Due to its heterogeneous morphology and its rarity, anaplastic lymphoma kinase gene-rearranged renal cell carcinoma ( ALK RCC) is a diagnostically challenging entity, often leading to labelling these tumors as RCC, not otherwise classified. This may have clinical and managerial implications, given that patients with ALK oncogene rearrangement may benefit from ALK -inhibitors. Therefore, we attempted to elucidate the clinicopathologic and immunophenotypical characteristics of ALK RCC in a large international cohort. Sixteen multi-institutional tumors were included in the study. Clinical, macroscopic, microscopic, immunohistochemical (IHC), molecular (DNA and RNA sequencing, FISH) and follow-up data were evaluated. There were 9 male and 7 female patients with tumor size ranging from 2 to 12.2cm (mean=7.1cm). All tumors had solid, tan-white with focal cystic changes and gelatinous appearance. Cystic changes and necrosis were seen in 7 and 6 tumors, respectively. Microscopically, a heterogeneous growth pattern was observed including solid (12), tubular (7), papillary (5), tubulocystic (2), pleomorphic epithelioid cells (6), sarcomatoid (2), rhabdoid (4), and intranuclear pseudoinclusions. All tumors were ALK-positive, coexpressing PAX8, KRT7, SDH, FH, and variably CD10, Vimentin, and AMACR. Molecular analysis through next-generation sequencing (NGS) was performed on 14/16 tumors. EML4::ALK (n=5) was the most common gene fusion observed; others included TPM1::ALK (n=4) , TPM3::ALK (n=2), SLIT1::ALK (n=2)and VCL::ALK (n=1). Despite focal TFE3 immunoreactivity in 4/13 cases, the absence of TFE3 gene rearrangement by molecular analysis excludes TFE3 - rearranged RCC as a differential diagnosis. Our study further expands the clinicopathologic, morphologic, and molecular genetic spectrum of ALK -RCC. ALK -RCC can be morphologically heterogeneous and mimic other well-established entities posing a misdiagnosis if appropriate IHC and/or molecular studies are not performed. Accurate diagnosis is of clinical significance as patients with this neoplasm may potentially benefit from ALK- inhibitors, particularly in a metastatic setting. As TFE3 immunoreactivity is not uncommon in ALK -RCC, documentation of ALK gene rearrangement is critical, either by surrogate IHC staining or cytogenetic/molecular analysis is essential.

Decoding azole resistance mechanisms and pathogenicity in Aspergillus section Fumigati through genomic analysis.

Pleural mesothelioma (PM) is a rare and aggressive malignancy. The development of novel therapeutic strategies targeting PM remains an unmet clinical need. However, comprehensive genomic data from Asian populations, particularly from Japanese patients, are limited. This study aimed to elucidate the genomic landscape of PM in Japanese patients using the nationwide Center for Cancer Genomics and Advanced Therapeutics (C-CAT) genomic database. A total of 211 patients registered between June 2019 and March 2025 were analyzed. The most frequent genetic alterations were in BAP1, NF2, TP53, CDKN2A/B, and MTAP. The median tumor mutation burden (TMB) was 1.26, and no microsatellite instability-high patients were detected. The median overall survival (OS) after first-line treatment was 30.6 months. Patients treated with immune checkpoint inhibitors (ICIs) had a significantly better OS than those who did not receive ICIs. In univariate and multivariate analyses, TP53 alterations and high TMB (cutoff value of 1.6) were associated with poor prognosis. These results suggest that integrating clinical and genomic data can enhance prognostic stratification and contribute to the development of precision medicine for PM. This study provides the first large-scale genomic characterization of Japanese PM patients with C-CAT and highlights potential biomarkers for future therapeutic development.

