Genome-wide association studies (GWAS) identify genomic regions associated with complex traits, but interpreting these regions requires further analysis of local linkage disequilibrium (LD), marker–trait relationships and multilocus genotype patterns. In this study, a Python-based graphical user interface (GUI) was developed to integrate regional post-GWAS analyses and provide a button-driven alternative to command-line workflows. The platform includes PLINK-based regional marker extraction, LD calculation, LD decay visualisation, exploratory regional SNP–trait analysis, candidate LD region detection, graph-based LD clustering and pairwise genotype-combination analysis. As a proof of concept, the workflow was applied to publicly available genotype data from the Bengal and Assam Aus Panel and grain iron (Fe) concentration data from the same population. The analysis focused on a previously reported quantitative trait locus (QTL) located at approximately 29.28 megabase pairs (Mbp) on rice chromosome 7, near the candidate gene nicotianamine synthase 3 (OsNAS3). The GUI extracted 4894 single-nucleotide polymorphisms (SNPs) from a 1-Mbp interval and generated 11,973,171 pairwise LD comparisons. Exploratory regional SNP–trait analysis evaluated 4448 eligible SNPs. Because the GUI screening procedure did not account for population structure or relatedness, the regional SNPs were additionally evaluated using a mixed linear model incorporating five genome-wide principal components and a genomic kinship matrix. The adjusted analysis identified 133 Bonferroni-significant and 1635 Benjamini–Hochberg-significant SNPs, including marginal adjusted signals within the genomic neighbourhood of OsNAS3. At the specified clumping parameters, all 133 Bonferroni-significant SNPs were assigned to a single clump indexed by SNP 7:29285763, and no residual association remained significant after conditioning on this lead marker and correcting for multiple testing. Among 43,639 SNP pairs separated by no more than 1000 base pairs (bp), 22,952 had r2 ≥ 0.6. These strong-LD pairs were summarised as 116 candidate LD regions using interval-based aggregation and 339 graph-based LD clusters using connected-components analysis. Of the 22,952 strong-LD SNP pairs, 21,646 met the genotype-call eligibility criteria for pairwise genotype-combination analysis. Although exploratory pairwise analysis of variance (ANOVA) identified numerous corrected genotype–phenotype differences, nested-model comparisons provided no corrected evidence that any two-locus genotype combination explained additional phenotypic variation beyond its better constituent single-marker model. These findings reflect the extensive LD within the selected QTL interval and represent regional prioritisation rather than identification of causal variants. Overall, the GUI provides a reproducible framework for characterising GWAS-derived QTL intervals, regional LD patterns and genotype–phenotype relationships. Its broader applicability should be further evaluated using additional traits, populations, genomic regions and organisms with compatible PLINK genotype data.
Oral squamous cell carcinoma (OSCC) is typically diagnosed at an advanced stage and thus there is a need for biomarkers for OSCC which can be detected with minimum invasiveness. Although circulating microRNAs (miRNAs) have demonstrated great diagnostic promise, there are limited publicly available circulating miRNA datasets associated with OSCC. To exploratively integrate public tissue miRNA datasets using machine learning methods to identify candidate miRNAs as potential OSCC biomarkers. The miRNA profiling of multiple OSCC tissue datasets from GEO and TCGA containing both tumor and control samples were integrated for biomarker discovery. Tumor samples only were excluded prior to the model development. Machine learning models were developed using differentially expressed set of miRNAs. Five classifiers were evaluated and the best model was externally validated by applying it to the only publicly available independent circulating serum miRNA dataset. The signature was subjected to a continuous probability score. The model interpretation was performed using SHAP. A 30-miRNA candidate signature associated with OSCC was identified. In internal cross-validation, the panel separated tumor from normal tissue (AUC 0.96). The logistic regression model had the highest AUC, while XGBoost and SVM-RBF had slightly higher accuracy (Table 5) and gave an AUC of 0.67 in external validation. The held-out tissue-level performance was variable and lower (AUC 0.54–0.88, Table 7) and the external serum AUC was modest (0.67). The panel identified included known OSCC-related miRNAs (miR-21, miR-146b-5p) and a number of less studied candidate biomarkers whose function remains to be tested. The study demonstrates that a machine learning method can be applied to integrate public tissue miRNA datasets for identification of circulating miRNA biomarkers in OSCC. This is an exploratory study; the signature should not be considered as a validated serum test, but rather a tissue-derived candidate panel that offers limited transferability to serum samples and needs to be validated in more, larger, and independent circulating samples before it can be applied to a diagnostic purpose.
