Identifying behavior regulatory leverage over mental disorders transcriptomic network hubs toward lifestyle-dependent psychiatric drugs repurposing

Abstract

Background: There is a vast prevalence of mental disorders, but patient responses to psychiatric medication fluctuate. As food choices and daily habits play a fundamental role in this fluctuation, integrating machine learning with network medicine can provide valuable insights into disease systems and the regulatory leverage of lifestyle in mental health.

Methods: This study analyzed coexpression network modules of MDD and PTSD blood transcriptomic profile using modularity optimization method, the first runner-up of Disease Module Identification DREAM challenge. The top disease genes of both MDD and PTSD modules were detected using random forest model. Afterward, the regulatory signature of two predominant habitual phenotypes, diet-induced obesity and smoking, were identified. These transcription/translation regulating factors (TRFs) signals were transduced toward the two disorders' disease genes. A bipartite network of drugs that target the TRFS together with PTSD or MDD hubs was constructed.

Results: The research revealed one MDD hub, the CENPJ, which is known to influence intellectual ability. This observation paves the way for additional investigations into the potential of CENPJ as a novel target for MDD therapeutic agents development. Additionally, most of the predicted PTSD hubs were associated with multiple carcinomas, of which the most notable was SHCBP1. SHCBP1 is a known risk factor for glioma, suggesting the importance of continuous monitoring of patients with PTSD to mitigate potential cancer comorbidities. The signaling network illustrated that two PTSD and three MDD biomarkers were co-regulated by habitual phenotype TRFs. 6-Prenylnaringenin and Aflibercept were identified as potential candidates for targeting the MDD and PTSD hubs: ATP6V0A1 and PIGF. However, habitual phenotype TRFs have no leverage over ATP6V0A1 and PIGF.

Conclusion: Combining machine learning and network biology succeeded in revealing biomarkers for two notoriously spreading disorders, MDD and PTSD. This approach offers a non-invasive diagnostic pipeline and identifies potential drug targets that could be repurposed under further investigation. These findings contribute to our understanding of the complex interplay between mental disorders, daily habits, and psychiatric interventions, thereby facilitating more targeted and personalized treatment strategies.

Keywords: Depression; Drug repurposing; Machine learning; Mental disorders; Network biology; Obesity; PTSD; Signal transduction; Smoking; Unhealthy food.

Fig. 1

The plots discern the Quality Control (QC) of the harmonization process, a shows the density and Q-Q plots generated by the ComBat function using the datasets mean and variance on the upper and lower parts, respectively. While b shows that the array variance differed significantly before and after batch correction. The last QC stage is the MDS plot in c, which illustrates that batch correction was successful for the MDD discovery metadata datasets

Fig. 2

Represents the enriched biological processes of MDD and PTSD DEGs. a shows a tree plot for the five ancestral terms enriched by MDD DEGs colored in blue, light-green, red, purple, and green. On the other hand, b displays an emaplot for the five ancestral biological terms enriched by PTSD DEGs, the clusters colored uniquely in the following: purple, pink, brown, mint, and light blue. Two scales were found on the right of both plots: the upper scale represents the number of genes detected in each term; the larger the size of the circle, the larger the gene set size of the term, while the lower scale classifies the terms according to its sign the negatively enriched terms found in red, while the positively enriched terms are colored in blue

Fig. 3

The figure collectively represents a comparison between the MDD and PTSD DEGs in two terms. a: compares the commonly enriched C7 gene sets between MDD and PTSD in the form of a ridge plot; the higher the significance of the set, the more it tends to be represented in blue, and vice versa. b, c: display the cyclic chord and bar plots of enriched pathways of MDD and PTSD DEGs, respectively. The chord color is a code for each enriched pathway, and the square color of each gene represents the LogFC value, whereas red indicates upregulation and cyan indicates downregulation. While the color of each dot in the PTSD bar plot identifies the source database, pathways are ordered from top to bottom in descending order according to the NES value

Fig. 4

The upper part shows the ROC curve for OOB sampling built for MDD modules. This plot illustrates the performance of the model in distinguishing between patients with MDD and controls. The color scale ranges from blue to red, such that blue indicates the lowest AUC value and red indicates higher values. The lower section represents the distribution of the mean minimum depth for top-ranked MDD genes. The genes with the lowest minimal depth are colored pink, and the color of the scale changes until it reaches gray at the end of the permutations

Fig. 5

The upper part shows the ROC curve for OOB sampling built for PTSD modules. This plot illustrates the performance of the model in distinguishing between patients with PTSD and controls. The color scale ranges from blue to red, such that blue indicates the lowest AUC value and red indicates higher values. The lower section represents the distribution of the mean minimum depth for top-ranked PTSD genes. The genes with the lowest minimal depth are colored pink, and the color of the scale changes until it reaches gray at the end of the permutations

Fig. 6

The figure shows the signal transduction network and drug-gene interaction network. a, b shows the four signal transduction paths of the four TRFs toward TTF1 gene in the upper part and FOXO1 gene in the lower part. The blue line in the network represents positive regulation, and the red-colored edges represent negative regulation of the target node. In c the drugs targeting lifestyle-based and disease genes are shown. Drug nodes are colored as follows: purple, dark blue, and faint yellow for drugs targeting MDD, PTSD hubs, and TRFs (MYC, GATA2, TCF4), respectively. In contrast, the gene source nodes are green