Computational protocol: Biomarkers of inflammation in infants with cystic fibrosis

Similar protocols

Protocol publication

[…] Descriptive statistics include the mean and standard deviation or the median and range, where specified. Linear mixed effects models were used to model the logarithm of targeted biomarker levels that were longitudinally assessed with subject-specific, additive random effects. Analysis of urine samples controlled for creatinine and specific gravity by taking the log and including them as covariates. Sputum and BALF biomarker concentrations were not normalized prior to analysis. To test for an association between urinary cathepsin B levels and CF status, a model with the log of cathepsin B as the response variable and the log of specific gravity, the log of creatinine, age, CF status and the interaction between age and the indicator for CF status was fit. The test for an association between BALF CCSP and plasma CCSP was based on a model which had BALF CCSP as a response variable and included no other covariates (although the results were not sensitive to inclusion of gender and age). A similar analysis detected the association between CCSP and IL-8. Two sample t-tests were used to test for differences in biomarker concentration between infants and older subjects with CF []. Receiver operator characteristic (ROC) curves were generated to examine the sensitivity and specificity of the biomarkers. All calculations were conducted using R version 2.15.2 and mixed effects models were fit using the lme function in the nlme package []. […]

Pipeline specifications

Software tools lme4, nlme
Application Mathematical modeling
Organisms Homo sapiens
Diseases Cystic Fibrosis, Lung Diseases
Chemicals Desmosine