Skip to main content
Figure 7 | BMC Bioinformatics

Figure 7

From: Stronger findings from mass spectral data through multi-peak modeling

Figure 7

Model 1 (All peaks) had a better accuracy at the prediction of signs of covariate effects for previously unseen lipids in the lipidomic gene-silencing data set compared to Model 3 (Single peak). The difference became pronounced when simulated noise was added to the data. The prediction was based on the inferred covariate effects of compounds from the same lipid family and was done in a cross-validation setting. In the task, the effects of the seven gene-silencing treatments were predicted on the three most abundant families of lipids in two time points. Points σ=0 and σ>0 on the x-axis show the prediction accuracy (y-axis) for the original data and the data with added noise, respectively.

Back to article page