COVID-19 patient transcriptomic and genomic profiling reveals comorbidity interactions with psychiatric disorders

By mid-August 2020, the global COVID-19 pandemic due to novel coronavirus SARS-CoV-2 resulted in over 20 million infections and three quarters of a million deaths worldwide. Asymptomatic spread and a long incubation period contributed to rapid transmission, necessitating public health measures such as quarantine, social distancing, isolation, school and business closures, and border controls. While vital, these strategies increase psychological stress that may exacerbate psychiatric disorders in susceptible individuals. It is unclear whether genomic functional changes in COVID-19 are involved in biological pathways inherent to psychiatric disorders. COVID-19 commonly presents with headache and other central nervous system (CNS) symptoms such as dizziness, seizures, confusion, and stroke. SARS-CoV-2 induces a complex immune response in which typical antiviral responses (e.g., type I interferon) can be unusually low, yet severe cases exhibit cytokine storms with highly elevated inflammatory factors, potentially interacting with comorbid cardiovascular and pulmonary conditions. Evidence also suggests SARS-CoV-2 infects neurons and may directly affect CNS function, and strong humoral immune responses may rapidly affect CNS function indirectly, especially in susceptible individuals. Characterizing these interactions and the immune and inflammatory underpinnings is important. Given the multifaceted pathophysiology in severe COVID-19, comparing common inflammatory substrates and shared signaling and ontology pathways with psychiatric disorders holds promise. SARS-CoV-2 enters many cell types, including neurons, via ACE2, and there are documented pathogen-triggered acute psychiatric manifestations (e.g., PANDAS) driven by autoimmune mechanisms. Dysregulated immune responses and resulting neuroinflammation can cause or exacerbate neurological dysfunction. Psychological distress from the pandemic and infection stress may activate the HPA axis and sympathetic nervous system, altering glucocorticoids and catecholamines; immunosuppressive glucocorticoid effects have been linked to PTSD symptoms and may exacerbate emotional dysregulation and physical comorbidities. Neuroendocrine-immune dysregulation in PTSD can yield a pro-inflammatory state associated with multiple systemic diseases. Bipolar disorder and schizophrenia are also linked to immune dysfunction and higher risks of respiratory diseases, and individuals with schizophrenia have increased infection risks and poorer outcomes during the pandemic. These lines of evidence suggest COVID-19 infection may share pathological determinants with psychiatric disorders through which they may interact. To investigate, the authors applied computational and bioinformatics approaches to study SARS-CoV-2 blood cell and immune panel transcriptomes and GWAS data, identifying SARS-CoV-2 acute response genes concordant with bipolar disorder, PTSD, and schizophrenia, and used functional and data mining analyses to determine how SARS-CoV-2 response genes and pathways may interact with these disorders.

• Study design and data sources: Publicly available datasets were used, including RNA-Seq transcriptomes, targeted immune panel expression data, GWAS, and whole-genome sequencing (WGS) datasets. COVID-19 patient data and datasets for bipolar disorder, PTSD, and schizophrenia were curated from established repositories (including COVID-19 Host Genetics Initiative results for COVID-19 GWAS; GWAS Catalogue, GWAS ATLAS, UK Biobank, dbGaP, PheGenI, and ClinVar for psychiatric GWAS/WGS).
• Transcriptomic analyses (PBMC): RNA-Seq gene expression profiles from peripheral blood mononuclear cells (PBMCs) of COVID-19 patients were analyzed to identify differentially expressed genes (DEGs) versus controls using stringent thresholds: absolute log2 fold change > 1 and adjusted p-value < 0.05. DEGs from COVID-19 were compared to significant upregulated/downregulated genes from PBMC RNA-Seq datasets of bipolar disorder, PTSD, and schizophrenia to identify shared dysregulated genes. Visualization included Venn diagrams, heat maps of log fold changes and adjusted p-values, and bubble plots combining effect sizes and significance.
• Targeted immune panel profiling: A NanoString-based targeted immune panel comprising 759 immune-related genes in PBMCs from COVID-19 patients and controls was analyzed. Genes significantly altered in COVID-19 (versus controls) were identified and overlapped with significant genes from the psychiatric disorder datasets. Overlaps were visualized with bubble plots and heat maps.
• Pathway and ontology enrichment: Gene Ontology (GO) biological processes and curated signaling pathway databases (WikiPathways, BioCarta, Reactome, Panther) were used to perform functional enrichment analyses on shared/concordant genes between COVID-19 and each psychiatric disorder. Gene Set Enrichment Analysis (GSEA) and complementary functional enrichment approaches were applied to assess immune and inflammatory pathway perturbations common to conditions.
• Co-expression and network analyses: Co-expression analyses and network-based clustering were conducted on concordant genes to identify gene clusters shared across COVID-19 and the psychiatric disorders, assessing network-level concordance.
• Genomic association analyses: COVID-19 GWAS summary results were queried (p-value threshold < 1.0E-6) to identify 146 COVID-19-associated genes. Psychiatric disorder-associated genes were obtained from multiple GWAS/WGS resources. Cross-comparisons were performed to identify concordant biomarkers between COVID-19-associated genes and those implicated in bipolar disorder, PTSD, and schizophrenia.
• Integration and interpretation: Findings from transcriptomic overlaps, immune panel profiling, pathway/ontology enrichment, co-expression networks, and GWAS cross-comparisons were integrated to infer potential biological interactions and shared mechanisms between COVID-19 and the psychiatric disorders.

