Extending DeepTrio for sensitive detection of complex de novo mutation patterns.

De novo mutations (DNMs), and among them clustered DNMs within 20 bp of each other (cDNMs) are known to be a potential cause of genetic disorders. However, identifying DNM in whole genome sequencing (WGS) data is a process that often suffers from low specificity. We propose a deep learning framework for DNM and cDNM detection in WGS data based on Google's DeepTrio software for variant calling, which considers regions of 110 bp up- and downstream from possible variants to take information from the surrounding region into account. We trained a model each for the DNM and cDNM detection tasks and tested it on data generated on the HiSeq and NovaSeq platforms. In total, the model was trained on 82 WGS trios generated on the NovaSeq and 16 on the HiSeq. For the DNM detection task, our model achieves a sensitivity of 95.7% and a precision of 89.6%. The extended model adds confidence information for cDNMs, in addition to standard variant classes and DNMs. While this causes a slight drop in DNM sensitivity (91.96%) and precision (90.5%), on HG002 cDNMs can be isolated from other variant classes in all cases (5 out of 5) with a precision of 76.9%. Since the model emits confidence probabilities for each variant class, it is possible to fine-tune cutoff thresholds to allow users to select a desired trade-off between sensitivity and specificity. These results show that DeepTrio can be retrained to identify complex mutational signatures with only little modification effort.

The Bjøntegaard Bible Why Your Way of Comparing Video Codecs May Be Wrong.

In this paper, we provide an in-depth assessment on the Bjøntegaard Delta. We construct a large data set of video compression performance comparisons using a diverse set of metrics including PSNR, VMAF, bitrate, and processing energies. These metrics are evaluated for visual data types such as classic perspective video, 360° video, point clouds, and screen content. As compression technology, we consider multiple hybrid video codecs as well as state-of-the-art neural network based compression methods. Using additional supporting points in-between standard points defined by parameters such as the quantization parameter, we assess the interpolation error of the Bjøntegaard-Delta (BD) calculus and its impact on the final BD value. From the analysis, we find that the BD calculus is most accurate in the standard application of rate-distortion comparisons with mean errors below 0.5 percentage points. For other applications and special cases, e.g., VMAF quality, energy considerations, or inter-codec comparisons, the errors are higher (up to 5 percentage points), but can be halved by using a higher number of supporting points. We finally come up with recommendations on how to use the BD calculus such that the validity of the resulting BD-values is maximized. Main recommendations are as follows: First, relative curve differences should be plotted and analyzed. Second, the logarithmic domain should be used for saturating metrics such as SSIM and VMAF. Third, BD values below a certain threshold indicated by the subset error should not be used to draw recommendations. Fourth, using two supporting points is sufficient to obtain rough performance estimates.

Mobile Apps Aimed at Preventing and Handling Unintentional Injuries in Children Aged <7 Years: Systematic Review.

BACKGROUND: Despite various global health crises, the prevention and handling of unintentional childhood injuries remains an important public health objective. Although several systematic reviews have examined the effectiveness of different child injury prevention measures, these reviews did not address the evaluation of mobile communication intervention tools. Whether and how mobile apps were evaluated provides information on the extent to which communication theories, models, and evidence-based knowledge were considered. Previous studies have shown that the effectiveness of mobile apps increases when theories and evidence are considered during their development. OBJECTIVE: This systematic review aimed to identify research on mobile apps dealing with the prevention and handling of unintentional injuries in children and examine the theoretical and methodological approaches thereof. In addition, this review analyzed the different needs of various target groups of the mobile apps described in the articles. METHODS: In total, 8 electronic databases, ranging from interdisciplinary to medical and technical as well as social sciences databases, were searched for original research articles or brief reports in peer-reviewed journals or conference proceedings. Moreover, this review encompassed a systematic scan of articles published in the BMJ journal Injury Prevention. These steps were followed by a snowball search based on the literature references in the articles identified through the initial screening. The articles had to be written in English or German, published between 2008 and 2021, and evaluate mobile apps dealing with the prevention and handling of unintentional child injuries. The identified 5 studies were analyzed by 5 independent researchers using an inductive approach. Furthermore, the quality of the studies was assessed using the Mixed Methods Appraisal Tool. RESULTS: A total of 5 articles were included and assessed with regard to overall quality of theoretical and methodological foundations, assessed variables, the focal app's architecture, and the needs of the study participants. The overall study quality was moderate, although part of this classification was due to a lack of details reported in the studies. Each study examined 1 mobile app aimed at parents and other caregivers. Each study assessed at least 1 usability- or user experience-related variable, whereas the needs of the included study participants were detailed in only 20% (1/5) of the cases. However, none of the studies referred to theories such as the Technology Acceptance Model during the development of the apps. CONCLUSIONS: The future development and evaluation of apps dealing with the prevention and handling of child injuries should combine insights into existing models on user experience and usability with established theories on mobile information behavior. This theory-based approach will increase the validity of such evaluation studies.

