Search Results (233)

This article introduces a diffractive neural network-enabled spectral object detection approach (DNN-SOD) to efficiently process massive sky-based multidimensional light field data. DNN-SOD combines the novel exploitation of target spectral features with the intrinsic parallelism of optical computing to process multidimensional information efficiently. DNN-SOD detects targets by segmenting the spectral data cube and processing it with the DNN. The DNN maps spectral intensity to the designated area of the detector, then reconstructs spectral curves, and differentiates targets by comparing them with reference spectral signatures. Classification results from individual sub-spectral data cubes are compiled in sequence, enabling accurate target detection. Simulation results indicate that the architecture achieved an accuracy of 91.56% on the MNIST multi-spectral dataset and 84.27% on the infrared target multi-spectral dataset, validating its feasibility for target detection. This architecture represents an innovative outcome at the intersection of remote sensing and optical computing, significantly advancing the dissemination and practical adoption of optical computing in the field. Full article

Recommender systems are data-driven tools designed to assist or automate users’ decision-making. With the growing demand of personalized sequential recommendations in business intelligence or e-commerce, effectively capturing temporal information from massive user-sequence data has become a crucial challenge. State-of-the-art attention-based models often struggle to balance performance with computational cost, while traditional convolutional neural networks suffer from limited receptive fields and rigid architectures that inadequately model dynamic user interests. To address these limitations, this paper proposes TimeWeaver, a time-aware dual-stream network for sequential recommendation, whose core innovations comprise three key components. First, it employs a re-parameterized large-kernel convolution to expand the effective receptive field. Second, we design a Time-Aware Augmentation mechanism that integrates inter-event time-interval information into positional encodings of items. This allows it to perceive the temporal dynamics of user behavior. Finally, we propose a dual-stream architecture to jointly capture dependencies across different time scales. The context stream employs a modern Temporal Convolutional Network (TCN) structure to strengthen the memorization of users’ medium- and long-term interests. In parallel, the dynamic stream leverages an Exponential Moving Average (EMA) mechanism to weight recent behaviors for sensitively capturing users’ immediate interests. This dual-stream design allows TimeWeaver to comprehensively extract both long- and short-term sequential features. Extensive experiments on three public e-commerce datasets demonstrate TimeWeaver’s superiority. Compared to the strongest baseline model, TimeWeaver achieves average relative improvements of 4.62%, 9.59%, and 4.59% across all metrics on the Beauty, Sports, and Toys datasets, respectively. Full article

Background/Objectives: Interpretation of mixture profiles generated from crime scene samples is an important element in forensic genetics. Here, a workflow for mixture deconvolution of sequenced microhaplotypes (MHs) and STRs using the probabilistic genotyping software MPSproto v0.9.7 was developed, and the performance of the two types of loci was compared. Methods: Sequencing data from a custom panel of 74 MHs (the MH-74 plex) and a commercial kit with 26 autosomal STRs (the ForenSeq™ DNA Signature Prep Kit) were used. Single-source profiles were computationally combined to create 360 two-person and 336 three-person mixtures using the Python script MixtureSimulator v1.0. Additionally, 72 real mixtures typed with the MH-74 plex and 18 real mixtures typed with the ForenSeq Kit from a previous study were deconvoluted using MPSproto. Results: The deconvoluted MH profiles were more complete and had fewer wrong genotype calls than the deconvoluted STR profiles. The contributor proportion estimates were more accurate for MH profiles than for STR profiles. Wrong genotype calls were mostly caused by locus and heterozygous imbalances, noise reads, or an inaccurate contributor proportion estimation. The latter was especially problematic in STR sequencing data, when two contributors contributed equally to the mixture. A total of 34,800 deconvolutions of the simulated mixtures were performed with two defined hypotheses: H_p, “The sample consists of DNA from one/two unknown contributor(s) and the suspect” and H_d, “The sample consists of DNA from two/three unknown individuals”. All true contributors were identified (LR > 10¹⁵ for MHs and LR > 10⁹ for STRs) and all non-contributors excluded (LR < 10⁻⁶ for MHs and LR < 0.2 for STRs). Conclusions: In simulated and real mixtures, the MHs performed better than STRs. Full article

