Table 1.
Comparison of enriched pathways and biological processes (annotated by Hallmark molecular signatures database) between COPD stable-state and exacerbation conditions
| Pathways and biological processes (Hallmark MSigDB) | Hallmark MSigDB systematic name | MNGCD analysis | GSEA | |
|---|---|---|---|---|
| Exacerbation, average p value (278 samples) | Stable, average p value 209 samples) | Exacerbation–stable contrast, corrected p value | ||
| E2F_TARGETS | M5925 | 10−5 | NS | NS |
| ESTROGEN_RESPONSE_LATE | M5907 | 5•10−4 | NS | NS |
| G2M_CHECKPOINT | M5901 | 10−7 | NS | NS |
| KRAS_SIGNALING_DN | M5956 | 3•10−3 | NS | NS |
| MITOTIC_SPINDLE | M5893 | 10−3 | NS | NS |
| HEME_METABOLISM | M5945 | 6•10−3 | 5•10−4 | 3•10−3 |
| INTERFERON_ALPHA_RESPONSE | M5911 | 8•10−6 | 2•10−4 | 3•10−3 |
| INTERFERON_GAMMA_RESPONSE | M5913 | 7•10−8* | 10-4* | 3•10−3 |
| ALLOGRAFT_REJECTION | M5950 | NS | 6•10−3 | NS |
Results of the MNGCD analysis are reported separately for stable and exacerbation conditions, and as direct contrast for the GSEA analysis (significance p < 0.01 after Benjamini-Hochberg correction). In the MNGCD analysis, for multiple communities enriched with the same Hallmark term, the minimum p value for that repetition was taken and p values of the 20 iterations were then log-averaged (details in Supplementary Information Section 3).
COPD chronic obstructive pulmonary disease, GSEA gene set enrichment analysis, MNGCD multi-network gene community detection, MSigDB molecular signatures database, NS not significant.
*p value was significant in more than 80% of the iterations but floating.