Explore the vast range of titles available.

Ancestral remains should be regarded not as “artifacts” but as human relatives who deserve respect Recent scientific developments have drawn renewed attention to the complex relationships among Indigenous peoples, the scientific community, settler colonial governments, and ancient human remains ( 1 , 2 ). Increasingly, DNA testing of ancestral remains uncovered in the America s is being used in disputes over these remains ( 3 ). However, articulations of ethical principles and practices in paleogenomics have not kept pace ( 4 ), even as results of these studies can have negative consequences, undermining or complicating community claims in treaty, repatriation, territorial, or other legal cases. Paleogenomic narratives may also misconstrue or contradict community histories, potentially harming community or individual identities. Paleogenomic data can reveal information about descendant communities that may be stigmatizing, such as genetic susceptibilities to disease. Given the potential consequences for Indigenous communities, it is critical that paleogenomic researchers consider their ethical obligations more carefully than in the past.

The concentration of atmospheric methane (CH₄) continues to increase with microbial communities controlling soil–atmosphere fluxes. While there is substantial knowledge of the diversity and function of prokaryotes regulating CH₄ production and consumption, their active interactions with viruses in soil have not been identified. Metagenomic sequencing of soil microbial communities enables identification of linkages between viruses and hosts. However, this does not determine if these represent current or historical interactions nor whether a virus or host are active. In this study, we identified active interactions between individual host and virus populations in situ by following the transfer of assimilated carbon. Using DNA stable-isotope probing combined with metagenomic analyses, we characterized CH₄-fueled microbial networks in acidic and neutral pH soils, specifically primary and secondary utilizers, together with the recent transfer of CH₄-derived carbon to viruses. A total of 63% of viral contigs from replicated soil incubations contained homologs of genes present in known methylotrophic bacteria. Genomic sequences of 13C-enriched viruses were represented in over one-third of spacers in CRISPR arrays of multiple closely related Methylocystis populations and revealed differences in their history of viral interaction. Viruses infecting nonmethanotrophic methylotrophs and heterotrophic predatory bacteria were also identified through the analysis of shared homologous genes, demonstrating that carbon is transferred to a diverse range of viruses associated with CH₄-fueled microbial food networks.

Interferons (IFNs) are key controllers of viral replication, with intact IFN responses suppressing virus growth and spread. Using the murine norovirus (MNoV) system, we show that IFNs exert selective pressure to limit the pathogenic evolutionary potential of this enteric virus. In animals lacking type I IFN signaling, the nonlethal MNoV strain CR6 rapidly acquired enhanced virulence via conversion of a single nucleotide. This nucleotide change resulted in amino acid substitution F514I in the viral capsid, which led to >10,000-fold higher replication in systemic organs including the brain. Pathogenicity was mediated by enhanced recruitment and infection of intestinal myeloid cells and increased extraintestinal dissemination of virus. Interestingly, the trade-off for this mutation was reduced fitness in an IFN-competent host, in which CR6 bearing F514I exhibited decreased intestinal replication and shedding. In an immunodeficient context, a spontaneous amino acid change can thus convert a relatively avirulent viral strain into a lethal pathogen.

Homologous recombination repair (HRR) is a cellular pathway for high-fidelity double strand DNA break repair that uses the sister chromatid as a guide to ensure chromosomal integrity and cell viability. Deficiency in the HRR pathway (HRD) can sensitize tumors to poly (ADP-ribose) polymerase inhibitors (PARPi) and platinum-based chemotherapy, offering an avenue to identify patients who may benefit from targeted therapies. HRD signature (HRDsig) is a pan-solid-tumor biomarker on the FoundationOne®CDx (F1CDx®) assay that employs a DNA scar-based approach to calculate a score based on copy number features (e.g., segment size, oscillation patterns, and breakpoints per chromosome arm) and does not rely on HRR gene alterations, enabling detection of genomic and epigenetic mechanisms of HRD. After finalizing the HRDsig algorithm, analytical validation was conducted in a CAP-accredited, CLIA-certified laboratory on 278 solid tumor and normal tissue specimens. HRDsig results were compared with an independent HRD biomarker, defined by the presence of a reversion mutation restoring HRR gene function. In this evaluation, 100 HRD-positive and 126 HRD-negative samples showed a positive percent agreement of 90.00% and a negative percent agreement of 94.44%. The limit of detection (LoD) was estimated at 23.04% tumor purity, with the limit of blank (LoB) confirmed as zero in 60 normal tissue replicates. Reproducibility testing on 11 positive and 11 negative samples across multiple labs, reagent lots, and sequencers yielded agreement in 99.49% of positive and 99.73% of negative replicates. HRDsig status remained consistent in the presence of interfering substances, demonstrating 100% concordance in spiked samples. These validation results underscore the high analytical concordance, low false-positive rate, and overall robustness of HRDsig for reliable assessment of homologous recombination deficiency.