High-grade serous ovarian cancer (HGSOC) is the most common and aggressive subtype of ovarian cancer, accounting for approximately 70% of cases. This study investigates genetic mutations and their associations with overall survival (OS), complete cytoreduction (R0), and platinum response in patients undergoing either primary debulking surgery followed by adjuvant chemotherapy (PDS) or neoadjuvant chemotherapy followed by interval debulking surgery (NACT). Genetic analysis was performed on 43 primary HGSOC tumor samples using targeted massive parallel sequencing via next-generation sequencing (NGS). Clinical and molecular data were evaluated collectively and through subgroup comparisons between PDS and NACT cohorts. All analyzed samples harbored genetic alterations. Univariate survival analysis revealed that the total number of mutations (p = 0.0035), as well as mutations in HRAS (p = 0.044), FLT3 (p = 0.023), TP53 (p = 0.03), and ERBB4 (p = 0.007), were significantly associated with poorer OS. Multivariate Cox regression integrating clinical and molecular data confirmed that ERBB4 mutations are independently associated with adverse outcomes. These findings reveal a distinctive mutational landscape between the PDS and NACT groups and suggest that ERBB4 alterations may define a particularly aggressive tumor phenotype. This study contributes to a deeper understanding of HGSOC biology and may support the development of novel therapeutic targets and personalized treatment strategies in the context of precision oncology. Full article

Massively parallel or next-generation sequencing (NGS) has enabled the genetic characterization of cancer patients, allowing the identification of somatic and germline variants associated with their diagnosis, tumor classification, and therapy response. Despite its benefits, NGS testing is not yet available in the Chilean public health system, rendering it both costly and time-consuming for patients and clinicians. Using a retrospective cohort of 67 formalin-fixed, paraffin-embedded (FFPE) colorectal cancer (CRC) samples, we aimed to implement the identification, annotation, and prioritization of relevant actionable tumor somatic variants in our laboratory, as part of the public health system. We compared two different library preparation methodologies (amplicon-based and capture-based) and different bioinformatics pipelines for sequencing analysis to assess advantages and disadvantages of each one. We obtained 80.5% concordance between actionable variants detected in our analysis and those obtained in the Cancer Genomics Laboratory from the Universidad de Chile (62 out of 77 variants), a validated laboratory for this methodology. Notably, 98.4% (61 out of 62) of variants detected previously by the validated laboratory were also identified in our analysis. Then, comparing the hybridization capture-based library preparation methodology with the amplicon-based strategy, we found ~94% concordance between identified actionable variants across the 15 shared genes, analyzed by the TumorSec^TM bioinformatics pipeline, developed by the Cancer Genomics Laboratory. Our results demonstrate that it is entirely viable to implement an NGS-based analysis of actionable variant identification and prioritization in cancer samples in our laboratory, being part of the Chilean public health system and paving the way to improve the access to such analyses. Considering the economic realities of most Latin American countries, using a small NGS panel, such as TumorSec^TM, focused on relevant variants of the Chilean and Latin American population is a cost-effective approach to extensive global NGS panels. Furthermore, the incorporation of automated bioinformatics analysis in this streamlined assay holds the potential of facilitating the implementation of precision medicine in this geographic region, which aims to greatly support personalized treatment of cancer patients in Chile. Full article

Grain yield establishment is a complex progress and the genetic basis of one of the most important yield components, 100-kernel weight, remains largely unknown. Here, we employed a double haploid (DH) population containing 477 lines which was developed from a cross of two maize elite inbred lines, PHBA6 and Chang7-2, to identify quantitative trait loci (QTL) that related to 100-kernel weight. The phenotypes of the DH population were acquired over three years in two different locations, while the DH lines were genotyped by next-generation sequencing technology of massively parallel 3ʹ end RNA sequencing (MP3RNA-seq). Eventually, 28,874 SNPs from 436 DH lines were preserved after SNP calling and filtering and a genetic map with a length of 837 cM was constructed. Then, single environment QTL analysis was performed using the R/qtl program, and it was found that a total of 17 QTLs related to 100-kernel weight were identified and distributed across the whole genome except chromosomes 5 and 6. The total phenotypic variation explained by QTLs detected in three different environments (BJ2016, BJ2107, and HN2018) was 22.2%, 32.9%, and 51.38%, respectively. Among these QTLs, three of them were identified across different environments as environmentally stable QTLs and explained more than 10% of the phenotypic variance each. Together, the results provided in this study preliminarily revealed the genetic basis of 100-kernel weight and will enhance molecular breeding for key agronomic kernel-related traits in maize. Full article

