrix can support, because each step adds its own artefacts. Doublets merge two cells into one droplet and mimic a fake intermediate type. Empty droplets carry only background. Ambient RNA from burst cells leaks into every droplet and blurs differences. Dropout leaves true expression uncounted, so a zero may mean absence or mere failure to capture.
To read a matrix honestly, flag artefacts before any clustering. Distrust a brand new cell type built from doublet-like mixed profiles. Treat zeros as guilty until proven innocent: ask whether capture failure explains them. Good analysis starts by asking what each zero and each odd profile owes to the protocol.
Tag, collapse, count, then doubt: barcodes map molecules to cells and artefacts decide what zeros mean.
2 · Watch
Take it off screen
Where it sits
8 questions wait behind this lesson, each with its answer explained. Every answer feeds the sky: stars light as they are learned, and dim when it is time to come back.