Anaplastic ependymoma (AE) is a rare and aggressive central nervous system tumor that predominantly affects children and remains inadequately characterized at the genomic level. This study aimed to delineate the genomic and demographic landscape of histologically defined AE while identifying potential therapeutic targets. We conducted a retrospective analysis of AE cases from the American Association for Cancer Research (AACR) Project Genomics Evidence Neoplasia Information Exchange (GENIE) repository via cBioPortal, examining recurrent somatic mutations, copy number alterations, mutation co-occurrence, and exploratory sex- and race-based enrichment using descriptive and non-parametric statistics. The most frequent alterations included mutations in the telomerase reverse transcriptase (TERT) promoter, followed by recurrent changes in lysine methyltransferase 2D (KMT2D), lysine methyltransferase 2A (KMT2A), lysine methyltransferase 2C (KMT2C), E1A binding protein p300 (EP300), additional sex combs like 1 (ASXL1), and SET domain containing 2 (SETD2), indicating significant disruption of chromatin remodeling. Recurrent alterations in tumor protein p53 (TP53), ataxia telangiectasia mutated (ATM), and cyclin-dependent kinase inhibitor 2A (CDKN2A) suggested dysregulation of the p53 and DNA damage response pathways. Additionally, alterations in notch receptor 1 (NOTCH1) and notch receptor 2 (NOTCH2) indicated aberrant NOTCH signaling. Neurofibromin 2 (NF2) mutations were observed in male patients, and exploratory subgroup differences emerged across racial groups. Overall, AE appears to be driven by recurrent alterations in chromatin remodeling, p53, DNA damage response, and NOTCH signaling pathways, highlighting these areas as priorities for future biological validation and therapeutic investigation.

Abstract Background Next generation sequencing (NGS) has driven the development of precision oncology. While adult cancer genomics has been extensively studied, pediatric cancer, particularly among Asian populations, has received less attention. Methods We screened gene alteration profiles of 99 pediatric cancer patients using the targeted NGS panel and compared the results with those from Western cohorts. The relationship between genes and clinical characteristics was also analyzed. Results The most frequently mutated gene was KMT2D (18.2%), followed by TP53 (11.1%) and DICER1 (9.1%). The frequency of KMT2D mutations in our cohort is significantly higher than in Western cohorts, whereas TP53 mutations showed no significant difference. Pathogenic or likely pathogenic (P/LP) germline mutations were identified in only 3.5% of the patients. Notably, 69%-75% of patients had at least one genomic alteration with potential clinical significance. In neuroblastoma (NB), the most commonly altered genes were MYCN (7/48), DICER1 (6/48), ARID1B, EGFR, KMT2D, and TCF3 (5/48). In a subset of intermediate-/high-risk NB patients (n = 33), MYCN amplification was associated with disease progression during induction chemotherapy, and gene alterations in the MAPK pathway were associated with poor survival. We also identified novel alterations in specific tumor types, including SFPQ-TACC1 fusion in NB. Conclusions Our findings enhance understanding of the genomic landscape of Chinese pediatric cancer populations. Further research is required to validate these findings and fully explore the potential of genomic data to improve outcomes for pediatric cancer patients.

Genomic landscape of cholangiocarcinoma in India: ethnic variants and implications for targeted therapy.

Natural products (NPs) are a rich source of therapeutic and agricultural compounds. Unfortunately, many promising metabolites are not expressed under the standard laboratory conditions. Deepening our understanding of the regulatory networks governing NP biosynthetic genes is essential to unlocking this hidden chemical diversity. One of the most common regulators of NP expression are TetR-family repressors encoded within NP biosynthetic gene clusters (BGCs). These repressors inhibit transcription of NP biosynthetic genes until interaction with their cognate small-molecule quorum-sensing ligands relieves this repression. While a small number of quorum-sensing cluster-situated regulators (qsCSRs) are known to control NP biosynthesis, the full extent of their regulatory scope has remained unclear due to limited genome-wide data. Using DNA Affinity Purification Sequencing (DAP-seq), we defined the predicted regulons for 84 qsCSR homologues across 78 Streptomyces strains. qsCSRs in this cohort exert influence across multiple cellular processes, with particularly strong impacts on other transcription factors throughout the genome. Approximately 30% of predicted NP BGCs contained qsCSR-regulated genes. In strains encoding multiple qsCSRs, we observed substantial overlap in the BGC regulation. Together, these results greatly expand the genomic landscape of qsCSR activity and provide a foundation for improved bioinformatic strategies to predict and interpret the regulatory control of NP biosynthesis.

