Deep learning predicts DNA methylation regulatory variants in the human brain and elucidates the genetics of psychiatric disorders.

There is growing evidence for the role of DNA methylation (DNAm) quantitative trait loci (mQTLs) in the genetics of complex traits, including psychiatric disorders. However, due to extensive linkage disequilibrium (LD) of the genome, it is challenging to identify causal genetic variations that drive DNAm levels by population-based genetic association studies. This limits the utility of mQTLs for fine-mapping risk loci underlying psychiatric disorders identified by genome-wide association studies (GWAS). Here we present INTERACT, a deep learning model that integrates convolutional neural networks with transformer, to predict effects of genetic variations on DNAm levels at CpG sites in the human brain. We show that INTERACT-derived DNAm regulatory variants are not confounded by LD, are concentrated in regulatory genomic regions in the human brain, and are convergent with mQTL evidence from genetic association analysis. We further demonstrate that predicted DNAm regulatory variants are enriched for heritability of brain-related traits and improve polygenic risk prediction for schizophrenia across diverse ancestry samples. Finally, we applied predicted DNAm regulatory variants for fine-mapping schizophrenia GWAS risk loci to identify potential novel risk genes. Our study shows the power of a deep learning approach to identify functional regulatory variants that may elucidate the genetic basis of complex traits.

Bispectral EEG (BSEEG) Algorithm Captures High Mortality Risk Among 1,077 Patients: Its Relationship to Delirium Motor Subtype.

OBJECTIVE: Delirium is dangerous and a predictor of poor patient outcomes. We have previously reported the utility of the bispectral EEG (BSEEG) with a novel algorithm for the detection of delirium and prediction of patient outcomes including mortality. The present study employed a normalized BSEEG (nBSEEG) score to integrate the previous cohorts to combine their data to investigate the prediction of patient outcomes. We also aimed to test if the BSEEG method can be applicable regardless of age, and independent of delirium motor subtypes. METHODS: We calculated nBSEEG score from raw BSEEG data in each cohort and classified patients into BSEEG-positive and BSEEG-negative groups. We used log-rank test and Cox proportional hazards models to predict 90-day and 1-year outcomes for the BSEEG-positive and -negative groups in all subjects and motor subgroups. RESULTS: A total of 1,077 subjects, the BSEEG-positive group showed significantly higher 90-day (hazard ratio 1.33 [95% CI 1.16-1.52] and 1-year (hazard ratio 1.22 [95% CI 1.06-1.40] mortality rates than the negative group after adjustment for covariates such as age, sex, CCI, and delirium status. Among patients with different motor subtypes of delirium, the hypoactive group showed significantly higher 90-day (hazard ratio 1.41 [95% CI 1.12-1.76] and 1-year mortality rates (hazard ratio 1.32 [95% CI 1.05-1.67], which remained significant after adjustment for the same covariates. CONCLUSION: We found that the BSEEG method is capable of capturing patients at high mortality risk.

The genome-wide DNA methylation profiles among neurosurgery patients with and without post-operative delirium.

AIMS: There is no previous study demonstrating the differences of genome-wide DNA methylation (DNAm) profiles between patients with and without postoperative delirium (POD). We aimed to discover epigenetic (DNAm) markers that are associated with POD in blood obtained from patients before and after neurosurgery. METHODS: Pre- and post-surgical blood DNA samples from 37 patients, including 10 POD cases, were analyzed using the Illumina EPIC array genome-wide platform. We examined DNAm differences in blood from patients with and without POD. Enrichment analysis with Gene Ontology and Kyoto Encyclopedia of Genes and Genomes terms were also conducted. RESULTS: When POD cases were tested for DNAm change before and after surgery, enrichment analyses showed many relevant signals with statistical significance in immune response related-pathways and inflammatory cytokine related-pathways such as "cellular response to cytokine stimulus", "regulation of immune system process", "regulation of cell activation", and "regulation of cytokine production". Furthermore, after excluding the potential effect of common factors related to surgery and anesthesia between POD cases and non-POD controls, the enrichment analyses showed significant signals such as "immune response" and "T cell activation", which are same pathways previously identified from an independent non-surgical inpatient cohort. CONCLUSIONS: Our first genome-wide DNAm investigation of POD showed promising signals related to immune response, inflammatory response and other relevant signals considered to be associated with delirium pathophysiology. Our data supports the hypothesis that epigenetics play an important role in the pathophysiological mechanism of delirium and suggest the potential usefulness of an epigenetics-based biomarker of POD.

