Predicting candidate therapeutic drugs for sepsis-induced acute respiratory distress syndrome based on transcriptome profiling

Figures & data

Table 1. Top ranking list of up-related and down-related genes sepsis-induced ARDS

Gene	logFC	pValue	FDR
PURG	3.582800226	4.51E-17	8.82E-13
ADAMTS6	3.740993555	2.45E-16	2.24E-12
WDR66	3.716725258	3.43E-16	2.24E-12
B3GNT7	3.820669571	4.92E-16	2.41E-12
KRTAP20-3	3.236491687	6.48E-16	2.53E-12
CPS1	−4.742218398	3.32E-11	1.10E-08
TM4SF1	−4.994941596	1.60E-10	3.68E-08
GAGE2B	−3.854115188	4.68E-10	8.72E-08
KRT17	−3.462657465	6.84E-10	1.23E-07
NTS	−3.585856923	1.03E-09	1.62E-07

Figure 1. Differential gene expression analysis of sepsis-induced ARDS and control samples. (a) Bidirectional hierarchical clustering of sepsis-induced ARDS and control samples. Each row represented a gene, each column represented a sample. Sepsis-induced ARDS samples are labeled as red and the control samples are labeled as blue. The heatmap shows the Z-score of relative gene expression of each samples; (b) Volcano plot for DEGs between sepsis-induced ARDS and control samples. X-axes indicates -log (FDR) and y-axes showes the log2 fold change. Red dots represent significantly up-regulated genes and green ones represents the down-regulated genes

Figure 2. Functional enrichment analysis of the DEGs between sepsis-induced ARDS and control samples. DAVID bioinformatic tools were used for functional enrichment analysis of the DEGs between sepsis-induced ARDS and control samples. Gene number, enrichment score (rich factor) and the enrichment p value of each biological process were displayed in the bubble plot

Figure 3. GSEA analysis shows significant up-regulation of genes in anti-bacterial and anti-fungus defense response in sepsis-induced ARDS samples

Figure 4. Protein-protein interaction (PPI) network analysis of DEGs between sepsis-induced ARDS and control samples. Each node represents a gene each edge represents an interaction pair. The size and color of each node shows the importance of the gene in the network

Table 2. Prediction of potential therapeutic drugs for sepsis-induced ARDS by connectivity map

No.	Cmap name	Dose	Cell	Score	Up	Down
1	SC-19,220	10 µM	PC3	−1	−0.148	0.296
2	meteneprost	10 µM	PC3	−0.982	−0.15	0.286
3	doxorubicin	7 µM	MCF7	−0.913	−0.126	0.279
4	BCB000040	10 µM	PC3	−0.906	−0.136	0.266
5	articaine	12 µM	PC3	−0.892	−0.124	0.272
6	isoflupredone	10 µM	PC3	−0.888	−0.14	0.254
7	vinblastine	100 nM	PC3	−0.879	−0.123	0.267
8	heptaminol	22 µM	PC3	−0.87	−0.149	0.238
9	AR-A014418	10 µM	PC3	−0.864	−0.145	0.239
10	3-acetamidocoumarin	20 µM	PC3	−0.864	−0.172	0.211
11	5,253,409	17 µM	MCF7	−0.86	−0.133	0.249
12	alsterpaullone	10 µM	PC3	−0.859	−0.106	0.275
13	bacampicillin	8 µM	MCF7	−0.858	−0.163	0.218
14	chlorpropamide	100 µM	MCF7	−0.857	−0.108	0.272
15	metampicillin	10 µM	PC3	−0.855	−0.128	0.251
16	estropipate	9 µM	MCF7	−0.852	−0.149	0.229
17	etilefrine	18 µM	PC3	−0.848	−0.138	0.239
18	homatropine	11 µM	MCF7	−0.845	−0.132	0.243
19	trichostatin A	100 nM	MCF7	−0.844	−0.132	0.243
20	fursultiamine	9 µM	MCF7	−0.843	−0.127	0.247

Figure 5. Molecular docking analysis of candidate drugs predicted by Cmap analysis. The lower panel shows the 2D schema chart of TOP2A endogenous ligands and Doxorubicin docking with TOP2A