Chemical Correlation DataBase (CCDB)

Core chemical datasets in Exposomics are :

1. Targeted biomonitoring : ( NHANES, HHEAR, targeted panels in individual publications )

2. Untargeted data with high-confidence annotations : data reported in Metabolomics WorkBench, EBI Metabolights, NIH consortiums data centers,  or supplementary tables in publications

3. Untargeted data without any annotations : peak-intensity matrices generated by untargeted data processing method such as IDSL.IPA

As part of the data curation pipeline, we first compute the pair-wise correlations among the chemicals detected in these datasets.  All of them are indexed into an opensearch full text database of inter-chemical correlations if they pass a set of our curation criteria. We are using the InChikey to represent a known chemical in our database, whereas un-identified peaks are assigned an unique identifier (STUDYID-ANALYSISID-MZ-RT).  The list of curated chemical identifiers by our team are provided in Zenodo.org.