In silico — carrying out some experiment by means of a computer simulation — is closely based on two older Latin phrases that are key terms in the jargon of every biologist and biochemist: in vivo and in vitro, both of which came into use at the end of the nineteenth century. The first translates as “in that which is alive”, and refers to some experiment carried out within a living organism, such as a drug test on an animal. The second means “in glass” and is used for experiments that take place in an artificial environment, such as a test tube or culture dish.
When it became possible to run computer-based simulations or mathematical models relating to animal biochemistry, behaviour and environment, in silico, in silicon, was coined for a trial carried out, as it were, on a computer chip. The online encyclopedia Wikipedia asserts that it was first used in 1989 by the mathematician Pedro Miramontes. Purists dislike it because the correct Latin form would be in silicio but in silico is now too firmly established to be seriously challenged.
Here’s an example from Kevin Kelly’s Out of Control: the New Biology of Machines: “To complement their studies in vitro, Pimm also set up experiments in silico — simplified ecological models in a computer”.
It’s now common in the specialist scientific vocabulary:
The computers were programmed to mathematically represent many of the processes in the body associated with type 2 diabetes, the condition this new drug aimed to treat. And, like human patients, the virtual patients were each slightly different from one another — they were given various weights, for instance, or had slightly different liver functions. By trying out a variety of doses “in silico”, from below expected efficacy to several-fold beyond their target dose, they learned that hypoglycaemia should not prove a problem.
New Scientist, 06 Aug. 2005.