PMC:2846953 / 24352-25062 JSONTXT

{"target":"http://pubannotation.org/docs/sourcedb/PMC/sourceid/2846953","sourcedb":"PMC","sourceid":"2846953","source_url":"https://www.ncbi.nlm.nih.gov/pmc/2846953","text":"The primary purpose of this analysis was to explore the quantitative associations of biomarkers of exposure and other variables with biomarkers of potential harm related to cigarette smoking. We examined the covariates (or secondary variables) to see whether they could mediate or modulate the effect of biomarkers of exposure. In pursuing this, we developed a systematic approach to the mining of a complex biomarker database. This approach helped us find interpretable linear models that could explain a substantial proportion of the variability for all four of the biomarkers of potential harm (WBC, EPI8, DEH11, HDL). These regression models summarize the information originally contained in 169 variables.","tracks":[]}