BackgroundPIK3CA-related overgrowth spectrum (PROS) is a group of disorders caused by somatic mosaic mutations in the PIK3CA gene, characterized by tissue overgrowth affecting multiple organs, including the skin, adipose tissue, blood vessels, bones, and nervous system. Although constitutive hyperactivation of PI3K-AKT-mTOR signaling in PROS promotes abnormal tissue development, remodeling, and segmental overgrowth, the downstream signaling networks are highly complex, and the precise molecular mechanisms underlying tissue overgrowth remain incompletely understood.MethodsIn this exploratory study, we investigated molecular alterations associated with overgrowth in PROS by profiling mRNA and microRNA expression in fibroblasts derived from affected skin (pv-Fibro) and unaffected skin (non-pv-Fibro) from the same patient. Additionally, fibroblasts from three healthy individuals were pooled and analyzed as a single control sample.ResultsWe identified 24,034 human mRNAs using microarray-based gene expression profiling. Ten mRNAs were selected as candidate lesion-associated upregulated genes in pv-Fibro (≥5-fold vs. nc-Fibro and <2-fold change in non-pv-Fibro vs. nc-Fibro). In contrast, three mRNAs were selected as candidate lesion-associated downregulated genes in pv-Fibro (≤0.2-fold vs. nc-Fibro and >0.5-fold in non-pv-Fibro vs. nc-Fibro). For miRNA profiling, 4,603 human microRNAs were analyzed. hsa-miR-146a-5p was markedly upregulated in pv-Fibro (≥5-fold vs. nc-Fibro and <2-fold change in non-pv-Fibro vs. nc-Fibro), while hsa-miR-146b-5p was markedly downregulated (≤0.2-fold vs. nc-Fibro and >0.5-fold in non-pv-Fibro vs. nc-Fibro).DiscussionThe observed changes in mRNA and miRNA expression represent candidate molecular alterations associated with inflammatory and tissue remodeling pathways in PROS pathology.
Pathogenic and likely pathogenic (P/LP) variants in ClinVar guide variant interpretation in clinical genetic testing, yet their population-specific frequency differences have not been systematically examined in epilepsy. We compiled 199 epilepsy-associated genes, extracted 46,746 unique P/LP variants from ClinVar, and retrieved allele frequencies from gnomAD v4.1.1 gene-level tables using nucleotide-resolved matching. Fisher’s exact test with Benjamini-Hochberg correction (q < 0.05) identified nine variants significantly enriched in East Asian (EAS) relative to non-Finnish European (NFE) populations, with odds ratios of 36.3–867.9. Eight additional low-frequency EAS signals met the same enrichment criterion with at least two EAS alleles and an uncorrected p < 0.01 and are reported as exploratory. Thirteen of the seventeen variants reside in seven autosomal recessive genes, namely, ALDH7A1, ST3GAL5, GOSR2, AARS1, BRAT1, WWOX, and UBA5. In parallel, an rsID-level workflow, the standard approach in database mining, yielded 60 candidate EAS-enriched variants. Nucleotide-resolved matching confirmed only fifteen of these and additionally recovered two signals the rsID workflow missed, a false-positive rate of 75.0%. These findings show that epilepsy P/LP variant frequencies differ substantially between EAS and NFE populations and concentrate in recessive genes with direct relevance to carrier screening. rsID-level ClinVar-gnomAD integration systematically overestimates population specificity, and nucleotide-resolved allele matching should become a standard step in population frequency analyses.
BackgroundSports-genomics research in Turkish athlete populations has examined diverse sports, genetic approaches, and sport-related phenotypes. This systematic review mapped the scope, study populations, and methodological characteristics of this literature.MethodsThis systematic review followed PRISMA 2020. Web of Science Core Collection, Scopus, PubMed/MEDLINE, SPORTDiscus with Full Text, Google Scholar, TRDizin, and DergiPark were searched for records published from 1 January 2019 to 1 April 2026. Original human studies were eligible when they included an identifiable Turkish athlete population and investigated inherited genetic or genomic variation in relation to a sport-related phenotype. Eligibility was determined by the characteristics of the study population and the availability of separately extractable data for Turkish athletes. Records were deduplicated and independently screened. Reports from confirmed shared cohorts were retained when they contributed non-overlapping eligible information and were linked during synthesis.ResultsThe database search yielded 1,740 records. After removal of 717 duplicates, 1,023 records underwent title and abstract screening and 396 full texts were assessed. Sixty-one database-derived reports and nine reports identified through backward reference-list checking and forward citation searching were included, yielding 70 reports from 58 operationally unique study populations. These populations comprised 3,201 athletes and 3,683 non-athlete controls. Eighteen reports originated from six confirmed shared-cohort clusters.ConclusionThis review provides a transparent evidence base for a cautious descriptive synthesis of Turkish sports-genomics research. The evidence is reported as a structured account of the study populations, genetic approaches, and phenotypic themes represented in the literature.Systematic Review Registrationhttps://doi.org/10.17605/OSF.IO/739CG.