In this study, 172 Single-Nucleotide Polymorphisms (SNPs) (94 identity-informative SNPs, 56 ancestry-informative SNPs, and 22 phenotypic-informative SNPs) included in the ForenSeq™ DNA Signature Prep kit/DNA Primer Mix B (Verogen) were used for genotyping DNA samples from a population of twenty-one unrelated subjects, native to Northeast Italy. SNP sequencing was performed with the MiSeq FGx™ Forensic Genomics System (Illumina-Verogen), and data were analyzed using the Universal Analysis Software (UAS) v1.2. Raw data underwent further examination with STRait Razor v3 (SRv3) to compare the target SNPs’ genotype calls made with UAS and to identify the presence of microhaplotypes (MHs) due to SNPs associated with the same target SNP’s amplicon. The allele (haplotype) frequencies, Hardy–Weinberg equilibrium, linkage disequilibrium, number of effective alleles (A_e), and relevant forensic statistic parameters were calculated. Among the 172 SNPs evaluated, 45 unique microhaplotypes were found, comprising a novel sequence variant never previously described. The presence of MHs resulted in an 8.00% rise in the typologies of unique sequences, leading to changes in A_e. Notably, for 12 out of the 94 iiSNPs, the values of A_e exceeded 2.00, which is generally associated with a higher expected heterozygosity and increased power of discrimination. Full article

Forensic genetics has experienced remarkable advancements over the past decades, evolving from the analysis of a limited number of DNA segments to comprehensive genome-wide investigations. This progression has significantly improved the ability to establish genetic profiles under diverse conditions and scenarios. Beyond individual identification, forensic genetics now enables the inference of physical traits (e.g., eye, hair, and skin color, as well as body composition), biogeographic ancestry, lifestyle habits such as alcohol and tobacco use, and even the transfer of genital microbiomes post-coitus, among other characteristics. Emerging trends point to a future shaped by the integration of cutting-edge technologies, including CRISPR-Cas systems, artificial intelligence, and machine learning, which promise to further revolutionize the field. This review provides a thorough exploration of forensic genetics, tracing its evolution from its foundational methods (past) to its diverse modern applications (present) and offering insights into its potential future directions. Full article

Noninvasive prenatal paternity testing (NIPPT) is a crucial tool in forensic contexts, particularly in cases involving post-rape pregnancies. It enables judicial authorities and victims to promptly address these situations by determining the paternity of the fetus within a few weeks of pregnancy. NIPPT relies on the analysis of cell-free fetal DNA (cffDNA) found in the maternal bloodstream. However, the abundance of maternal DNA presents a significant challenge in detecting fetal DNA. As a result, research has focused on improving methods for isolating or enriching fetal DNA and, specifically in the context of forensic genetics, on the development of suitable genetic markers. The use of Single Nucleotide Polymorphisms (SNPs) along with novel compound markers or composite multiplexes, has shown promising results. Despite significant advances, partly driven by the increased use of Massive Parallel Sequencing (MPS), challenges remain in validating markers-based NIPPT assays for forensic casework. Further studies are required to enhance the sensitivity of these tests, particularly during the early stages of pregnancy, such as the first trimester. Additionally, improving and standardizing statistical frameworks for result evaluation and interpretation is essential to ensure compatibility with forensic standards. Full article

Microhaplotypes (MHs) are a novel class of genetic markers, exhibiting features that position them as an alternative to STRs and SNPs in addressing challenges commonly encountered in forensic investigations. Additionally, MHs can also offer valuable insights for ancestry inference. However, due to the novelty of MHs, extensive research in different global populations is required before implementation in forensic casework and general research. In this study, individuals from Afghanistan and Somalia were characterized with the Ion AmpliSeq™ MH-74 Plex Research Panel previously developed for forensic genetic purposes. A total of 84 Afghan and 89 Somalian samples were sequenced on the Ion GeneStudio™ S5 System. This led to the identification of 32 and 42 single nucleotide variants in the Afghan and Somalian populations, respectively, that were not included in the former MH definitions. Most of the observed variants were considered to be rare occurrences, being observed one or two times in the dataset. The average values of the effective number of alleles (A_e) were 3.7 for Somalia and 3.6 for Afghanistan—pointing to elevated intrapopulation diversities compared to Europeans. Other parameters (H_o, H_e, PIC, PD, and PE) consistently showed higher average values in the Afghans and Somalis compared to the previously studied populations. PCA and STRUCTURE analyses with 1000 Genomes samples assigned the Somalis to a different cluster than the other sub-Saharan African populations. The analyses also showed higher European and East Asian co-ancestry in the Afghans than in the remaining South Asian populations. The capability of the MH-74 plex to address common kinship problems was evaluated through computational simulations, considering generic thresholds differing by one order of magnitude to assess the FDRs. The median LR > 10¹³ for true siblings when the hypotheses ‘full siblings’ and ‘unrelated individuals’ were compared. As expected, the median LRs were much lower for simulated half-siblings and cousins. This work evaluated the forensic potential of MHs in understudied populations. Overall, the studied panel was versatile and capable of being applied in different forensic applications. Full article

