Fig. 4

Distribution of compounds found by one or both of the two automatically curated databases. Percentage of compounds that are in SciFinder for the set of 45 patents and are only in SureChEMBL (orange), IBM SIIP (blue), or both (green). US4231938 is not included since none of compounds in SciFinder have been found in either SureChEMBL or IBM SIIP (cf. Table 1). The results used to generate this visualisation can be found in Additional file 3