Evaluation of point-of-care thumb-size bispectral electroencephalography device to quantify delirium severity and predict mortality.

BACKGROUND: We have developed the bispectral electroencephalography (BSEEG) method for detection of delirium and prediction of poor outcomes. AIMS: To improve the BSEEG method by introducing a new EEG device. METHOD: In a prospective cohort study, EEG data were obtained and BSEEG scores were calculated. BSEEG scores were filtered on the basis of standard deviation (s.d.) values to exclude signals with high noise. Both non-filtered and s.d.-filtered BSEEG scores were analysed. BSEEG scores were compared with the results of three delirium screening scales: the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU), the Delirium Rating Scale-Revised-98 (DRS) and the Delirium Observation Screening Scale (DOSS). Additionally, the 365-day mortalities and the length of stay (LOS) in the hospital were analysed. RESULTS: We enrolled 279 elderly participants and obtained 620 BSEEG recordings; 142 participants were categorised as BSEEG-positive, reflecting slower EEG activity. BSEEG scores were higher in the CAM-ICU-positive group than in the CAM-ICU-negative group. There were significant correlations between BSEEG scores and scores on the DRS and the DOSS. The mortality rate of the BSEEG-positive group was significantly higher than that of the BSEEG-negative group. The LOS of the BSEEG-positive group was longer compared with that of the BSEEG-negative group. BSEEG scores after s.d. filtering showed stronger correlations with delirium screening scores and more significant prediction of mortality. CONCLUSIONS: We confirmed the usefulness of the BSEEG method for detection of delirium and of delirium severity, and prediction of patient outcomes with a new EEG device.

Increased mortality in patients with standard EEG findings of 'diffuse slowing'.

BACKGROUND: We aimed to confirm the association between slow brain wave activity typically described as "diffuse slowing" on standard electroencephalogram (EEG) and patient outcomes, including mortality. METHODS: This retrospective study was conducted with patient chart data from March 2015 to March 2017 at a tertiary care academic hospital in the midwestern United States. In total, 1,069 participants age ≥55 years on an inpatient floor or ICU received a standard 24-hour EEG. The primary outcome was all-cause mortality at 30, 90, 180, and 365 days. Secondary outcomes were time to discharge, and discharge to home. RESULTS: Having diffuse slowing on standard EEG was significantly associated with 30-, 90-, 180-, and 365-day mortality compared with patients who had normal EEG findings, after controlling for age, sex, and Charlson Comorbidity Index score. When controlling for these factors, patients with diffuse slowing had a significantly longer time to discharge and were significantly less likely to discharge to home. Our findings showed that a standard EEG finding of diffuse slowing for inpatients age ≥55 years is associated with poor outcomes, including greater mortality. CONCLUSIONS: This study suggested that the finding of diffuse slowing on EEG may be an important clinical marker for predicting mortality in geriatric inpatients.

Increased mortality in patients with standard EEG findings of 'diffuse slowing'.

BACKGROUND: We aimed to confirm the association between slow brain wave activity typically described as "diffuse slowing" on standard electroencephalogram (EEG) and patient outcomes, including mortality. METHODS: This retrospective study was conducted with patient chart data from March 2015 to March 2017 at a tertiary care academic hospital in the midwestern United States. In total, 1,069 participants age ≥55 years on an inpatient floor or ICU received a standard 24-hour EEG. The primary outcome was all-cause mortality at 30, 90, 180, and 365 days. Secondary outcomes were time to discharge, and discharge to home. RESULTS: Having diffuse slowing on standard EEG was significantly associated with 30-, 90-, 180-, and 365-day mortality compared with patients who had normal EEG findings, after controlling for age, sex, and Charlson Comorbidity Index score. When controlling for these factors, patients with diffuse slowing had a significant longer time to discharge and were significantly less likely to discharge to home. Our findings showed that a standard EEG finding of diffuse slowing for inpatients age ≥55 years is associated with poor outcomes, including greater mortality. CONCLUSIONS: This study suggested that the finding of diffuse slowing on EEG may be an important clinical marker for predicting mortality in geriatric inpatients.

Genome-wide DNA methylation investigation of glucocorticoid exposure within buccal samples.

