Loading animations
Some things are far easier to watch than to read about. These explainers cover the concepts students get stuck on — how sequencing actually works, why we run FastQC, how a read finds its place in the genome, and how a neural network learns.
Molecular biology
Follow a single read through every file format an NGS pipeline touches, watching exactly what each format adds. Click any field to find out what it means.
Molecular biology
Nothing in DNA says where a codon starts. Read the same 123 bases six ways, then take one base out and watch the protein become unrelated — and put three back to see it recover.
Molecular biology
Mutate any base of the real beta-globin gene and see what it does. Enumerating all 1,332 possible substitutions shows why “it is a missense variant” says almost nothing on its own.
Molecular biology
Follow one gene to one protein: 300 bases of DNA become a 168-nucleotide message and 37 amino acids. The finished mRNA cannot be found anywhere in the gene.
Molecular biology
Thirty cycles is not thirty times more DNA — it is about a billion. Then watch the reaction run out of reagents, and four samples starting a million-fold apart finish at the same place.
Molecular biology
A letter’s height is its frequency times the information in its column, in bits — not its frequency. And a logo built from a handful of sites is inflated by a bias you can calculate.
Molecular biology
Five cuts in a circle give five fragments, not six. Then run them out and watch two land in the same band — on an axis that is logarithmic, not linear.
Molecular biology
There are 4²⁰ possible 20-mers and only three billion bases, so a guide must be unique. It is — until you allow three mismatches, and the target becomes thirty thousand sequences instead of one.
Molecular biology
Twelve beads on a grid have 120,292 shapes, and every one of them gets checked. A hydrophobic core falls out of the arithmetic — and most sequences turn out to have no fold at all.
Molecular biology
Take the insulin B chain and rewrite every codon for E. coli. The protein comes out character for character identical, two thirds of the DNA does not, and the adaptation score that measures the difference turns out to be easy to game.