model organism

You cannot do most experiments on humans, and even if you could, we grow slowly and are hard to control. So molecular biologists do the bulk of their work on a small cast of well-chosen stand-ins: simpler, faster, cheaper organisms that are easy to study in the lab and whose lessons carry over to us. These are the model organisms, and a huge fraction of what we know about how life works at the molecular level was learned from them first.

A good model is chosen for convenience that does not sacrifice relevance. It should grow fast and cheaply, be easy to manipulate genetically, and have biology similar enough to ours that the findings transfer. The classic roster is small: the gut bacterium Escherichia coli; baker's yeast, a simple eukaryote; the roundworm Caenorhabditis elegans, whose every cell has been traced; the fruit fly Drosophila melanogaster, a century-old workhorse of genetics; the mouse, our closest standard mammalian model; and the small weed Arabidopsis thaliana for plants. Each occupies a sweet spot of being simple enough to dissect yet relevant enough to teach us something general.

Models work because the deep machinery of life is shared. The genetic code, the way DNA is copied and read, and many core proteins are nearly the same from bacteria to humans, so a gene studied in yeast often has a close human counterpart doing a similar job. The catch is that models are simplifications: a discovery in a worm is a strong lead, not a guarantee, and confirming that a result holds in humans is its own essential step.