Explore the vast range of titles available.

In broilers, the effects of crude protein (CP) reduction on animal performance are heterogeneous. This could limit the use of this strategy in commercial farms despite its potential to improve the sustainability of production. The objective of this meta-analysis was to study the effect of lowering dietary CP in fast-growing broilers with a focus on growth performance. A database was built from 29 papers published after 2016, for a total of 106 trials and 268 treatments. To be included in the database, trials had to be iso-energy and iso-lysine. Trials in which the growth rate for the control treatment was below 90% of the genetic potential of the birds were not included. The effect of the CP level was analyzed by multiple linear regression, with the trial as a random effect. A subsample of 33 trials (AACON) met the recommended amino acid (AA)-to-lysine ratios for indispensable AAs. In this subdatabase, average daily gain and average daily feed intake were maintained when dietary CP was reduced. The feed conversion ratio increased by 1.3% when CP was reduced by a one percentage point. The same CP reduction decreased daily nitrogen (N) excretion by 10.4%, whereas N retention was not affected by CP. In conclusion, respecting the ideal protein concept with the use of feed-grade AA allows for a decrease in dietary CP and N excretion with a limited impact on growth performance.

Background Salivary agglutinin, also known as deleted in malignant brain tumors 1 (DMBT1), is an anti-microbial protein. DMBT1 is low in saliva from patients with oral squamous cell carcinoma (OSCC) and dramatically increases after treatment, with accompanying microbial changes. While this suggests an association between DMBT1 suppression and changes in the oral microbiota, causation has not been established. DMBT1 is also a tumor suppressor protein; its loss promotes OSCC progression, but its role in OSCC development is unknown. In this study, OSCC development was investigated in a murine carcinogen model that simulates human OSCC. Microbiota were standardized between Dmbt1 knockout ( Dmbt1 −/− ) and wild-type ( Dmbt1 + / + ) mice via interbreeding and co-housing. Saliva was collected at baseline and at 4, 8, 12, 16, and 22 weeks post-carcinogen initiation (stopped at 16 weeks). Tongues were harvested at week 22 for histopathology, and the salivary microbiome was profiled by 16S rRNA sequencing. Microbial diversity metrics and conditional dependence networks assessed community structure, while longitudinal patterns were analyzed using a locally sparse varying coefficient mixed model and functional principal component analysis (fPCA). Results Despite microbiota standardization, Dmbt1 −/− and Dmbt1 + / + displayed differences in microbiome composition based on β-diversity metrics. At endpoint, carcinogen-treated Dmbt1 −/− showed higher OSCC prevalence and more aggressive invasion than Dmbt1 + / + . Several OTUs, including those from Lachnospiraceae, Sphingomonas , Carnobacteriaceae, and Candidatus Saccharibacteria families, demonstrated differential abundance patterns over time, either genotype-specific, diagnosis-specific, or both. Notably, Sphingomonas and Lachnospiraceae exhibited time-dependent abundance differences in mice that developed OSCC. fPCA identified taxa with abundance trajectories that were different between OSCC and precancer and genotype specific. Conclusions Thus, DMBT1 shapes salivary microbiota composition and protects against OSCC development. Dynamic, genotype-specific microbial shifts during carcinogenesis underscore the complex interplay between the oral microbiota and cancer progression. 52sTgPoWwfY7A3YogZ1rT1 Video Abstract Graphical Abstract

The aim of this paper is to define the performance of natural ventilation from different levels, taking into account the climatic potential, the energy performance of the building and the use of the building in terms of thermal comfort of the occupants. To do this, Building Energy Simulation has been used to simulate simplified office buildings with different parameters to quantify their impact on the natural ventilation efficiency. These parameters include three levels of energy performance of the envelope, three different cooling modes with air conditioning only, natural ventilation only and a mixed mode between both, two weather files, and two scenarios of natural ventilation (possible all the day long or only during occupied period). Finally, various KPIs were calculated to show the diversity of ways of looking at natural ventilation performance, and the impacts of changes in the models’ input parameters on the three defined performance levels. For example, we have shown that a particular building can have good thermal comfort without having the best building potential, demonstrating the need for multiple levels of natural ventilation performance.