AIM: Glucocorticoids play a major role in regulating the stress response, and an imbalance of glucocorticoids has been implicated in stress-related disorders. Within mouse models, CpGs across the genome have been shown to be differentially methylated in response to glucocorticoid treatment, and using the Infinium 27K array, it was shown that humans given synthetic glucocorticoids had DNA methylation (DNAm) changes in blood. However, further investigation of the extent to which glucocorticoids affect DNAm across a larger proportion of the genome is needed. METHODS: Buccal samples were collected before and after synthetic glucocorticoid treatment in the context of a dental procedure. This included 30 tooth extraction surgery patients who received 10 mg of dexamethasone. Genome-wide DNAm was assessed with the Infinium HumanMethylationEPIC array. RESULTS: Five CpGs showed genome-wide significant DNAm changes that were >10%. These differentially methylated CpGs were in or nearest the following genes: ZNF438, KLHDC10, miR-544 or CRABP1, DPH5, and WDFY2. Using previously published datasets of human blood gene expression changes following dexamethasone exposure, a significant proportion of genes with false-discovery-rate-adjusted significant CpGs were also differentially expressed. A pathway analysis of the genes with false-discovery-rate-adjusted significant CpGs revealed significant enrichment of olfactory transduction, pentose and glucuronate interconversions, ascorbate and aldarate metabolism, and steroid hormone biosynthesis pathways. CONCLUSION: High-dose synthetic glucocorticoid administration in the setting of a dental procedure was significantly associated with DNAm changes within buccal samples. These findings are consistent with prior findings of an influence of glucocorticoids on DNAm in humans.

Delirium detection by a novel bispectral electroencephalography device in general hospital.

AIM: Delirium is common and dangerous among elderly inpatients; yet, it is underdiagnosed and thus undertreated. This study aimed to test the diagnostic characteristics of a noninvasive point-of-care device with two-channel (bispectral) electroencephalography (EEG) for the screening of delirium in the hospital. METHODS: Patients admitted to the University of Iowa Hospitals and Clinics were assessed for the presence of delirium with a clinical assessment, the Confusion Assessment Method for Intensive Care Unit and Delirium Rating Scale. Subsequently, we obtained a 10-min bispectral EEG (BSEEG) recording from a hand-held electroencephalogram device during hospitalization. We performed power spectral density analysis to differentiate between those patients with and without delirium. RESULTS: Initially 45 subjects were used as a test dataset to establish a cut-off. The BSEEG index was determined to be a significant indicator of delirium, with sensitivity 80% and specificity 87.7%. An additional independent validation dataset with 24 patients confirmed the validity of the approach, with a sensitivity of 83.3% and specificity of 83.3%. CONCLUSION: In this pilot study, the BSEEG method was able to distinguish delirious patients from non-delirious patients. Our data showed the feasibility of this technology for mass screening of delirium in the hospital.

Treatment outcomes of depression: the pharmacogenomic research network antidepressant medication pharmacogenomic study.

BACKGROUND: The effectiveness of selective serotonin reuptake inhibitors (SSRIs) in patients with major depressive disorder (MDD) is controversial. AIMS: The clinical outcomes of subjects with nonpsychotic MDD were reported and compared with the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study outcomes to provide guidance on the effectiveness of SSRIs. METHODS: Subjects were treated with citalopram/escitalopram for up to 8 weeks. Depression was measured using the Quick Inventory of Depressive Symptomatology-Clinician Rated (QIDS-C16) and the 17-item Hamilton Depression Rating Scale. RESULTS: The group of subjects with at least 1 follow-up visit had a remission (QIDS-C16 ≤ 5) rate of 45.8% as well as a response (50% reduction in QIDS-C16) rate of 64.8%, and 79.9% achieved an improvement of 5 points or higher in QIDS-C16 score. The Pharmacogenomic Research Network Antidepressant Medication Pharmacogenomic Study subjects were more likely to achieve a response than STAR*D study subjects. After adjustment for demographic factors, the response rates were not significantly different. When reporting the adverse effect burden, 60.5% of the subjects reported no impairment, 31.7% reported a minimal-to-mild impairment, and 7.8% reported a moderate-to-severe burden at the 4-week visit. CONCLUSIONS: Patients contemplating initiating an SSRI to treat their MDD can anticipate a high probability of symptom improvement (79.9%) with a low probability that their symptoms will become worse. Patients with lower baseline severity have a higher probability of achieving remission. The Pharmacogenomic Research Network Antidepressant Medication Pharmacogenomic Study replicates many findings of the first phase of the STAR*D study after controlling for the differences between the studies.

