Fig. 6: Proteome-based pseudotime captures the progression of dementia.
From: Inferring protein expression changes from mRNA in Alzheimer’s dementia using deep neural networks
a The trajectory of brain proteome and transcriptome in older adults. Each individual was mapped to 2-dimensional space using the spectral embedding algorithm given the top 40 PCs of the estimated proteome. Then, we used SCORPIUS to estimate the trajectory of the protein system across individuals and pseudotime for each individual. The stream plot on the bottom shows that the proportion of AD people is increased as the pseudotime proceeds. The density plots represent relationships of b global AD pathology, c cognitive decline, d hemisphere volume, and e hippocampus volume with pseudotime based on estimated proteome and actual transcriptome. Pearson’s correlation and p-value are indicated in each plot. The shaded areas represent the 95% confidence level interval for predictions from a linear model.