Dependency among observations can arise from a multitude of sources, including spatial or temporal correlation, grouped, clustered or repeated measurements, hierarchical structures, and dyadic interactions. Neglecting these interdependencies in statistical analyses may result in incorrect inferences or loss of statistical power. Conversely, adequately modeling these dependencies not only enhances the validity of our statistical inferences but also deepens our comprehension of the intricate dynamics that generate the data. In this dissertation, we propose innovative methodologies tailored to three cases, each exemplifying unique challenges of dependent data.First, we consider an imputation task where dependency among subjects is captured by a network structure. The dual nature of the data features pairwise binary relationships between subjects and subject-specific attributes that are partially observed and need to be imputed. While it is possible to conduct imputation leveraging solely subject-wise information, the relational data embedded in the network could offer auxiliary information that enhance the imputation accuracy. To capitalize on this, we explore a joint latent space model that employs subject-specific latent variables to bridge the two data modes, simultaneously addressing the interdependencies among both subjects and attributes. By adopting a Bayesian framework, we achieve an optimal integration of all available information, leading to more accurate imputed values. We ensure the practical viability of our model through the use of variational approximations.Second, we study a differential analysis task within the context of longitudinal data, driven by a study on microbial abundance data. To accommodate the temporal dependencies and the continuous nature of the biological data, we opt for a varying-coefficient mixed model with kernel smoothing. Through a sparsity-inducing penalty, we identify time periods of differences between experimental conditions. Our novel estimation method is the first of its kind to simultaneously account for longitudinal effects while obtaining smooth, locally-sparse estimates and permitting any sampling design, such as irregularly-sample time points or missing data. Crucially, accounting for time dependence is shown to improve estimation accuracy as well as support recovery, in comparison to an equivalent approach with independence assumptions. This provides greater confidence in the discoveries obtained when applied to the motivating data, which revealed novel scientific insights that eluded detection by cross-sectional or independent approaches.Third, we proceed with a joint inference and prediction task emerging from a brain-computer interface setting. Specifically, we aim to construct a model of the brain’s electrical activity, as captured by multiple electroencephalogram (EEG) channels over time, in relation to the brain’s response to visual stimuli presented under an oddball paradigm, in which only certain stimuli are pertinent. To that end, we propose a Bayesian model explaining the measured response to target and nontarget stimuli, with particular interest in the difference in responses. This model not only explains the observed data but also generates predictive probabilities, thus enabling the classification of unlabeled responses. Our model innovates in three ways over existing approaches. We aggregate EEG channels in a collection of latent factors, which abstracts the EEG system design and naturally capture spatial correlation. Additionally, our model allows temporal variations in spatial covariance, thus supporting dynamic modeling of the covariance structure. Lastly, our approach not only characterizes mean differences in response based on stimulus type but also allows for the dynamic spatial covariance to be modulated by the type of stimulus being presented. Experimental validation confirms that our framework yields prediction accuracy on par with that of discriminative methods. However, the flexibility inherent in our generative model provides deeper understanding of the underlying brain functions responsible for the responses.

Actuarial practitioners now have access to multiple sources of insurance data corresponding to various situations: multiple business lines, umbrella coverage, multiple hazards, and so on. Despite the wide use and simple nature of single-target approaches, modeling these types of data may benefit from an approach performing variable selection jointly across the sources. We propose a unified algorithm to perform sparse learning of such fused insurance data under the Tweedie (compound Poisson) model. By integrating ideas from multitask sparse learning and sparse Tweedie modeling, our algorithm produces flexible regularization that balances predictor sparsity and between-sources sparsity. When applied to simulated and real data, our approach clearly outperforms single-target modeling in both prediction and selection accuracy, notably when the sources do not have exactly the same set of predictors. An efficient implementation of the proposed algorithm is provided in our R package MStweedie, which is available at https://github.com/fontaine618/MStweedie . Supplementary materials for this article are available online.

Microbiome differential abundance analysis remains a challenging problem despite multiple methods proposed in the literature. The excessive zeros and compositionality of metagenomics data are two main challenges for differential abundance analysis. We propose a novel method called \"analysis of differential abundance by pooling Tobit models\" (ADAPT) to overcome these two challenges. ADAPT uniquely treats zero counts as left-censored observations to facilitate computation and enhance interpretation. ADAPT also encompasses a theoretically justified way of selecting non-differentially abundant microbiome taxa as a reference for hypothesis testing. We generate synthetic data using independent simulation frameworks to show that ADAPT has more consistent false discovery rate control and higher statistical power than competitors. We use ADAPT to analyze 16S rRNA sequencing of saliva samples and shotgun metagenomics sequencing of plaque samples collected from infants in the COHRA2 study. The results provide novel insights into the association between the oral microbiome and early childhood dental caries.

