Data mining techniques are becoming increasingly important in chemistry as
databases become too large to examine manually. Data mining methods from th
e field of Inductive Logic Programming (ILP) have potential advantages for
structural chemical data. In this paper we present Warmr, the first ILP dat
a mining algorithm to be applied to chemoinformatic data. We illustrate the
value of Warmr by applying it to a well studied database of chemical compo
unds tested for carcinogenicity in rodents. Data mining was used to find al
l frequent substructures in the database, and knowledge of these frequent s
ubstructures is shown to add value to the database. One use of the frequent
substructures was to convert them into probabilistic prediction rules rela
ting compound description to carcinogenesis. These rules were found to be a
ccurate on test data, and to give some insight into the relationship betwee
n structure and activity in carcinogenesis. The substructures were also use
d to prove that there existed no accurate rule, based purely on atom-bond s
ubstructure with less than seven conditions, that could predict carcinogeni
city. This results put a lower bound on the complexity of the relationship
between chemical structure and carcinogenicity. Only by using a data mining
algorithm, and by doing a complete search, is it possible to prove such a
result. Finally the frequent substructures were shown to add value by incre
asing the accuracy of statistical and machine learning programs that were t
rained to predict chemical carcinogenicity. We conclude that Warmr, and ILP
data mining methods generally, are an important new tool for analysing che
mical databases.