Where does our data come from?

The data on System is currently sourced from the entire PubMed corpus: the leading health and biomedical database used by millions of researchers every day.

So far, we’ve extracted over 14 million findings from more than 5 million studies, including all English-language abstracts on PubMed and open-access full-text articles. We add new findings daily as the corpus grows. Importantly, we extract relationships at the source, from the original research studies, along with key metadata like covariates and study population, before structuring those findings and storing in our graph database.