Fréchet regression provides a versatile framework for modeling responses in metric spaces with Euclidean predictors, yet current methodologies rely almost exclusively on frequentist approaches. We propose a Bayesian framework for Fréchet regression that offers a principled way of incorporating prior information into nonlinear global Fréchet regression. By targeting a novel Fréchet Bayes rule, we reduce the object-valued regression problem to a collection of tractable scalar regression tasks. Our approach allows for a controlled interpolation between the prior and the data-driven frequentist estimate, facilitating effective shrinkage toward informed values. While initially derived under Gaussian assumptions, we demonstrate that our framework is robust to model misspecification by establishing its validity under moment conditions via weak conditional expectations. The numerical properties of the proposed methodology are demonstrated in simulation studies and an application to microbiome compositional data, where we show that leveraging an auxiliary cohort to inform the prior significantly enhances predictive performance in a targeted, small-scale study

Differential abundance (DA) analysis in microbiome studies has recently been used to uncover a plethora of associations between microbial composition and various health conditions. While current approaches to DA typically apply only to cross-sectional data, many studies feature a longitudinal design to better understand the underlying microbial dynamics. To study DA in longitudinal microbial studies, we introduce a novel varying coefficient mixed-effects model with local sparsity. The proposed method can identify time intervals of significant group differences while accounting for temporal dependence. Specifically, we exploit a penalized kernel smoothing approach for parameter estimation and include a random effect to account for serial correlation. In particular, our method operates effectively regardless of whether sampling times are shared across subjects, accommodating irregular sampling and missing observations. Simulation studies demonstrate the necessity of modeling dependence for precise estimation and support recovery. The application of our method to a longitudinal study of mice oral microbiome during cancer development revealed significant scientific insights that were otherwise not discernible through cross-sectional analyses. An R implementation is available at https://github.com/fontaine618/LSVCMM.

SARS-CoV-2 has infected millions of people globally and continues to undergo evolution. Emerging variants can be partially resistant to vaccine induced immunity and therapeutic antibodies, emphasizing the urgent need for accessible, broad-spectrum therapeutics. Here, we report a comprehensive study of ensovibep, the first trispecific clinical DARPin candidate, that can simultaneously engage all three units of the spike protein trimer to potently inhibit ACE2 interaction, as revealed by structural analyses. The cooperative binding of the individual modules enables ensovibep to retain inhibitory potency against all frequent SARS-CoV-2 variants, including Omicron BA.1 and BA.2, as of February 2022. Moreover, viral passaging experiments show that ensovibep, when used as a single agent, can prevent development of escape mutations comparably to a cocktail of monoclonal antibodies (mAb). Finally, we demonstrate that the very high in vitro antiviral potency also translates into significant therapeutic protection and reduction of pathogenesis in Roborovski dwarf hamsters infected with either the SARS-CoV-2 wild-type or the Alpha variant. In this model, ensovibep prevents fatality and provides substantial protection equivalent to the standard of care mAb cocktail. These results support further clinical evaluation and indicate that ensovibep could be a valuable alternative to mAb cocktails and other treatments for COVID-19. Competing Interest Statement Authors from Molecular Partners own performance share units and/or stock of the company. H.K.B. owns stock of the company. I.D. is an employee of Thermo Fisher Scientific. C.G.K.; K.K.B. and K.R. are employees of Novartis. The other authors declare no competing interests. Footnotes * The revision includes additional data, highly relevant for the COVID-19 pandemic, incl. data from the viral strains delta and omicron BA.1 as well as BA.2.

Globally accessible therapeutics against SARS-CoV-2 are urgently needed. Here, we report the generation of the first anti-SARS-CoV-2 DARPin molecules with therapeutic potential as well as rapid large-scale production capabilities. Highly potent multivalent DARPin molecules with low picomolar virus neutralization efficacies were generated by molecular linkage of three different monovalent DARPin molecules. These multivalent DARPin molecules target various domains of the SARS-CoV-2 spike protein, thereby limiting possible viral escape. Cryo-EM analysis of individual monovalent DARPin molecules provided structural explanations for the mode of action. Analysis of the protective efficacy of one multivalent DARPin molecule in a hamster SARS-CoV-2 infection model demonstrated a significant reduction of pathogenesis. Taken together, the multivalent DARPin molecules reported here, one of which has entered clinical studies, constitute promising therapeutics against the COVID-19 pandemic. Competing Interest Statement Molecular Partners authors own performance share units and/or stock of the company. HKB owns stock of the company. Footnotes * This version of the manuscript has been revised to update Figures and add additional data (e.g. Fig 3).