To determine the contribution of a range of different biological, genetic and technical factors to the proportion of variance explained in plasma protein levels, we’ve run linear mixed models including all relevant variables. The plot shows the proportion of explained variation of plasma levels of the protein of interest attributed to each of the biological, genetic and technical factors, from a joint model. Explanatory factors are grouped into higher-level classes, shown in the plot legend and represented by the different colours.
BMI: Body mass index
HDL: High density lipoproteins
LDL: Low density lipoproteins
eGFR: estimate glomerular filtration rate
HbA1c: Glycated haemoglobin
ALP: Alkaline Phosphatase
ALT: Alanine aminotransferase
T3: Triiodothyronine
T4: Thyroxine
GRS: Genetic risk score
Proteomic PC1 (0.005%): Principal component 1 for 0.005% dilution proteins
Proteomic PC1 (0.5%): Principal component 1 for 0.005% dilution proteins
Proteomic PC1 (20%): Principal component 1 for 0.005% dilution proteins
To determine the contribution of a range of different biological, genetic and technical factors to the proportion of variance explained in plasma protein levels, we’ve run linear mixed models including all relevant variables. The plot enables comparing the proportion of explained variation in plasma levels of two proteins attributed to each of the biological, genetic and technical factors. Explanatory factors are grouped into higher-level classes, shown in the plot legend and represented by the different colours.
The plot enables visualizing the top proteins for which a variable of interest explains the largest proportion of plasma variation. The colour of the tiles at the bottom of the plot represents the variable class which explains most of the variation in plasma of the protein. This shows whether the variable of interest is the major contributor the total variation of plasma levels of the proteins, or whether other variables explain a larger proportion of variation in plasma. The bar represents the proportion of variance explained by the variable looked-up and is coloured accordingly.