The Permian Lucaogou Formation (PLF) shale oil reservoirs in the Junggar Basin exhibit significant lithological heterogeneity, which limits the understanding of the relationship between macroscopic and microscopic reservoir characteristics, as well as insights into reservoir quality. To address this gap, thirty core samples, exhibiting typical sedimentary features, were selected from a 46 m section of the PLF for sedimentological analysis, thin section examination, high-performance microarea scanning, and scanning electron microscopy. Seven main lithofacies were identified, including massive bedding slitstone/fine-grained sandstone (LS1), cross to parallel bedding siltstone (LS2), climbing ripple laminated argillaceous siltstone (LS3), paired graded bedding argillaceous siltstone (LS4), irregular laminated argillaceous siltstone (LS5), irregular laminated silty mudstone (LM2), and horizontal laminated mudstone (LM2). The paired graded bedding sequences with internal erosion surfaces, massive bedding, and terrestrial plant fragments suggest a lacustrine hyperpycnal flow origin. The channel subfacies of hyperpycnal flow deposits, primarily consisting of LS1 and LS2, reflect strong hydrodynamic conditions, with a single-layer thickness ranging from 1.3 to 3.8 m (averaging 2.2 m) and porosity between 7.8 and 14.2% (averaging 12.5%), representing the primary sweet spot. The lobe subfacies, composed mainly of LS3, LS4, and LS5, reflect relatively strong hydrodynamic conditions, with a single-layer thickness ranging from 0.5 to 1.4 m (averaging 0.8 m) and porosity between 4.2 and 13.8% (averaging 9.6%), representing the secondary sweet spot. In conclusion, strong hydrodynamic conditions and depositional microfacies are key factors in the formation and distribution of sweet spots. The findings of this study are valuable for identifying sweet spots in the PLF and provide useful guidance for the exploration of lacustrine shale oil reservoirs in the context of hyperpycnal flow deposition globally. Full article

One major cause of prenatal mortality and morbidity is congenital abnormalities. Knowing the prevalence and etiology of congenital malformations is essential for analyzing trends and improving neonatal care. Objective: the team aimed to evaluate the utility of whole-exome sequencing (WES) in Romanian prenatal care, highlighting its diagnostic efficacy in comparison to molecular karyotyping, particularly in cases with negative genetic results prior to WES, unfavorable pregnancy outcomes, and consanguinity. Methods: Initially, we identified pregnancies with abnormal ultrasounds unrelated to known syndromes. Subsequently, we performed SNP (single nucleotide polymorphism)-array testing, yielding negative results. We then applied prenatal WES, utilizing Massive Parallel Sequencing on the NovaSeq 6000 platform (average coverage > 100× read length: 2 × 100 bp) with library preparation using the Twist Human Core Exome kit RefSeq & Mitochondrial panel (Twist Bioscience). The bioinformatic analysis involved direct comparison to the human reference sequence (hg38). Results: We achieved a 50% diagnostic rate. After receiving results, two couples chose pregnancy termination, five had uneventful births, and one pregnancy ended in stillbirth. Additionally, we identified three incidental findings that enhanced patient and at-risk member management. This article details ten prenatal cases tested with WES, highlighting its superior diagnostic performance compared to the SNP array. WES detected the genetic diagnostic in 50% of cases that the SNP array did not. We emphasize the advantages of WES in prenatal diagnostics while acknowledging the need for further investigations to comprehensively evaluate its diagnostic utility in the Romanian population. Full article

