Joshua Lederberg

In the early 1960s, the biologist Joshua Lederberg (1925-2008), a 1958 Nobel Prize laureate for his discoveries of genetic transfer in bacteria, started working with computers. Over the summer of 1962, he learned to program on BALGOL (Burroughs Algol) for the Burroughs 220 computer (a 1957 vacuum-tube computer with core memory) and quickly succumbed to the hacker syndrome. Lederberg soon became tremendously interested in creating interactive computers to help him in his exobiology research. Specifically, he was interested in designing computing systems to help him study alien organic compounds.

As he was not an expert in either chemistry or computer programming, Lederberg collaborated with two other prominent Jewish-American scientists from Stanford—chemistry professor Carl Djerassi (1923-2015) to help him with chemistry, and the chairman of Stanford computer science department Edward Feigenbaum (b. 1936) with programming, to automate the process of determining chemical structures from raw mass spectrometry data. Feigenbaum was an expert in programming languages and heuristics (in the late 1950s he developed EPAM, one of the first computer models of how people learn) and helped Lederberg design a system that replicated the way Djerassi solved structure elucidation problems. They devised a system called DENDRitic ALgorithm (Dendral) that was able to generate possible chemical structures corresponding to the mass spectrometry data as an output.

DENDRAL (see a historical note), the first expert system in the world, was written in the Lisp programming language of John McCarthy, which was considered the language of artificial intelligence (AI) because of its flexibility. DENDRAL ran on a computer system called ACME (Advanced Computer for Medical Research), installed at Stanford Medical School in 1965 for use by resident researchers through time-sharing.

The project consisted of research on two main programs Heuristic Dendral (see the description) and Meta-Dendral, and several sub-programs. Heuristic Dendral is a performance system and Meta-Dendral is a learning system. The initial program was coded by the programist Georgia Sutherland, but later the Dendral team recruited Bruce Buchanan to extend the system. Buchanan wanted Dendral to make discoveries on its own, not just help humans make them. Thus he, Lederberg, and Feigenbaum designed “Meta-Dendral”, which was a “hypothesis maker”.

The greatest significance of DENDRAL lies in its theoretical and scientific contribution to the development of knowledge-based computer systems. Many later expert systems were derived from Dendral, including SUMEX, MYCIN, MOLGEN, PROSPECTOR, XCON, and STEAMER.