Fig. 1 | Scientific Reports

Fig. 1

From: Unsupervised learning reveals novel disease-associated proteins in high-dimensional human proteomic data

Fig. 1

Schematic of the DIRAM/COD framework for the creation of the clusters. The blue path on the left represents the DIRAM method, where the dataset was sliced to avoid missing values. The yellow path on the right represents the DIRCOD method, where the Leiden community algorithm was used. The central medallion illustrates our definition of high- and low-abundance proteins.

Back to article page