Fig. 1
From: Identifying therapeutic targets by combining transcriptional data with ordinal clinical measurements
Combining transcriptomic data with ordinal clinical information. a Schematic representation of postmortem brain tissues, which are categorized in four groups based on their degree of neurodegeneration. The formula shows an ordinal regression model in which the expression of a gene explains the degree of neuronal loss. This model is implemented using the proportional odds assumption, where the slope of the fitted lines (β) is considered equal for all categories. b In this plot, the X axis shows the expression of a gene, while the Y axis shows neurodegenerative grades. Each dot represents the expression of a gene from a sample with neurodegenerative grade y. c The plot shows the ordinal regression model, in which a linear function of the expression of a gene is related to the probability of an ordinal category. d In this plot, the expression of a gene is represented in the X axis, while the Y axis shows a function of the cumulative probability of ordinal categories. The parallel lines display fitted lines by implementing the proportional odds assumption
