Skip to main content

Table 1 Omics data description and technical variability management

From: Variability of multi-omics profiles in a population-based child cohort

Omics profile

Matrix

Sample size (omics available for both visits)

Number of features

Laboratory processing

Batch correction

Criteria for feature exclusion

DNA methylation

Blood leukocytes

149

91601

Randomized by cohort and sex, and panel samples paired in plate and array

Residuals of SVs protecting for cohort, sex and age. Cell type composition also corrected with SVs.

< 98% call rate and < 62.5% ICC

Gene expression

Whole blood

105

45438

Randomized by cohort and sex, and panel samples paired in plate.

Residuals of SVs protecting for cohort, sex and age. Cell type composition also corrected with SVs.

< 25% call rate

miRNA expression

Whole blood

100

453

Randomized by cohort and sex, and panel samples paired in plate and array.

Residuals of SVs protecting for cohort, sex and age. Cell type composition also corrected with SVs.

< 25% call rate

Proteins

Plasma

149

36

Randomized by cohort

Overall protein average minus plate specific protein average subtracted for each individual and each protein

< 30% measurements in the linear range (LIN)

Serum metabolites

Serum

154

177

Fully randomized

-

> 30% CV and > 30% BLD + zeros

Urine metabolites

Urine

154

44

Fully randomized

-

> 30% CV

  1. Definitions. Call rate (for DNA methylome): proportion of detection of a given CpG among samples; Call rate (for miRNA and gene expression): proportion of detection of gene or miRNA among samples. Abbreviations. SV surrogate variables, ICC interclass correlation coefficient [35], CV coefficient of variation, BLD below limit of detection