Fig. 2
From: Longitudinal multi-omics transition associated with fatality in critically ill COVID-19 patients
Transcriptome features of phase transition between Cluster 1 and Cluster 2. a Left, PCA map for transcriptomics of 113 samples; right, the percentage of terminal status in each cluster. Dots represent samples. Three colors represent three clusters. Round, triangle, and square shapes represent patients’ terminal status. b Longitudinal distribution of samples in colors of clusters. 113 transcriptomic data were represented at time-points. Scatter plot shows the RNA sample collection time with dots colored by clusters. Red box-shadow represents Cluster 1-to-2 transition. Blue box-shadow represents Cluster 2-to-1 transition. c Heatmap of differential expression genes (logFC > 1) between Cluster 1 and Cluster 2. d Soft clustering of longitudinal gene dynamics in six patients containing Cluster 1-to-2 transition (FDR < 0.05). Each patient is individually analyzed by using mfuzz. X-axis represents the sampling time. e Validation of differential expressed gene from Cluster 1-to-2 by using Cluster 2-to-1 transcriptome data. Color represents the patient id. 85.2% up-regulated genes and 99.2% down-regulated genes show reversed expression levels. f GO terms and KEGG pathways for differentially expressed genes between Cluster 1 and Cluster 2. Top 3 terms and pathways are displayed. The size of dots denotes the -log10 of the p-value, the color denotes the expression level