Deep neural networks have been widely adopted in numerous domains due to their high performance and accessibility to developers and application-specific end-users. Fundamental to image-based applications is the development of Convolutional Neural Networks (CNNs), which possess the ability to automatically extract features from data. However, comprehending these complex models and their learned representations, which typically comprise millions of parameters and numerous layers, remains a challenge for both developers and end-users. This challenge arises due to the absence of interpretable and transparent tools to make sense of black-box models. There exists a growing body of Explainable Artificial Intelligence (XAI) literature, including a collection of methods denoted Class Activation Maps (CAMs), that seek to demystify what representations the model learns from the data, how it informs a given prediction, and why it, at times, performs poorly in certain tasks. We propose a novel XAI visualization method denoted CAManim that seeks to simultaneously broaden and focus end-user understanding of CNN predictions by animating the CAM-based network activation maps through all layers, effectively depicting from end-to-end how a model progressively arrives at the final layer activation. Herein, we demonstrate that CAManim works with any CAM-based method and various CNN architectures. Beyond qualitative model assessments, we additionally propose a novel quantitative assessment that expands upon the Remove and Debias (ROAD) metric, pairing the qualitative end-to-end network visual explanations assessment with our novel quantitative “yellow brick ROAD” assessment (ybROAD). This builds upon prior research to address the increasing demand for interpretable, robust, and transparent model assessment methodology, ultimately improving an end-user’s trust in a given model’s predictions. Examples and source code can be found at: https://omni-ml.github.io/pytorch-grad-cam-anim/ .

Early diagnosis and access to resources, support and therapy are critical for improving long-term outcomes for children with autism spectrum disorder (ASD). ASD is typically detected using a case-finding approach based on symptoms and family history, resulting in many delayed or missed diagnoses. While population-based screening would be ideal for early identification, available screening tools have limited accuracy. This study aims to determine whether machine learning models applied to health administrative and birth registry data can identify young children (aged 18 months to 5 years) who are at increased likelihood of developing ASD. We assembled the study cohort using individually linked maternal-newborn data from the Better Outcomes Registry and Network (BORN) Ontario database. The cohort included all live births in Ontario, Canada between April 1st, 2006, and March 31st, 2018, linked to datasets from Newborn Screening Ontario (NSO), Prenatal Screening Ontario (PSO), and Canadian Institute for Health Information (CIHI) (Discharge Abstract Database (DAD) and National Ambulatory Care Reporting System (NACRS)). The NSO and PSO datasets provided screening biomarker values and outcomes, while DAD and NACRS contained diagnosis codes and intervention codes for mothers and offspring. Extreme Gradient Boosting models and large-scale ensembled Transformer deep learning models were developed to predict ASD diagnosis between 18 and 60 months of age. Leveraging explainable artificial intelligence methods, we determined the impactful factors that contribute to increased likelihood of ASD at both an individual- and population-level. The final study cohort included 707,274 mother-offspring pairs, with 10,956 identified cases of ASD. The best-performing ensemble of Transformer models achieved an area under the receiver operating characteristic curve of 69.6% for predicting ASD diagnosis, a sensitivity of 70.9%, a specificity of 56.9%. We determine that our model can be used to identify an enriched pool of children with the greatest likelihood of developing ASD, demonstrating the feasibility of this approach.This study highlights the feasibility of employing machine learning models and routinely collected health data to systematically identify young children at high likelihood of developing ASD. Ensemble transformer models applied to health administrative and birth registry data offer a promising avenue for universal ASD screening. Such early detection enables targeted and formal assessment for timely diagnosis and early access to resources, support, or therapy.