New developments in the genetics of bipolar disorder.

The last several years have been breakthrough ones in bipolar disorder (BPD) genetics, as the field has identified robust risk variants for the first time. Leading the way have been genome-wide association studies (GWAS) that have assessed common genetic markers across very large groups of patients and controls. These have resulted in findings in genes including ANK3, CACNA1C, SYNE1, ODZ4, and TRANK1. Additional studies have begun to examine the biology of these genes and how risk variants influence aspects of brain and behavior that underlie BPD. For example, carriers of the CACNA1C risk variant have been found to exhibit hippocampal and anterior cingulate dysfunction during episodic memory recall. This work has shed additional light on the relationship of bipolar susceptibility variants to other disorders, particularly schizophrenia. Even larger BPD GWAS are expected with samples now amassed of 21,035 cases and 28,758 controls. Studies have examined the pharmacogenomics of BPD with studies of lithium response, yielding high profile results that remain to be confirmed. The next frontier in the field is the identification of rare bipolar susceptibility variants through large-scale DNA sequencing. While only a couple of papers have been published to date, many studies are underway. The Bipolar Sequencing Consortium has been formed to bring together all of the groups working in this area, and to perform meta-analyses of the data generated. The consortium, with 13 member groups, now has exome data on ~3,500 cases and ~5,000 controls, and on ~162 families. The focus will likely shift within several years from exome data to whole genome data as costs of obtaining such data continue to drop. Gene-mapping studies are now providing clear results that provide insights into the pathophysiology of the disorder. Sequencing studies should extend this process further. Findings could eventually set the stage for rational therapeutic development.

Genome-wide DNA methylation analysis in female veterans with military sexual trauma and comorbid PTSD/MDD.

BACKGROUND: Military sexual trauma (MST) is a prevalent issue within the U.S. military. Victims are more likely to develop comorbid diseases such as posttraumatic stress disorder (PTSD) and major depressive disorder (MDD). Nonetheless, not everyone who suffers from MST develops PTSD and/or MDD. DNA methylation, which can regulate gene expression, might give us insight into the molecular mechanisms behind this discrepancy. Therefore, we sought to identify genomic loci and enriched biological pathways that differ between patients with and without MST, PTSD, and MDD. METHODS: Saliva samples were collected from 113 female veterans. Following DNA extraction and processing, DNA methylation levels were measured through the Infinium HumanMethylationEPIC BeadChip array. We used limma and bump hunting methods to generate the differentially methylated positions and differentially methylated regions (DMRs), respectively. Concurrently, we used Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genome to find enriched pathways. RESULTS: A DMR close to the transcription start site of ZFP57 was differentially methylated between subjects with and without PTSD, replicating previous findings and emphasizing the potential role of ZFP57 in PTSD susceptibility. In the pathway analyses, none survived multiple correction, although top GO terms included some potentially relevant to MST, PTSD, and MDD etiology. CONCLUSION: We conducted one of the first DNA methylation analyses investigating MST along with PTSD and MDD. In addition, we found one DMR near ZFP57 to be associated with PTSD. The replication of this finding indicates further investigation of ZFP57 in PTSD may be warranted.

Correlation of telomere length in brain tissue with peripheral tissues in living human subjects.

Telomeres are important to chromosomal stability, and changes in their length correlate with disease, potentially relevant to brain disorders. Assessing telomere length in human brain is invasive, but whether peripheral tissue telomere length correlates with that in brain is not known. Saliva, buccal, blood, and brain samples were collected at time points before, during, and after subjects undergoing neurosurgery (n = 35) for intractable epilepsy. DNA was isolated from samples and average telomere length assessed by qPCR. Correlations of telomere length between tissue samples were calculated across subjects. When data were stratified by sex, saliva telomere length correlated with brain telomere length in males only. Buccal telomere length correlated with brain telomere length when males and females were combined. These findings indicate that in living subjects, telomere length in peripheral tissues variably correlates with that in brain and may be dependent on sex. Peripheral tissue telomere length may provide insight into brain telomere length, relevant to assessment of brain disorder pathophysiology.