Background/Objectives: Massively parallel sequencing technologies have advanced chronic lymphocytic leukemia (CLL) diagnostics and precision oncology. Illumina platforms, while offering robust performance, require substantial infrastructure investment and a large number of samples for cost-efficiency. Conversely, third-generation long-read nanopore sequencing from Oxford Nanopore Technologies (ONT) can significantly reduce sequencing costs, making it a valuable tool in resource-limited settings. However, nanopore sequencing faces challenges with lower accuracy and throughput than Illumina platforms, necessitating additional computational strategies. In this paper, we demonstrate that integrating publicly available short-read data with in-house generated ONT data, along with the application of machine learning approaches, enables the characterization of the CLL transcriptome landscape, the identification of clinically relevant molecular subtypes, and the assignment of these subtypes to nanopore-sequenced samples. Methods: Public Illumina RNA sequencing data for 608 CLL samples were obtained from the CLL-Map Portal. CLL transcriptome analysis, gene module identification, and transcriptomic subtype classification were performed using the oposSOM R package for high-dimensional data visualization with self-organizing maps. Eight CLL patients were recruited from the Hematology Center After Prof. R. Yeolyan (Yerevan, Armenia). Sequencing libraries were prepared from blood total RNA using the PCR-cDNA sequencing-barcoding kit (SQK-PCB109) following the manufacturer’s protocol and sequenced on an R9.4.1 flow cell for 24–48 h. Raw reads were converted to TPM values. These data were projected into the SOMs space using the supervised SOMs portrayal (supSOM) approach to predict the SOMs portrait of new samples using support vector machine regression. Results: The CLL transcriptomic landscape reveals disruptions in gene modules (spots) associated with T cell cytotoxicity, B and T cell activation, inflammation, cell cycle, DNA repair, proliferation, and splicing. A specific gene module contained genes associated with poor prognosis in CLL. Accordingly, CLL samples were classified into T-cell cytotoxic, immune, proliferative, splicing, and three mixed types: proliferative–immune, proliferative–splicing, and proliferative–immune–splicing. These transcriptomic subtypes were associated with survival orthogonal to gender and mutation status. Using supervised machine learning approaches, transcriptomic subtypes were assigned to patient samples sequenced with nanopore sequencing. Conclusions: This study demonstrates that the CLL transcriptome landscape can be parsed into functional modules, revealing distinct molecular subtypes based on proliferative and immune activity, with important implications for prognosis and treatment that are orthogonal to other molecular classifications. Additionally, the integration of nanopore sequencing with public datasets and machine learning offers a cost-effective approach to molecular subtyping and prognostic prediction, facilitating more accessible and personalized CLL care. Full article

Background/Objectives: Genetics and epigenetics play an important role in the pathogenesis of cutaneous melanoma. The majority of cases harbor mutations in genes associated with the MAPK signaling pathway, i.e., BRAF, NRAS, or NF1. The remaining neoplasms, often located on acral sites, are condensed as the triple-wildtype subtype and are characterized by other molecular drivers. This study aimed to elucidate genetic and epigenetic differences within cutaneous melanoma and to compare it with melanocytic nevi. Methods: DNA was extracted from archived tissue samples of cutaneous melanoma (n = 19), melanocytic nevi (n = 11), and skin controls (n = 11) and subsequently analyzed by massive parallel (next generation) gene panel sequencing and genome-wide DNA methylation array analysis. The sample size was increased by including repository data from an external study. Results: There were major differences in the genomic landscape of MAPK-altered and triple-wildtype cutaneous melanoma, the latter presenting with a lower number of mutations, a different pattern of copy number variants, and a low frequency of TERT promoter mutations. Dimensional reduction of DNA methylation array analysis clearly separated cutaneous melanoma from melanocytic nevi but revealed no major differences between classical cutaneous melanoma and the triple-wildtype cases. However, it identified a possible biological subgroup characterized by intermediately methylated CpGs. Conclusions: Dimensional reduction of methylation array data is a useful tool for the analysis of melanocytic tumors to differentiate between malignant and benign lesions and may be able to identify biologically distinct subtypes of cutaneous melanoma. Full article

Background: Investigating the way in which individuals are genetically related has been a long-standing application of forensic DNA typing. Whilst capillary electrophoresis (CE)-based STR analysis is likely to provide sufficient data to resolve regularly encountered paternity cases, its power to adequately resolve more distant or complex relationships can be limited. Massively parallel sequencing (MPS) has become a popular alternative method to CE for analysing genetic markers for forensic applications, including kinship analysis. Data workflows used in kinship testing are well-characterised for CE-based methodologies but are much less established for MPS. When incorporating this technology into routine relationship casework, modifications to existing procedures will be required to ensure that the full power of MPS can be utilised whilst maintaining the authenticity of results. Methods: Empirical data generated with MPS for forensically relevant STRs and SNPs and real-world case experience have been used to determine the necessary workflow adaptations. Results: The four considerations highlighted in this work revolve around the distinctive properties of sequence-based data and the need to adapt CE-based data analysis workflows to ensure compatibility with existing kinship software. These considerations can be summarised as the need for a suitable sequence-based allele nomenclature; methods to account for mutational events; appropriate population databases; and procedures for dealing with rare allele frequencies. Additionally, a practical outline of the statistical adjustments required to account for genetic linkage between loci, within the expanded marker sets associated with MPS, has been presented. Conclusions: This article provides a framework for laboratories wishing to implement MPS into routine kinship analysis, with guidance on aspects of the data analysis and statistical interpretation processes. Full article