BackgroundImmune checkpoint inhibitor responses in cutaneous melanoma are heterogeneous. We investigated immune checkpoint gene (ICG)-based tumor phenotypes, their immune and survival associations, treatment-associated transcriptional changes, and interferon-γ (IFN-γ)-responsive gene expression in melanoma cell lines.MethodsWe clustered 47 immune checkpoint genes in 472 TCGA-SKCM tumors and assessed survival using one tumor per patient. Cox models adjusted for AJCC stage, age and sex, with a stage-stratified sensitivity model. GSE91061 comprised 109 RNA-seq samples from 65 patients, including 43 paired pretreatment/on-treatment cases. We reanalyzed all 16,291 GSE120575 immune cells using the deposited log2(TPM+1) values, aggregated expression and composition within 48 biopsies from 32 patients, and accounted for repeated patients, time point and therapy. Canonical marker plots and full gene-set definitions were supplied. Cell-line RT-qPCR assessed CD274, CXCL10 and HLA-A expression after recombinant human IFN-γ treatment at 10 or 100 ng/mL for 24 or 48 h (PeproTech), with four independent biological replicates measured in technical triplicate.ResultsICG-high, ICG-mid and ICG-low phenotypes included 136, 203 and 133 tumors. The ICG score correlated with the CD8 T-cell signature (Spearman ρ = 0.828, P = 5.06 × 10-120). In 454 evaluable unique patients, median overall survival was 4930, 2192, 1487 days (log-rank P = 1.50e-05). Among 404 complete cases with 190 deaths, ICG-high versus ICG-low was associated with survival after adjustment (HR 0.34, 95% CI 0.22–0.51; P = 2.68e-07), with a concordant stage-stratified estimate (HR 0.35, 95% CI 0.23–0.53; P = 6.40e-07). Paired on-treatment ICG scores increased (median change 0.115; P = 0.012). At the biopsy level, immune-cell IFN-α and IFN-γ scores were higher in nonresponders after adjustment for time point and therapy (P = 1.29e-04 and P = 0.004, respectively).ConclusionICG phenotypes were associated with immune-expression profiles and overall survival after adjustment for AJCC stage, age and sex. Paired bulk transcriptomes showed increased IFN-related expression during treatment, whereas the single-cell cohort showed higher immune-cell IFN scores in nonresponders. Together, these findings describe the relationship of checkpoint and IFN programs to immune context, treatment time and clinical outcome.
High-throughput genotyping supports marker-assisted selection of known loci and genomic prediction of complex traits. Yet, routine adoption of these molecular breeding tools remains out of reach for many breeding programs. Here, we present a multi-species single-nucleotide polymorphism (SNP) array designed for standardized genotyping of wheat, barley, and oat. The platform contains 9,468 small-grain SNPs (3,421 wheat, 3,058 barley, and 2,989 oat), strategically selected to provide genome-wide coverage using genetic selection bins of 1 cM for wheat and oat and 0.5 cM for barley. Validation across diverse germplasm collections (>10,000 samples) revealed robust technical performance, with 98.7%, 89.0%, and 91.5% of SNPs in barley, wheat, and oat, respectively, meeting all quality thresholds. A useful feature of the array is the ability to genotype samples in multi-species hybridization mode, thereby lowering the array-processing cost per sample. Multi-species hybridization produced partner-specific signal responses while retaining high genotype-call concordance, ranging from 93.5% to 98.7% across the evaluated dual- and tri-hybridization configurations. Oat signal quality was highest under tri-hybridization, although its performance was only marginally greater than in the oat-wheat dual configuration. Independent applications demonstrated that the array produced genomic prediction accuracies comparable to those obtained using higher-density arrays or sequence-based genotyping platforms and supported the detection of marker-trait associations in genome-wide association studies. By combining species-specific marker sets for wheat, barley, and oat on one platform, the array provides a standardized genomic resource for routine molecular breeding.