[Optimisation of risk and crisis communication of governments, authorities and public health institutions-challenges in long-lasting crises illustrated by the COVID-19 pandemic]. / Optimierung der Risiko- und Krisenkommunikation von Regierungen, Behörden und Organisationen der Gesundheitssicherung ­ Herausforderungen in lang anhaltenden Krisen am Beispiel der COVID-19-Pandemie.

The COVID-19 pandemic demonstrates the great importance of risk and crisis communication. In a dynamic situation, authorities and policymakers face the challenge of dealing with a large amount of data, reviewing it and communicating it in a way that is appropriate for diverse target groups. Comprehensible and unambiguous information on risks and options for action make a significant contribution to the objective and subjective safety of the population. Hence, there is a great need to use the experience gained from the pandemic to optimize risk and crisis communication.Digitalization enables multimodal arrangements - that is, the combination of text, graphics, icons, images, animations and sound. These arrangements play an increasingly important role in risk and crisis communication. It is of interest to what extent the communicative interaction of authorities, media and other public actors in crisis preparation and management in view of a complex public can be improved with the help of target group-specific communication and how legal certainty can be ensured for official and media practice. Accordingly, the article pursues three objectives:1. It describes the challenges faced by authorities and media actors in pandemic communication.2. It shows the role of multimodal arrangements as well as the necessary research perspectives to grasp the complexity of communicative crisis management in the federal system.3. It provides a rationale for how an interdisciplinary research network from the fields of media, communication and law can gain insights into the evidence-based use of multimodal communication.

Digital dashboards visualizing public health data: a systematic review.

Introduction: Public health is not only threatened by diseases, pandemics, or epidemics. It is also challenged by deficits in the communication of health information. The current COVID-19 pandemic demonstrates that impressively. One way to deliver scientific data such as epidemiological findings and forecasts on disease spread are dashboards. Considering the current relevance of dashboards for public risk and crisis communication, this systematic review examines the state of research on dashboards in the context of public health risks and diseases. Method: Nine electronic databases where searched for peer-reviewed journal articles and conference proceedings. Included articles (n = 65) were screened and assessed by three independent reviewers. Through a methodological informed differentiation between descriptive studies and user studies, the review also assessed the quality of included user studies (n = 18) by use of the Mixed Methods Appraisal Tool (MMAT). Results: 65 articles were assessed in regards to the public health issues addressed by the respective dashboards, as well as the data sources, functions and information visualizations employed by the different dashboards. Furthermore, the literature review sheds light on public health challenges and objectives and analyzes the extent to which user needs play a role in the development and evaluation of a dashboard. Overall, the literature review shows that studies that do not only describe the construction of a specific dashboard, but also evaluate its content in terms of different risk communication models or constructs (e.g., risk perception or health literacy) are comparatively rare. Furthermore, while some of the studies evaluate usability and corresponding metrics from the perspective of potential users, many of the studies are limited to a purely functionalistic evaluation of the dashboard by the respective development teams. Conclusion: The results suggest that applied research on public health intervention tools like dashboards would gain in complexity through a theory-based integration of user-specific risk information needs. Systematic review registration: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=200178, identifier: CRD42020200178.

Prioritization of non-coding elements involved in non-syndromic cleft lip with/without cleft palate through genome-wide analysis of de novo mutations.

Non-syndromic cleft lip with/without cleft palate (nsCL/P) is a highly heritable facial disorder. To date, systematic investigations of the contribution of rare variants in non-coding regions to nsCL/P etiology are sparse. Here, we re-analyzed available whole-genome sequence (WGS) data from 211 European case-parent trios with nsCL/P and identified 13,522 de novo mutations (DNMs) in nsCL/P cases, 13,055 of which mapped to non-coding regions. We integrated these data with DNMs from a reference cohort, with results of previous genome-wide association studies (GWASs), and functional and epigenetic datasets of relevance to embryonic facial development. A significant enrichment of nsCL/P DNMs was observed at two GWAS risk loci (4q28.1 (p = 8 × 10-4) and 2p21 (p = 0.02)), suggesting a convergence of both common and rare variants at these loci. We also mapped the DNMs to 810 position weight matrices indicative of transcription factor (TF) binding, and quantified the effect of the allelic changes in silico. This revealed a nominally significant overrepresentation of DNMs (p = 0.037), and a stronger effect on binding strength, for DNMs located in the sequence of the core binding region of the TF Musculin (MSC). Notably, MSC is involved in facial muscle development, together with a set of nsCL/P genes located at GWAS loci. Supported by additional results from single-cell transcriptomic data and molecular binding assays, this suggests that variation in MSC binding sites contributes to nsCL/P etiology. Our study describes a set of approaches that can be applied to increase the added value of WGS data.

Exome-wide association study to identify rare variants influencing COVID-19 outcomes: Results from the Host Genetics Initiative.

