RT Journal Article
SR Electronic
T1 Mining common pathway deregulation profiles in lung diseases
JF European Respiratory Journal
JO Eur Respir J
FD European Respiratory Society
SP P2021
VO 44
IS Suppl 58
A1 Arsen Arakelyan
A1 Lilit Nersisyan
A1 Henry Wirth
A1 Hans Binder
YR 2014
UL http://erj.ersjournals.com/content/44/Suppl_58/P2021.abstract
AB This study is aimed at investigating lung diseases described by common pathomechanisms based on evaluation of gene expression profiles in molecular pathways. 16 datasets containing 428 samples for 22 health conditions were taken from Gene Expression Omnibus. Self organizing maps (Wirth, H. et al.BMC Bioinformatics 2011;12:306) and cluster analysis with dynamic tree cut were used for gene expression based disease clustering. In-house pathway signal flow algorithm and phylogenetic analysis were applied to find common pathway deregulation patterns in clusters.Analysis resulted in grouping the 22 conditions into 5 clusters (fig.1). PSF and phylogenetic analysis identified unique pathway deregulation patterns for each cluster (fig.2).Figure 1. Clustering of the 22 health conditions based on gene expression patterns.Figure 2. Venn diagram of cluster-specific pathways deregulation.Our results allowed defining lung disease clusters characterized by common profiles of molecular pathway deregulation, as well as identifying pathways deregulation patterns exclusive to each cluster. These findings may shed light on commonalities and specificities of pathomechanisms of those diseases.