The Genome-wide DNA methylation changes in gastrointestinal surgery patients with and without postoperative delirium: Evidence of immune process in its pathophysiology.

AIM: The pathophysiological mechanisms of postoperative delirium (POD) are still unclear, and there is no reliable biomarker to distinguish between those with and without POD. Our aim was to discover DNAm markers associated with POD in blood collected from patients before and after gastrointestinal surgery. METHOD: We collected blood samples from 16 patients including 7 POD patients at three timepoints; before surgery (pre), the first and third postoperative days (day1 and day3). We measured differences in DNA methylation between POD and control groups between pre and day1 as well as between pre and day3 using the Illumina EPIC array method. Besides, enrichment analysis with Gene Ontology and Kyoto Encyclopedia of Genes and Genomes terms were also performed after excluding influence of common factors related to surgery and anesthesia. RESULT: The results showed that pre and day1 comparisons showed that immune and inflammatory signals such as 'T-cell activation' were significantly different, consistent with our previous studies with non-Hispanic White subjects. In contrast, we found that these signals were not significant any more when pre was compared with day3. CONCLUSION: These results provide strong evidence for the involvement of inflammatory and immune-related epigenetic signals in the pathogenesis of delirium, including POD, regardless of ethnic background. These findings also suggest that DNAm, which is involved in inflammation and immunity, is dynamically altered in patients with POD. In summary, the present results indicate that these signals may serve as a new diagnostic tool for POD.

The Bispectral Electroencephalography Method Quantifies Postoperative Delirium-Like States in Young and Aged Male Mice After Head-Mount Implantation Surgery.

Delirium, a syndrome characterized by an acute change in attention, awareness, and cognition, is commonly observed in older adults, although there are few quantitative monitoring methods in the clinical setting. We developed a bispectral electroencephalography (BSEEG) method capable of detecting delirium and can quantify the severity of delirium using a novel algorithm. Preclinical application of this novel BSEEG method can capture a delirium-like state in mice following lipopolysaccharide administration. However, its application to postoperative delirium (POD) has not yet been validated in animal experiments. This study aimed to create a POD model in mice with the BSEEG method by monitoring BSEEG scores following EEG head-mount implantation surgery and throughout the recovery. We compared the BSEEG scores of C57BL/6J young (2-3 months old) with aged (18-19 months old) male mice for quantitative evaluation of POD-like states. Postoperatively, both groups displayed increased BSEEG scores and a loss of regular diurnal changes in BSEEG scores. In young mice, BSEEG scores and regular diurnal changes recovered relatively quickly to baseline by postoperative day (PO-Day) 3. Conversely, aged mice exhibited prolonged increases in postoperative BSEEG scores and it reached steady states only after PO-Day 8. This study suggests that the BSEEG method can be utilized as a quantitative measure of POD and assess the effect of aging on recovery from POD in the preclinical model.

Epigenetic signals associated with delirium replicated across four independent cohorts.

Delirium is risky and indicates poor outcomes for patients. Therefore, it is crucial to create an effective delirium detection method. However, the epigenetic pathophysiology of delirium remains largely unknown. We aimed to discover reliable and replicable epigenetic (DNA methylation: DNAm) markers that are associated with delirium including post-operative delirium (POD) in blood obtained from patients among four independent cohorts. Blood DNA from four independent cohorts (two inpatient cohorts and two surgery cohorts; 16 to 88 patients each) were analyzed using the Illumina EPIC array platform for genome-wide DNAm analysis. We examined DNAm differences in blood between patients with and without delirium including POD. When we compared top CpG sites previously identified from the initial inpatient cohort with three additional cohorts (one inpatient and two surgery cohorts), 11 of the top 13 CpG sites showed statistically significant differences in DNAm values between the delirium group and non-delirium group in the same directions as found in the initial cohort. This study demonstrated the potential value of epigenetic biomarkers as future diagnostic tools. Furthermore, our findings provide additional evidence of the potential role of epigenetics in the pathophysiology of delirium including POD.

Metformin use history and genome-wide DNA methylation profile: potential molecular mechanism for aging and longevity.

