search

dayoonkwon / BioAge

Biological Age Calculations Using Several Biomarker Algorithms
GNU General Public License v3.0
72 stars 24 forks source link

missing data in biomarkers #33

Closed lixinmei1235 closed 5 months ago

lixinmei1235 commented 5 months ago

Dear Authors, When using the kdm_calc function in the BioAge package, I found that when the number of missing biomarkers of an observation is <=2, the corresponding kdm value can be calculated. On what statistical treatment or references is this based? And why not ignore observations that contain any missing values of these biomarkers? Sincerely waiting for your answer, thank you！

The following code was taken from ‘kdm_calc.R’ dat$kdm = ifelse(BA_nmiss>2,NA,dat$kdm)

dayoonkwon commented 5 months ago

The decision to allow calculations of the kdm value when the number of missing biomarkers is two or fewer is based on statistical considerations aimed at maximizing data use while maintaining reliability.

If you want a complete set of biomarkers without any missing values, I recommend preprocessing your data to include only complete cases before applying the kdm_calc function.