From seeing cells to reading their code
The earlier guides in this rung handed you eyes. Microscopes, fluorescent tags, and the ability to grow cells in a dish let us *watch* life — but watching only goes so far. Two cells can look identical under the best microscope and yet carry completely different DNA, run different genes, or make different amounts of a single protein. To answer questions like *which* gene, *what* sequence, *how much* protein, you have to stop looking and start reading the molecules themselves. That shift — from seeing to reading — is what built modern molecular biology, and it rests on a surprisingly small set of techniques.
Three of the four tools in this guide rest on one idea you already met deep in the genome rungs: base-pairing. Because A always pairs with T and G with C, a single strand of DNA carries enough information to find, copy, or read its partner. Copying DNA (PCR), reading DNA (sequencing), and even the base-pairing address tag of the gene editors in the next guide all exploit that same rule. The fourth tool, the Western blot, steps sideways to ask about proteins instead. Together they let us copy, sort, read, and detect — the four verbs of the molecular trade.
PCR: a molecular photocopier
The first problem is brute scarcity. A drop of blood, a hair root, a swab from your cheek — each holds only a vanishingly tiny amount of any one stretch of DNA, far too little to read or test. PCR, the polymerase chain reaction, solves this by copying one chosen region over and over until a single starting molecule becomes hundreds of millions. Think of it as a photocopier that, instead of copying a whole book, copies only the one paragraph you point it at. PCR is the technique that makes almost everything else in this guide possible, because you usually need *abundant* DNA before you can sort it or read it.
Here is the lovely part: PCR is just DNA replication, taken out of the cell and run in a tube. Recall from the replication rung that copying needs the strands pulled apart, short primers to mark a starting point, and DNA polymerase to extend them. PCR does all three with nothing but temperature changes. Heat near boiling pries the double helix into single strands. Cooling lets two short, custom-designed primers — one for each end of your target — stick exactly where you want copying to begin and end. A warming step then lets the polymerase fill in fresh strands between the primers. One subtlety made it practical: ordinary polymerase is destroyed by the near-boiling step, so PCR uses a heat-resistant polymerase borrowed from microbes that live in hot springs.
- Denature: heat the sample near boiling so each double helix splits into two single strands.
- Anneal: cool down so the two short primers latch onto the exact ends of your target sequence.
- Extend: warm slightly so the heat-resistant polymerase builds a new strand from each primer.
- Repeat: each cycle doubles the copies, so ~30 cycles turn one molecule into hundreds of millions.
Gel electrophoresis: sorting by size
Now you have a tube full of DNA — but how do you know your PCR actually copied the right thing, and not some shorter junk? You sort the fragments by size. Gel electrophoresis pours a slab of jelly-like gel riddled with microscopic pores, loads your DNA into little wells at one end, and runs an electric current across it. Here the chemistry of DNA does the work for you: its backbone carries a uniform negative charge, so every fragment is dragged toward the positive electrode. Small fragments wriggle through the mesh quickly and travel far; large ones snag and lag behind. Gel electrophoresis turns an invisible mixture into a ladder of bands, each band a pile-up of fragments that are all the same size.
wells (load here) negative electrode (-) | | | | [===][===][===][===] <- DNA starts here, runs DOWN : = : = big fragments: slow, near top = : = : : = : small fragments: fast, near bottom = = v v v v positive electrode (+) lane1 marker sampleA sampleB (compare to known sizes)
Be honest about what a gel can and cannot tell you. By running a lane of known size markers beside your sample, you can read off roughly how large each band is — so a gel confirms *that a molecule of the right size is present*. But that is all. The gel cannot tell you the band's actual sequence, what gene it is, or what it does; two completely unrelated fragments of the same length sit in exactly the same place. Gel electrophoresis is a tool of separation and sizing, not identification — which is precisely why it so often serves as the *first step* feeding into sequencing or a Western blot, where the real identification happens.
Sequencing: reading the letters
Copying and sizing DNA still leaves the deepest question unanswered: what does it actually *say*? DNA stores meaning in the exact order of its four letters — A, T, G, C — and a smear of DNA, however abundant, tells you nothing until you read that order. DNA sequencing is the technology that reads it out letter by letter, turning a physical molecule into text you can store, search, and compare against any other genome on Earth. This is the step that converts biology into information.
Most modern sequencing leans, once again, on copying — but copying watched in slow motion. The machine builds a new strand opposite the DNA being read, and arranges for each letter to announce itself as it is added: in one widely used method, every incorporated base flashes a tiny burst of light in a colour that says which of the four it was. A camera records the colours in order, spelling out the sequence. The trick that made DNA sequencing cheap was doing this not for one molecule but for millions in parallel, so a human genome of three billion letters can now be read in a day or two — a task that took the first Human Genome Project over a decade and billions of dollars.
Yet a sequence is not understanding. Sequencing tells you what the recipe *says* — the order of letters — but not on its own which genes are switched on in a given cell, how the cell actually uses them, or what an unfamiliar stretch does. Knowing a mutation is present does not by itself tell you whether it causes disease; that still demands experiments, the kind you ran in the culture dish two guides ago, often in a model organism. And the reads themselves carry occasional errors, so any finding that matters is confirmed by reading the same region many times over. Sequencing reveals the text; the meaning of the text is a separate, harder problem.
The Western blot: catching a protein
DNA tells you the recipe; proteins are the dish that gets cooked. A cell can carry a gene yet make none of its protein, or pour out enormous amounts of it — and only the protein actually does the work. The Western blot answers two plain questions about one chosen protein: is it there, and roughly how much? It begins exactly where the gel left off: the cell's proteins are separated by size using gel electrophoresis (proteins, like DNA, can be coaxed to march through a gel by size), then transferred — *blotted* — onto a thin membrane that pins them in place for probing.
Then comes the clever part, which you have already met in the fluorescence guide: an antibody is used as a custom-shaped probe. An antibody raised against your target protein is washed over the membrane and sticks only to that one protein, ignoring the thousands of others. A second, labelled antibody then latches onto the first and produces a visible signal — a dark band or a glow — exactly where the target sits. The band's *position* up the gel reveals the protein's size; its *darkness* reflects roughly how much is present. This is the same antibody-as-probe logic as immunostaining, but read out as a band on a membrane rather than a glow inside a cell.
Four verbs, one workflow
These tools shine brightest in combination, each handing off to the next. A typical day might run like this: extract a scrap of DNA from a sample and use PCR to *copy* a target gene into abundance; run a gel to *sort* the product and confirm it is the right size; send that band off to *read* its sequence; and, if you care whether the gene is actually being made into protein, run a Western blot to *detect* and roughly measure that protein. Copy, sort, read, detect — four verbs that chain together into the everyday workflow of a molecular lab.
Notice the common thread running under all of it. PCR copies DNA, sequencing reads DNA, and the gene editor waiting in the final guide aims itself at DNA — and all three work only because A pairs with T and G with C, the same base-pairing rule that holds the double helix together. We have now learned to *see* cells, to *grow* and *sort* them, and to *read and copy* their code. Just one verb remains, the boldest of all: to *rewrite* it. That is where the next guide ends the ladder — with CRISPR, the tool that turned reading the code into editing it.