Circulating tumor DNA (ctDNA) testing has emerged as a cancer precision medicine approach. We investigated the genomic landscape and clinical utility of ctDNA in patients receiving 223Ra dichloride for bone metastatic castration-resistant prostate cancer (mCRPC). Methods: This prospective, observational, multicenter study enrolled patients treated with 223Ra for bone mCRPC. Targeted sequencing of cell-free DNA from plasma at baseline and end of treatment (EOT), along with paired leukocyte DNA, was performed using an 88-gene panel. Associations between ctDNA profiles and clinical outcomes, including biomarker response, radiographic progression-free survival (rPFS), and overall survival (OS), were analyzed. Results: Of 93 patients analyzed, ctDNA was successfully profiled in 84 baseline and 74 EOT samples, with matched data available for 68 patients. A ctDNA fraction of at least 5%, as well as TP53 alteration, PTEN alteration, and cell cycle pathway alterations at baseline were significantly associated with shorter rPFS and OS. Dynamic changes in ctDNA fraction and PTEN alteration between baseline and EOT correlated with distinct rPFS and OS. Conclusion: This study suggests the clinical utility of ctDNA profiling as both a prognostic and a monitoring tool in patients with bone mCRPC treated with 223Ra. The findings obtained in this study raise the possibility that ctDNA could contribute to future strategies for risk stratification or treatment monitoring during 223Ra therapy.

In clonally propagated crops, extensive divergence between haplotypes complicates transcriptional regulation. However, the contribution of the three dimensional (3D) genome organization to these allelic differences and agronomic traits remains unclear. Here, we generated haplotype-resolved 3D genome landscapes for grapevine cultivars with contrasting berry colors and seed traits, integrating them with genomic, epigenomic, and transcriptomic profiles. We found profound 3D architectural divergence between haplotypes, spanning from large-scale A/B compartments down to a level of topologically associated domain (TAD) boundary variation (18.53%∼23.01%) that approached inter-cultivar differences (18.74~21.62%) (p > 0.05). A key mechanism driving these effects involves large-scale, haplotype-specific transitions between distinct TAD states (Active, Inactive, Heterochromatic), which asymmetrically regulate transcription and alter local DNA methylation patterns. Importantly, these structural rearrangements, including TAD boundary shifts, are strongly associated with underlying structural variants (SVs). Critically, this regulatory cascade impacts key agronomic loci, genes controlling berry color (e.g. VvMYBA) and seedlessness determination (e.g. VvSUS2) were positioned at cultivar-specific TAD boundaries, exhibiting presence/absence variations and differential expression patterns. Our findings support a mechanistic model wherein phased 3D chromatin architecture and heterozygous SVs are strongly associated with the regulation of key agronomic traits, paving the way for accelerating the genetic improvement of clonal crops.

In livestock, understanding the genetic basis of adaptation to the environment is essential for enhancing resilience to climate change and sustaining productivity in diverse environments. Indigenous Ethiopian cattle represent an ideal model for such studies, as they have evolved across a wide range of environments from the cool, oxygen-limited highlands to the hot, pathogen-rich lowlands. These environmental gradients imposed intense selective pressures, shaping their genomic landscape. In this study, we performed the first comprehensive analysis of X-linked adaptive signatures in Ethiopian indigenous cattle using whole-genome sequencing data. Population structure analysis revealed clear genetic differentiations between Abigar and Barka cattle, while the remaining populations showed substantial shared ancestry and admixtures. Pairwise fixation index ([Formula: see text] estimates, runs of homozygosity (ROH) patterns, and linkage disequilibrium (LD) decay further supported historical isolation and stronger selection pressure in Barka, contrasting with the greater diversity and faster LD decay in Gojjam Highland cattle. Complementary selection signature detections ([Formula: see text], XP-EHH, and[Formula: see text]) revealed population-specific and shared genomic regions under selection on the X chromosome. Notably, signals associated with high-altitude adaptation were detected near the RBM3, RPS4X, and TSC22D3 loci. Additional signals were observed in genes related to thermoregulation and oxidative stress response (EDA, SUV39H1, and HDAC8), as well as immune regulation (IRAK1, BDA20, and IL1RAPL1), suggesting adaptation to hot and pathogen-rich environments. Functional enrichment analysis highlighted genes involved in extracellular matrix organization and immune signaling pathways, underscoring their roles in environmental adaptation. This study provides the first genome-wide evidence of X-linked adaptive divergence in the Ethiopian cattle. The findings highlight the contribution of the X chromosome to heat tolerance, hypoxia adaptation, and immune resilience, offering valuable genomic insights for breeding programs aimed at enhancing productivity and climate adaptability in tropical cattle.

The isogenic, tumor-bearing Dark Agouti Mammary Adenocarcinoma (DAMA) model is commonly utilized in breast cancer research to evaluate mechanisms of, and interventions for chemotherapy-induced toxicity and tumor response. However, the biological subtype, immune landscape, and genomic profile of DAMA have remained uncharacterized. In this study, we comprehensively evaluated DAMA tumors naïve and exposed to methotrexate, assessing their histopathological, molecular, and genomic features. Our findings reveal that the DAMA model presents a platform to model the human triple-negative breast cancer (TNBC), as it exhibits substantial macrophage infiltration, the tumors display dysregulation of oncogenes Bcl2, Egfr, and potentially Myc, suggesting apoptotic resistance as a key mechanism driving growth. These characteristics position the DAMA model as a potential preclinical model for investigating TNBC biology, therapeutic responses, and drug toxicity.

