Chemicals Informatics: Discover Structural Factors and Optimize Combinations of Compounds for Water Repellent Composite based on Public Literatures
Takashi Isobe, Yoshihiro Okada · 2021
Chemical industry pays much cost and long time to develop new compounds or composites that have aimed properties. The developers need to efficiently discover initial candidates before simulation, actual synthesis, optimization, and evaluation. To meet their needs, we have developed CI (Chemicals Informatics) to efficiently discover potential candidates of compounds or composites based on large number of public literatures. Our system currently has the data of 117M existing compounds, and 61 properties extracted by analyzing a public chemical database with the linked worldwide 33M papers and 30M patents in addition to 11M new structures generated based on existing compounds. The compounds are shown as vectors including 41 organic and 70 inorganic features. The properties are extracted from literatures using rule-based NLP (Natural Language Processing). Our system also made evolve to predicts 61 properties by each space of compound crossover for composite in addition to neighbor and structural crossover for compound written in past paper. This paper further shows users can extract structural factors and highly probable combinations of compounds that contribute to good properties. Moreover, they can find potential combinations without patents of aimed property from 128M to the fourth power. As the actual novel use case, we explored structural factors and highly probable new combinations for water repellent composites. CI could extract structural factors that contributed to good properties and highly probable combinations of compounds for composite that contributed to contact angle of 160 degree.