Tensor decomposition-based unsupervised feature extraction identifies candidate genes that induce post-traumatic stress disorder-mediated heart diseases

Abstract

Background: Although post-traumatic stress disorder (PTSD) is primarily a mental disorder, it can cause additional symptoms that do not seem to be directly related to the central nervous system, which PTSD is assumed to directly affect. PTSD-mediated heart diseases are some of such secondary disorders. In spite of the significant correlations between PTSD and heart diseases, spatial separation between the heart and brain (where PTSD is primarily active) prevents researchers from elucidating the mechanisms that bridge the two disorders. Our purpose was to identify genes linking PTSD and heart diseases.

Methods: In this study, gene expression profiles of various murine tissues observed under various types of stress or without stress were analyzed in an integrated manner using tensor decomposition (TD).

Results: Based upon the obtained features, ∼ 400 genes were identified as candidate genes that may mediate heart diseases associated with PTSD. Various gene enrichment analyses supported biological reliability of the identified genes. Ten genes encoding protein-, DNA-, or mRNA-interacting proteins-ILF2, ILF3, ESR1, ESR2, RAD21, HTT, ATF2, NR3C1, TP53, and TP63-were found to be likely to regulate expression of most of these ∼ 400 genes and therefore are candidate primary genes that cause PTSD-mediated heart diseases. Approximately 400 genes in the heart were also found to be strongly affected by various drugs whose known adverse effects are related to heart diseases and/or fear memory conditioning; these data support the reliability of our findings.

Conclusions: TD-based unsupervised feature extraction turned out to be a useful method for gene selection and successfully identified possible genes causing PTSD-mediated heart diseases.

Keywords: Feature extraction; Heart disease; Post-traumatic stress disorder; Tensor decomposition.

Fig. 1

The gene expression pattern of 10 gene sets. Each of which includes 100 genes from synthetic data (thus, 1000 of the total of 30,000 genes are being considered). The remaining 29,000 genes do not have any class specificity. Blue squares represent classes where the genes in each gene set are expressive. Ten tissues are assumed to be treated in 10 distinct ways. Thus, in total, there are 100 classes

Fig. 2

Scatter plots involving the second gene’s through fifth gene’s singular value vectors. $x_{{\ell }_1,i_1},2\le {\ell }_1\le 5$, of 1,000 genes (1≤i ₁≤1000) that belong to one of the 10 gene sets. These 10 gene sets are represented by distinct combinations of colors and symbols. The 29,000 genes not included in any of the 10 gene sets are omitted for clarity

Fig. 3

Performance of synthetic data (averaged over 100 trials). BH: Benjamini-Hochberg, FDR: false discovery rate. Red curves: true positive rates, black curves: false positive rates, solid curves: P=0.01, dashed curves: P=0.05, dotted curves: P=0.1, blue dash-and-dot curves: area under the curve (AUC)

Fig. 4

Singular value vectors employed a The second control-related or treatment-related singular value vector, $x_{{\ell }_1=2,j_1}$. Control: j ₁=1, and treatment (stress): j ₁=2. b The fourth tissue singular value vector, $x_{{\ell }_2=4,j_2}$, AY: j ₂=1, HC: j ₂=2, heart: j ₂=8, hemibrain: j ₂=9, and spleen: j ₂=10. Other tissue singular value vectors, ℓ ₂≠4, can be found in Additional file 2

Fig. 5

Graphical representation of relations between the identified biological terms and proteins. Biological temrs (orange) and various protein-, DNA-, or mRNA-binding proteins (cyan). A: “heart contraction” (GO BP), B: “cardiac muscle contraction” (GO BP), C: “protein targeted to ER” (GO BP), D: “SRP-dependent cotranslational protein targeted to membrane” (GO BP), E: “neuron part” (GO CC), F: “Huntington’s disease” (KEGG), G: “Parkinson’s disease” (KEGG), H: “Alzheimer’s disease” (KEGG), I: “Cardiac muscle contraction” (KEGG), J: “nonsense-mediated decay (NMD)” (REACTOME), K: “SRP-dependent cotranslational protein targeted to membrane” (REACTOME). Orange edges: genes shared by biological terms, blue edges: genes targeted by protein-, DNA-, or mRNA-binding proteins. Width of edges is proportional to the number of genes. Sizes of the orange circles representing biological terms are proportional to the number of genes enriched in each biological term among the 457 genes