Whole-genome and whole-transcriptome sequencing enable biologically and clinically meaningful T-ALL subtype classification with greater precision than immunophenotype-based approaches. The genomic landscape of T-ALL is heavily influenced by noncoding alterations, posing a challenge for clinical applications, as their detection often requires whole-genome sequencing. Studies have identified genomic subtypes and genetic alterations that improve the accuracy of patient outcome prediction.

India harbours a diverse range of indigenous goat breeds that have adapted to varied climatic zones over centuries. This study investigated the genomic basis of local adaptation in these populations (n= 11) divided into seven agro-climatic zones using genome-wide SNP data and century-scale environmental variables. A total of 2,295,833 SNPs and 15 non-collinear bioclimatic predictors were analyzed using the landscape genomics tool R SamBada for genotype-environment association. Models were selected based on G-score and q-value thresholds (q < 0.01). Several loci showed strong signatures of selection, with associated genes enriched in key adaptive pathways, including HIF-1 signalling, insulin signalling, and toll-like receptor pathways. Many key genes and pathways were identified with both direct and indirect roles in adaptation to specific agro-climatic zone. Only 9 SNP variants showed SIFT score < 0.05 (deleterious) out of which, only 2 variants each harbouring gene PTPRC and PLCB1 were predicted to be deleterious with high confidence. Further downstream technical validation for functionality was done using PTPRC and PLCB1 present in coding region and exhibited significant environmental associations. Missense mutations in these genes were further characterized using I-Mutant, ConSurf, and Phyre2. The PTPRC variant was predicted to reduce protein stability within a moderately conserved immune domain, and structural modelling indicated altered folding in mutant proteins. These adaptive variants likely contribute to resilience against heat, humidity, and pathogen-driven stress. This integrative landscape genomics approach reveals how natural selection and environmental pressures have shaped the adaptive genome of Indian indigenous goats and provides a foundation for marker-assisted selection to enhance climate resilience in future breeding programs. This study represents the first landscape genomics analysis in indigenous goat populations of India.

The human T-cell leukemia virus type 1 (HTLV-1) is a neglected oncogenic retrovirus responsible for adult T-cell leukemia/lymphoma (ATLL) and autoimmune diseases that disproportionately affects marginalized populations worldwide. Peru reports the highest global ATLL incidence, yet comprehensive genomic studies remain limited. We aimed to characterize the virological, immunological, and host genetic landscape of HTLV-1 infection to identify population-specific high-risk features. We conducted a prospective cohort study (2017-2018) of 67 HTLV-1-infected individuals from Lima, Peru, using an integrated high-throughput genomic approach. This included whole-genome HTLV-1 sequencing, targeted ultra-deep sequencing of 280 hematological malignancy-associated genes, high-resolution HLA typing, GATK-based variant calling, and comprehensive clinical follow-up over 7 years. Phylogenomic analyses were performed using maximum likelihood and Bayesian approaches. The cohort exhibited exceptionally high-risk characteristics with a median proviral load (PVL) of 4.5 ± 3.8, and 13.4% cumulative crude mortality over 7 years. Phylogenomic analysis revealed 96.8% of isolates belonged to the Transcontinental subtype, with most clustering in a newly identified Andean-Amazonian subgroup. HLA-I analysis demonstrated unique population-specific allele distributions with significantly reduced evolutionary divergence compared to Japanese cohorts (HLA-HED: 5.34 vs. 6.87, p = 0.0002), potentially differences in mechanisms of viral immune control. Ultra-deep sequencing identified early clonal hematopoiesis with prevalent mutations in cancer-associated genes including KMT2D (55%), NOTCH1 (49%), and TP53 (27%). Mutation burden correlated significantly with proviral load (r = 0.34, p = 0.003), and longitudinal analysis revealed progressive genomic instability with more than two-fold increase in mutations over 3 years (7.25 vs. 19.5 mutations per patient, p = 0.01). High PVL (> 4%) was the only independent predictor of crude mortality (OR: 1.07; 95% CI: 1.01-1.15; p = 0.033). Contrary to previous reports, Strongyloides coinfection was not associated with disease progression. This first comprehensive genomic characterization of HTLV-1 in South America reveals population-specific viral evolution, reduced HLA diversity, and evidence of early oncogenic transformation events. The exceptionally high proviral loads and unique mutational landscape provide novel insights into HTLV-1 pathogenesis and support the development of population-tailored risk stratification approaches. These findings emphasize the urgent need for expanded genomic surveillance and targeted interventions in underrepresented populations bearing disproportionate HTLV-1 burden.