• PBMC transcriptome overlap: Of the top 1289 COVID-19 response genes, 73 were similarly up- or downregulated in at least one of the psychiatric disorders. COVID-19 PBMC profiles shared 39, 19, and 22 dysregulated genes with bipolar disorder, PTSD, and schizophrenia, respectively.
• Common gene across all four conditions: RHOBTB3 was the only gene commonly dysregulated among COVID-19, bipolar disorder, PTSD, and schizophrenia. Additional overlaps included VCAN and RHOBTB3 (COVID-19, bipolar disorder, PTSD) and PTGDR, SH2D1B, AKR1C3, YES1, and RHOBTB3 (COVID-19, bipolar disorder, schizophrenia).
• Distinctive COVID-19 signature: Despite overlaps, most DEGs were uniquely affected in each condition, reflecting differing disease processes.
• Pathways and ontology: Enrichment analyses identified inflammatory and immune pathways shared between COVID-19 and the psychiatric disorders, consistent with cytokine-mediated CNS-immune interactions (e.g., IL-1β, IL-6, TNFα pathways) and upstream regulators (CLOCK, ERK1, GSK3β, P11) relevant to neuropsychiatric symptoms.
• Targeted immune panel results: Among 759 immune genes profiled, 145 were significantly altered in COVID-19 PBMCs versus controls. Overlaps with psychiatric disorders from this panel were 9 genes (bipolar disorder), 10 genes (PTSD), and 2 genes (schizophrenia), indicating convergent systemic immune responses.
• GWAS findings: COVID-19 GWAS identified 146 genes at p < 1e-6. Cross-comparison with psychiatric GWAS/WGS datasets revealed common biomarkers concordant between COVID-19 genetic susceptibility and those implicated in bipolar disorder, PTSD, and schizophrenia (details in supplementary resources referenced by the authors).
• Overall correlation: PTSD pathway profiles appeared most highly correlated with COVID-19 among the three psychiatric disorders, potentially reflecting strong stress-immune system interactions.
• Clinical implications: The identification of common inflammatory pathways suggests potential benefit of anti-inflammatory strategies in managing psychiatric comorbidities in the context of COVID-19.

The study addressed whether SARS-CoV-2 infection interacts with biological pathways underlying psychiatric disorders by integrating PBMC transcriptomics, targeted immune profiling, and genomic association data. Although the majority of transcriptional changes were distinct between COVID-19 and psychiatric disorders, a meaningful subset of genes and pathways overlapped, highlighting shared immune-inflammatory mechanisms. The central identification of RHOBTB3 as a gene dysregulated across COVID-19, bipolar disorder, PTSD, and schizophrenia underscores potential convergence in intracellular transport and immune-related processes affecting leukocytes and possibly CNS functions. Enrichment of cytokine and immune signaling pathways aligns with known neuroimmune interactions implicated in mood, cognition, sleep, and stress responses. PTSD exhibited the strongest pathway-level concordance with COVID-19, consistent with HPA axis dysregulation and a pro-inflammatory state in PTSD that could potentiate immune activation during infection. These findings suggest that COVID-19-related immune responses may exacerbate or unmask psychiatric symptoms in susceptible individuals and that modulating inflammation could be a therapeutic avenue. Network and co-expression analyses further support shared regulatory modules, and GWAS cross-comparisons indicate overlapping genetic determinants that may influence susceptibility and clinical outcomes. Collectively, the results provide a systems-level rationale for monitoring and addressing inflammatory processes in psychiatric populations during COVID-19 and for leveraging shared mechanisms to inform treatment strategies.

This work integrates transcriptomic, immune panel, and genomic data to demonstrate that COVID-19 shares immune-inflammatory gene expression signatures and pathways with bipolar disorder, PTSD, and schizophrenia. While overlaps at the gene level are modest, convergent pathways—particularly in cytokine signaling—are evident, with RHOBTB3 emerging as a gene common to all conditions examined. PTSD shows the strongest pathway correlation with COVID-19, supporting the role of stress-immune interactions. These insights suggest potential benefits of anti-inflammatory interventions for psychiatric patients affected by COVID-19 and highlight the value of multi-level data integration for identifying biomarkers and therapeutic targets. Future research should include longitudinal studies to track symptom-immune dynamics, analyses in brain-relevant tissues and single-cell contexts, validation of candidate pathways and genes (e.g., RHOBTB3) in mechanistic models, and refined GWAS meta-analyses to pinpoint causal variants and regulatory networks underlying the observed overlaps.

• Tissue specificity: Analyses relied primarily on PBMCs and peripheral immune panels; CNS tissue and cell-type–specific effects were not directly assessed.
• Heterogeneous datasets: Public datasets differ in cohort characteristics, sample sizes, platforms, and processing pipelines, introducing potential confounding and batch effects.
• Cross-sectional design: The transcriptomic comparisons are largely cross-sectional, limiting causal inference about directionality between infection and psychiatric symptom changes.
• Modest gene-level overlap: The sparse overlap in DEGs limits conclusions about specific shared genes; pathway-level inferences may be more robust but still indirect.
• Limited clinical detail: The provided summary lacks granular clinical metadata (e.g., disease severity, treatments, timing relative to infection), which could influence gene expression and immune profiles.
• GWAS thresholding and mapping: The use of a p < 1e-6 threshold and gene-mapping strategies may include non-causal loci; functional validation is needed to confirm implicated genes and pathways.