Fig. 4: Clustering of top 200 candidate antigens based on proximity measured from tree-based model. | npj Systems Biology and Applications

Fig. 4: Clustering of top 200 candidate antigens based on proximity measured from tree-based model.

From: Positive-unlabeled learning identifies vaccine candidate antigens in the malaria parasite Plasmodium falciparum

Fig. 4

First two dimensions of UMAP are shown. Top 200 candidate antigens from the final tree-filtered model were grouped based on k-means clustering. Points represent top 200 candidate antigens in three groups, 48 known antigens (light cyan), and four reference antigens (yellow; protein names noted by text).

Back to article page