Pediatric myelodysplastic syndromes (MDS) differ from adult MDS, and their genetic basis is poorly understood. This study characterizes the genomic features and clinical outcomes of advanced pediatric MDS. In this retrospective study, next-generation sequencing was performed on 63 pediatric patients with advanced MDS, including 46 with MDS with excess blasts (MDS-EB) and 17 with MDS-EB in transformation (MDS-EB-T, 2016 WHO classification), to detect somatic and potential germline variants. Mutations in 53 genes were detected in 49 patients (77.8%), with a median of 2 variants per patient (range, 0-8). Alterations in the RAS/MAPK pathway were most common, occurring in 28 patients (44.4%). Monosomy 7 significantly co-occurred with SETBP1, ETV6, and GATA2 mutations (p < 0.01). Time-dependent analyses showed HSCT improved 2-year overall survival (OS, 42.6% vs. 84.5%, p = 0.003) and event-free survival (EFS, 37.3% vs. 58.2%, p = 0.011). In multivariate analyses, HSCT was strongly associated with improved OS and EFS (p < 0.01). PTPN11 mutation remained an independent predictor of poorer OS and EFS (p < 0.05), and MDS-EB-T subtype independently predicted inferior EFS (p < 0.05). In advanced pediatric MDS, HSCT was associated with improved survival, whereas PTPN11 mutations emerged as an adverse prognostic factor.

Anti-PD-1 therapies improve survival in recurrent/metastatic (R/M) squamous cell carcinoma of the head and neck (SCCHN), but only a minority of patients achieve durable responses. The mechanisms driving resistance to anti-PD-1 in SCCHN remain poorly understood. Using the IMMUcan multiomics workflow, we characterized the molecular and immune profiles of R/M SCCHN progressing on anti-PD-1 treatment and compared them with an anti-PD-1-naïve cohort. Tumor biopsies from patients with anti-PD-1-resistant SCCHN exhibited significantly more EGFR and MYCL amplifications, along with increased MYC pathway alterations. Transcriptomic and proteomic analyses revealed that anti-PD-1-secondary resistant SCCHN had increased CD8+ T-cell infiltration with higher levels of immune exhaustion markers than primary resistant and naïve SCCHN. Additionally, high beta-2-microglobulin (B2M) expression correlated with greater T-cell infiltration and improved survival following anti-PD-1 therapy. Tumor cell B2M expression was independent of TMB and PD-1L expression, suggesting that B2M expression could serve as an additional biomarker for anti-PD-1 response.

This study presents the comparative analysis of 26 multidrug-resistant (MDR) C. striatum strains isolated in Brazil. Additional genomes from international sources were incorporated. The analyses encompassed in vitro antimicrobial susceptibility testing and an in silico workflow for genomic similarity comparison, phylogenetic reconstruction, genomic clustering, pangenome analysis, mobilome content, virulence prediction, and functional annotation of unique proteins and putative virulence clusters. Strong in silico evidence of clonality among several Brazilian isolates was obtained at the same time that some strains consistently indicated a divergent genomic profile. There are 196 unique coding sequences (CDSs) across the Brazilian IHPs. Of particular interest, strain IHP2030 carried an exclusive fimbria, sharing less than 50% similarity with other fimbriae in the dataset. Yet, structural predictions suggested conservation of key structural domains typically associated with fimbrial proteins. Mobilome content analysis revealed that IHPs strains were overall similar, differing primarily in the number of insertion sequences and in the presence or absence of CRISPR-Cas defense systems. Regarding virulence, an exclusive cluster in IHP2050 and IHP2060 suggests adaptive advantages associated with their respective environments of isolation. This study reveals a complex genomic landscape among Brazilian MDR C. striatum strains, marked by clonal dissemination alongside strain-level genetic variation in accessory genomes, mobilome composition, and virulence-associated gene repertoires, providing genomic evidence of diversification within hospital-associated lineages.