Abstract Genome-scale metabolic models (GSMMs) of Campylobacter jejuni ( C. jejuni ) have been studied mainly for reference strains, and their characteristics at the lineage level remain poorly understood. We reconstructed GSMMs for 86 C. jejuni isolates collected in Korea using gapseq and analyzed them together with in vitro substrate utilization data for seven substrates measured on the same isolates. Strain-specific GSMMs clustered by clonal complex (CC): in all nine CCs with ≥ 2 isolates the mean within-CC Jaccard distance was smaller than the mean between-CC distance; across all strain pairs the median distance was 0.082 within CCs and 0.130 between CCs ( P < 0.001). In the in vitro phenotypes, only five of the nine CCs showed smaller within- than between-CC distances, and the GSMM and in vitro distances were only weakly correlated (Mantel r = 0.13, P = 0.03). In an exploratory comparison of two lineages with contrasting profiles, CC-443 ( n = 7) showed negative mean z -scores for all seven substrates, whereas CC-658 ( n = 4) showed positive mean z -scores for all seven, the highest being sulfite (z = + 0.95). In GSMMs, CC-443 lacked reactions for the monoamine/polyamine oxidase family and pantothenate/CoA biosynthesis, whereas CC-658 was enriched in fatty acid/phospholipid biosynthesis, sialic acid metabolism, and anaerobic sulfur respiration. Anaerobic sulfur respiration-related reactions were more prevalent in CC-658 (75%) than in the other CCs (11%), consistent with its high in vitro sulfite utilization; this difference was retained when the comparison was restricted to chicken isolates (0% in CC-443 vs. 100% in CC-658). In silico growth simulations under two gut-mimicking diets (Western and high-fiber) showed consistently lower predicted growth for CC-443 than for CC-658 under both conditions (exact permutation test, P = 0.048), although the eleven models yielded only four distinct predicted values under each diet, which limits the strength of this comparison. These results indicate that metabolic modeling of C. jejuni benefits from considering lineage-level diversity rather than relying on a single reference-strain GSMM.
Abstract Genome-wide association studies (GWAS) have identified numerous non-coding risk variants for prostate and ovarian cancers, yet determining which are regulatorily active and which target genes they control remains challenging. We re-analysed a dinucleotide-barcode massively parallel reporter assay (DiR-seq) of 208 bi-allelic prostate cancer risk single-nucleotide polymorphisms (SNPs) in three prostate cancer cell lines under false-discovery-rate (FDR) control, identifying 50 regulatory SNPs at FDR < 0.10 (30 at FDR < 0.05), including 9 shared by at least two cell lines. Benchmarking standard in-silico functional annotations against this experimental ground truth revealed poor predictive performance (cross-validated ROC-AUC = 0.55–0.59), demonstrating that computational predictions cannot reliably substitute for measured allelic activity. We therefore treated the ovarian analysis as hypothesis-generating. Convergent annotation evidence prioritised 44 of 176 ovarian cancer GWAS leads (31 independent loci); a core of 21 SNPs (17 loci) was stable under alternative weights, leave-one-feature-out scoring and threshold changes, and candidates were enriched in ovarian and fallopian-tube regulatory chromatin (OR = 7.9, FDR = 5 × 10⁻5). Copy-number recurrence at candidate genes agreed between two tumour cohorts (Spearman ρ = 0.75, 95% CI 0.46–0.88) but no more than for chromosome-arm-matched control genes, so this observation is exploratory. Formal colocalisation with full summary statistics supported a shared causal variant for prostate cancer risk and TLCD3A expression at 17p13.3 (PP.H4 = 0.92), whereas the ovarian 17q21.31 signal could not be resolved to a single gene. Transcription-factor motifs disrupted in both cancers were over-represented for the interferon-regulatory factor (IRF) and STAT families (FDR = 3 × 10⁻6), a motif-level hypothesis. In an independent MPRA in primary ovarian epithelial cells, variants disrupting interferon-stimulated response elements (ISREs) were depleted, not enriched, for allele-specific activity (4.6% vs 9.2%; OR = 0.47, nominal p = 0.06, FDR = 0.10), so the IRF/STAT hypothesis was not supported; the same assay confirmed allele-specific activity at one prioritised ovarian candidate, rs2976384 (skew = − 0.67, FDR = 3 × 10⁻5). These results provide an audited prostate regulatory-SNP catalogue, quantify the limits of annotation-based fine-mapping, and offer testable cross-cancer hypotheses.
Periodontitis is a widespread chronic inflammatory condition that frequently coexists with diabetes mellitus (DM). Identifying non-invasive biomarkers is essential for early detection, targeted therapy, and better understanding of the bidirectional relationship between these two diseases. This study aimed to investigate salivary metabolomic profiles in diabetic individuals with and without periodontitis to identify candidate metabolites associated with periodontitis in patients with diabetes. This pilot study compared salivary 1H NMR metabolomic profiles of patients with diabetes with and without periodontitis (n = 13 per group). Unstimulated whole saliva was collected from both groups. Partial least squares discriminant analysis (PLS-DA) was applied to differentiate between the groups. Spectral variables with VIP > 1 were selected for metabolite annotation, followed by metabolite set enrichment analysis (MSEA) to identify dysregulated metabolic pathways. Using the PLS-DA model, a total of 78 discriminating variables were identified, leading to the annotation of ten key metabolites: glycolic acid, thymidine, tyramine, deoxyguanosine, cytidine, uridine, NADPH, ADP, serotonin, and nicotinate. MSEA revealed significant pathway disturbances, with pyrimidine metabolism emerging as a notably enriched pathway. The metabolomics study of saliva using 1H NMR showed distinct metabolic disturbances, particularly in pyrimidine metabolism, in diabetic periodontitis. Although these altered metabolites could serve as potential candidates for non-invasive biomarkers, independent validation from a large cohort is still required before they can be translated into clinical diagnostic use.