BACKGROUND: Metformin, a commonly prescribed anti-diabetic medication, has repeatedly been shown to hinder aging in pre-clinical models and to be associated with lower mortality for humans. It is, however, not well understood how metformin can potentially prolong lifespan from a biological standpoint. We hypothesized that metformin's potential mechanism of action for longevity is through its epigenetic modifications. METHODS: To test our hypothesis, we conducted a post-hoc analysis of available genome-wide DNA methylation (DNAm) data obtained from whole blood collected from inpatients with and without a history of metformin use. We assessed the methylation profile of 171 patients (first run) and only among 63 diabetic patients (second run) and compared the DNAm rates between metformin users and nonusers. RESULTS: Enrichment analysis from the Kyoto Encyclopedia of Genes and Genome (KEGG) showed pathways relevant to metformin's mechanism of action, such as longevity, AMPK, and inflammatory pathways. We also identified several pathways related to delirium whose risk factor is aging. Moreover, top hits from the Gene Ontology (GO) included HIF-1α pathways. However, no individual CpG site showed genome-wide statistical significance (p < 5E-08). CONCLUSION: This study may elucidate metformin's potential role in longevity through epigenetic modifications and other possible mechanisms of action.

Cross-tissue correlations of genome-wide DNA methylation in Japanese live human brain and blood, saliva, and buccal epithelial tissues.

Neuroepigenetics considers genetic sequences and the interplay with environmental influences to elucidate vulnerability risk for various neurological and psychiatric disorders. However, evaluating DNA methylation of brain tissue is challenging owing to the issue of tissue specificity. Consequently, peripheral surrogate tissues were used, resulting in limited progress compared with other epigenetic studies, such as cancer research. Therefore, we developed databases to establish correlations between the brain and peripheral tissues in the same individuals. Four tissues, resected brain tissue, blood, saliva, and buccal mucosa (buccal), were collected from 19 patients (aged 13-73 years) who underwent neurosurgery. Moreover, their genome-wide DNA methylation was assessed using the Infinium HumanMethylationEPIC BeadChip arrays to determine the cross-tissue correlation of each combination. These correlation analyses were conducted with all methylation sites and with variable CpGs, and with when these were adjusted for cellular proportions. For the averaged data for each CpG across individuals, the saliva-brain correlation (r = 0.90) was higher than that for blood-brain (r = 0.87) and buccal-brain (r = 0.88) comparisons. Among individual CpGs, blood had the highest proportion of CpGs correlated to the brain at nominally significant levels (19.0%), followed by saliva (14.4%) and buccal (9.8%). These results were similar to the previous IMAGE-CpG results; however, cross-database correlations of the correlation coefficients revealed a relatively low (brain vs. blood: r = 0.27, saliva: r = 0.18, and buccal: r = 0.24). To the best of our knowledge, this is the fifth study in the literature initiating the development of databases for correlations between the brain and peripheral tissues in the same individuals. We present the first database developed from an Asian population, specifically Japanese samples (AMAZE-CpG), which would contribute to interpreting individual epigenetic study results from various Asian populations.

NSAIDs use history: impact on the genome-wide DNA methylation profile and possible mechanisms of action.

BACKGROUND AND OBJECTIVE: NSAIDs inhibit cyclooxygenase, but their role in aging and other diseases is not well understood. Our group previously showed the potential benefit of NSAIDs in decreasing the risk of delirium and mortality. Concurrently, epigenetics signals have also been associated with delirium. Therefore, we sought to find differentially methylated genes and biological pathways related to exposure with NSAIDs by comparing the genome-wide DNA methylation profiles of patients with and without a history of NSAIDs use. METHODS: Whole blood samples were collected from 171 patients at the University of Iowa Hospital and Clinics from November 2017 to March 2020. History of NSAIDs use was assessed through a word-search function in the subjects' electronic medical records. DNA was extracted from the blood samples, processed with bisulfite conversion, and analyzed using Illumina's EPIC array. The analysis of top differentially methylated CpG sites and subsequent enrichment analysis were conducted using an established pipeline using R statistical software. RESULTS: Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genome (KEGG) showed several biological pathways relevant to NSAIDs' function. The identified GO terms included "arachidonic acid metabolic process," while KEGG results included "linoleic acid metabolism," "cellular senescence," and "circadian rhythm." Nonetheless, none of the top GO and KEGG pathways and the top differentially methylated CpG sites reached statistical significance. CONCLUSION: Our results suggest a potential role of epigenetics in the mechanisms of the action of NSAIDs. However, the results should be viewed with caution as exploratory and hypothesis-generating given the lack of statistically significant findings.