The Enterobacter cloacae complex (ECC) is an emerging global cause of healthcare-associated infections, marked by high mortality and limited treatment options due to extensive antimicrobial resistance. Cefiderocol is increasingly used against carbapenem-resistant pathogens, including metallo-β-lactamase producers. However, reports of treatment failure, particularly in isolates with borderline MICs, indicate uncertainty regarding the relationship between phenotypic susceptibility and underlying genomic resistance mechanisms. The objective of this study was to characterise the genomic resistance landscape of a cefiderocol-susceptible Enterobacter hormaechei isolate and to examine potential genomic-phenotypic discordance relevant to cefiderocol activity. We conducted genomic and phenotypic characterisation of a clinical Enterobacter hormaechei subsp. hoffmannii isolate recovered at Liverpool Clinical Laboratories (Royal Liverpool Hospital, United Kingdom). The isolate exhibited extensive resistance to β-lactams, carbapenems, aztreonam, most aminoglycosides, and multiple ancillary agents, yet remained phenotypically susceptible to cefiderocol (MIC = 2 mg/L; EUCAST susceptible breakpoint). Whole-genome sequencing revealed a multifactorial resistance architecture, including blaNDM-1 and widespread mutations across siderophore-mediated iron uptake pathways (tonB, fhu, fep, ent, wzz). Variants in global regulators (fur, ampR) suggested altered iron-acquisition and β-lactamase (blaACT-associated) regulatory networks. Multiple mutations affecting β-lactam targets, including ftsI (PBP3) indicated reduced PBP affinity. Together, these determinants form a complex genomic resistance landscape. This study highlights a genomic-phenotypic discordance in an E. hormaechei isolate with borderline cefiderocol susceptibility. The findings underscore the risk of overestimating cefiderocol activity based solely on MIC values and emphasise the need for genomic resistance assessment in high-risk Enterobacterales.

Genetic differentiation among populations often varies significantly across the genome due to factors such as selection and recombination, resulting in a heterogeneous genomic landscape. However, variation in low-differentiation regions-genomic valleys-remains poorly understood. Moreover, most insights into plant genomic landscapes come from flowering plants, while comparable genome-wide studies in other taxa, such as conifers, remain limited. We analyzed whole-genome sequencing data from 100 individuals of three pine species-Pinus banksiana, Pinus contorta, and Pinus nigra. We found substantial genome-wide variation in recombination rates, with intergenic regions exhibiting higher recombination than genic regions, and rates decreasing with increasing distance from genes. Recombination rate was negatively correlated with gene length, driven primarily by intron length, suggesting that long introns in conifers may promote the retention of exceptionally long genes by maintaining low recombination in these regions. Genomic scans further revealed that genomic valleys are maintained through either balancing, background, or parallel selection. Additionally, multiple forms of selection were strongly associated with local recombination rate variation, highlighting the significant role of recombination in shaping patterns of genomic differentiation. Our findings provide new insight into the evolution and maintenance of extremely long genes in conifers. Moreover, the results indicate that allopatric selection in regions of low recombination is a major force structuring genomic variation in these species.

Identifying genomic regions shaped by natural selection is a central goal in evolutionary genomics. Existing machine learning methods for this task are typically trained on labeled data simulated according to specific evolutionary scenarios. While effective in controlled settings, these models are limited by their reliance on explicit class labels, detecting only the processes they were trained to recognize. This limitation makes it difficult to interpret predictions for regions shaped by other evolutionary forces, a problem especially acute when analyzing genomes influenced by mixtures of adaptive and demographic factors. One-vs-rest strategies offer a potential alternative but suffer from the complexity of modeling processes as a catch-all "rest" class. Here, we explore positive-unlabeled learning as a flexible framework for detecting adaptive events. This semi-supervised approach permits identification of a target class using only positive labels and an unlabeled background, without requiring explicit modeling of negatives. To assess its utility, we focus on a binary classification setting for detecting selective sweeps against a mixed background of unlabeled sweeps and neutrally evolving regions. We introduce PULSe, a method that trains only on labeled sweep observations while treating remaining data as unlabeled. By avoiding assumptions about background composition, PULSe enables robust sweep discovery in realistic genomic landscapes. We evaluate performance across demographic, adaptive, and confounding contexts, including domain shift from misspecified models, and find that PULSe delivers strong generalizability. Finally, analyzing European and Bengali genomes, we recapitulate known sweep candidates, demonstrating PULSe as a versatile tool for detecting adaptive regions across diverse genomic landscapes.