Host genetics is a key determinant of COVID-19 outcomes. Previously, the COVID-19 Host Genetics Initiative genome-wide association study used common variants to identify multiple loci associated with COVID-19 outcomes. However, variants with the largest impact on COVID-19 outcomes are expected to be rare in the population. Hence, studying rare variants may provide additional insights into disease susceptibility and pathogenesis, thereby informing therapeutics development. Here, we combined whole-exome and whole-genome sequencing from 21 cohorts across 12 countries and performed rare variant exome-wide burden analyses for COVID-19 outcomes. In an analysis of 5,085 severe disease cases and 571,737 controls, we observed that carrying a rare deleterious variant in the SARS-CoV-2 sensor toll-like receptor TLR7 (on chromosome X) was associated with a 5.3-fold increase in severe disease (95% CI: 2.75-10.05, p = 5.41x10-7). This association was consistent across sexes. These results further support TLR7 as a genetic determinant of severe disease and suggest that larger studies on rare variants influencing COVID-19 outcomes could provide additional insights.

Next-generation phenotyping contributing to the identification of a 4.7 kb deletion in KANSL1 causing Koolen-de Vries syndrome.

Next-generation phenotyping (NGP) is an application of advanced methods of computer vision on medical imaging data such as portrait photos of individuals with rare disorders. NGP on portraits results in gestalt scores that can be used for the selection of appropriate genetic tests, and for the interpretation of the molecular data. Here, we report on an exceptional case of a young girl that was presented at the age of 8 and 15 and enrolled in NGP diagnostics on the latter occasion. The girl had clinical features associated with Koolen-de Vries syndrome (KdVS) and a suggestive facial gestalt. However, chromosomal microarray (CMA), Sanger sequencing, multiplex ligation-dependent probe analysis (MLPA), and trio exome sequencing remained inconclusive. Based on the highly indicative gestalt score for KdVS, the decision was made to perform genome sequencing to also evaluate noncoding variants. This analysis revealed a 4.7 kb de novo deletion partially affecting intron 6 and exon 7 of the KANSL1 gene. This is the smallest reported structural variant to date for this phenotype. The case illustrates how NGP can be integrated into the iterative diagnostic process of test selection and interpretation of sequencing results.

Reconstruction of the origin of the first major SARS-CoV-2 outbreak in Germany.

The first major COVID-19 outbreak in Germany occurred in Heinsberg in February 2020 with 388 officially reported cases. Unexpectedly, the first outbreak happened in a small town with little to no travelers. We used phylogenetic analyses to investigate the origin and spread of the virus in this outbreak. We sequenced 90 (23%) SARS-CoV-2 genomes from the 388 reported cases including the samples from the first documented cases. Phylogenetic analyses of these sequences revealed mainly two circulating strains with 74 samples assigned to lineage B.3 and 6 samples assigned to lineage B.1. Lineage B.3 was introduced first and probably caused the initial spread. Using phylogenetic analysis tools, we were able to identify closely related strains in France and hypothesized the possible introduction from France.

[Risk communication of policy advising scientific organisations: a thematic outline using the example of the German Federal Institute for Risk Assessment]. / Risikokommunikation politikberatender Wissenschaftsorganisationen: Ein Themenaufriss am Beispiel des Bundesinstituts für Risikobewertung.

Regulatory scientific agencies, such as the German Federal Institute for Risk Assessment (BfR), are confronted with various challenges in their science-based risk communication. On the one hand, the communication of health risks is becoming increasingly complex and, accordingly, more demanding, which is why - among other things - questions about the health literacy of consumers and target group-oriented risk communication are gaining importance. On the other hand, the outputs of regulatory scientific agencies are increasingly exposed to politicisation and public criticism, which increases the relevance of questions about the objectivity and trustworthiness of expert opinion, risk assessment, and official statements, as well as the legitimacy and reputation of such organisations. This is further intensified by the emergence of new social media actors who produce and publish their own information and communication materials. The misinformation, disinformation and malinformation consequently distributed in this context represent another challenge, which is closely related to questions of adequate communication of health risks and of stabilising legitimacy, reputation and trustworthiness.The article discusses various approaches to solving these problems, including the optimization and visual enhancement of health information, enabling social participation as well as embedding these measures in strategic stakeholder and reputation management. The article concludes with a call for a more open discussion of inherent dilemmas.

DeepCNV: a deep learning approach for authenticating copy number variations.

Copy number variations (CNVs) are an important class of variations contributing to the pathogenesis of many disease phenotypes. Detecting CNVs from genomic data remains difficult, and the most currently applied methods suffer from an unacceptably high false positive rate. A common practice is to have human experts manually review original CNV calls for filtering false positives before further downstream analysis or experimental validation. Here, we propose DeepCNV, a deep learning-based tool, intended to replace human experts when validating CNV calls, focusing on the calls made by one of the most accurate CNV callers, PennCNV. The sophistication of the deep neural network algorithm is enriched with over 10 000 expert-scored samples that are split into training and testing sets. Variant confidence, especially for CNVs, is a main roadblock impeding the progress of linking CNVs with the disease. We show that DeepCNV adds to the confidence of the CNV calls with an optimal area under the receiver operating characteristic curve of 0.909, exceeding other machine learning methods. The superiority of DeepCNV was also benchmarked and confirmed using an experimental wet-lab validation dataset. We conclude that the improvement obtained by DeepCNV results in significantly fewer false positive results and failures to replicate the CNV association results.