Different neural correlates of deception: Crafting high and low creative scams

Deception is a complex social behavior that manifests in various forms, including scams. To successfully deceive victims, liars have to continually devise novel scams. This ability to create novel scams represents one kind of malevolent creativity, referred to as lying. This study aimed to explore different neural substrates involved in the generation of high and low creative scams. A total of 40 participants were required to design several creative scams, and their cortical activity was recorded by functional near-infrared spectroscopy. The results revealed that the right frontopolar cortex (FPC) was significantly active in scam generation. This region associated with theory of mind may be a key region for creating novel and complex scams. Moreover, creativity-related regions were positively involved in creative scams, while morality-related areas showed negative involvement. This suggests that individuals might attempt to use malevolent creativity while simultaneously minimizing the influence of moral considerations. The right FPC exhibited increased coupling with the right precentral gyrus during the design of high-harmfulness scams, suggesting a diminished control over immoral thoughts in the generation of harmful scams. Additionally, the perception of the victim's emotions (related to right pre-motor cortex) might diminish the quality of highly original scams. Furthermore, an efficient and cohesive neural coupling state appears to be a key factor in generating high-creativity scams. These findings suggest that the right FPC was crucial in scam creation, highlighting a neural basis for balancing malevolent creativity against moral considerations in high-creativity deception.

Early minimally invasive image-guided endoscopic evacuation of intracerebral hemorrhage (EMINENT-ICH): a randomized controlled trial

BACKGROUND

Spontaneous supratentorial intracerebral hemorrhage is the deadliest form of stroke with mortality rates over 50%. Currently, no sufficiently effective treatment to improve both mortality and functional outcome rates exists. However, it seems that minimally invasive surgery, especially endoscopic surgery, might be beneficial in improving survival and functional outcome rates, yet large confirmatory studies thereof are lacking. The aim of this trial is to compare whether early minimally invasive endoscopic surgery leads to improved functional outcome rates compared to the best medical treatment.

METHODS

This is a prospective, parallel-arm, outcome assessor blinded multicenter trial across Switzerland. Endoscopic surgery will be compared to the best medical treatment in a 1:1 randomization over a total time of 12 months. The primary outcome is defined as improved functional outcome (mRS < 3) after 6 months; secondary outcomes include mortality and morbidity rates as well as patient reported outcomes and the temporal evolution of serum biomarkers for brain damage.

DISCUSSION

Currently, large, randomized trials assessing the role and potential effect of early endoscopic surgery in intracerebral hemorrhage are lacking. Potential practical and methodological issues faced in this trial are patient enrollment, adherence to the hematoma evacuation technique used, potential patient cross-over, and the adaptive Bayesian statistical design. Nonetheless, this trial would be among the first to research the effects of early minimally invasive endoscopic surgery for SSICH and can provide class I evidence for future treatment options in intracerebral hemorrhage.

TRIAL REGISTRATION

ClinicalTrials.gov NCT05681988. Registered on January 3, 2023.

Similar humoral responses but distinct CD4+ T cell transcriptomic profiles in older adults elicited by MF59 adjuvanted and high dose influenza vaccines

Older age (≥ 65 years) is associated with impaired responses to influenza vaccination, leading to the preferential recommendation of MF59-adjuvanted (MF59Flu) or high-dose (HDFlu) influenza vaccines for this age group in the United States. Herein, we characterized transcriptomic profiles of CD4+ T cells isolated from 234 recipients (≥ 65 years) of either MF59Flu or HDFlu vaccine, prior to vaccination and 28 days thereafter. We identified 412 and 645 differentially expressed genes (DEGs) in CD4+ T cells of older adults after receiving MF59Flu and HDFlu, respectively. DEGs in CD4+ T cells of MF59Flu recipients were enriched in 14 KEGG pathways, all of which were downregulated. DEGs in CD4+ T cells of HDFlu recipients were enriched in 11 upregulated pathways and 20 downregulated pathways. CD4+ T cells in both vaccine groups shared 50 upregulated genes and 75 downregulated genes, all of which were enriched in 7 KEGG pathways. The remaining 287 and 520 DEGs were specifically associated with MF59Flu and HDFlu, respectively. Unexpectedly, none of these DEGs was significantly correlated with influenza A/H3N2-specific HAI titers, suggesting these DEGs at the individual level may have a limited role in protection against influenza. Our findings emphasize the need for further investigation into other factors influencing immunity against influenza in older adults.

Combined effects of benzene, toluene, xylene, ethylbenzene, and styrene exposure on hearing loss mediated by oxidative stress at realistic low levels

The link between benzene, toluene, ethylbenzene, xylene, and styrene (BTEXS) exposure and hearing loss (HL) is not well-established. This study investigated 1694 petrochemical workers in southern China to examine the effects of BTEXS urinary metabolites on auditory function, considering oxidative stress (OS) as a potential cause. Using generalized linear models, elastic net regression, and quantile g-computation, we evaluated the single and combined effects of BTEXS, OS indicators, and HL. Subgroup analysis explored interactions between BTEXS and cumulative noise exposure (CNE), while mediation analysis assessed OS's role in BTEXS-related HL. Positive associations were found between hippuric acid (HA) and HL (OR = 1.20, P < 0.05) and high-frequency hearing loss (HFHL) (OR = 1.22, P < 0.05). The ENET model linked 3&4-methylhippuric acid (3&4-MHA) with increased HFHL risk. The qgcomp model showed a 23% increased HL risk and a 20% increased HFHL risk per quartile increase in BTEXS exposure. Toluene metabolites (SBMA and HA) were significant contributors to HL, HFHL, and speech-frequency hearing loss (SFHL). Higher BTEXS SBMA, MA and HA levels exacerbated HL risk in workers exposed to CNE. Interaction analysis revealed synergistic effects between tt-MA and other metabolites on HFHL risk. Total SOD (TSOD) significantly mediated the BTEXS-HL relationship. These findings highlight a dose-effect association between BTEXS exposure and HL due to oxidative damage, with toluene metabolites being critical pollutants. BTEXS exposure also synergistically increased HL risk with noise.

Unveiling new protein biomarkers and therapeutic targets for acne through integrated analysis of human plasma proteomics and genomics

Background

The current landscape of acne therapeutics is notably lacking in targeted treatments, highlighting a critical need for the discovery of new drug targets to improve treatment outcomes.

Objectives

This study aims to investigate the connections between proteomics and genetics in relation to acne across extensive population cohorts, aspiring to identify innovative preventive and therapeutic approaches.

Methods

Employing a longitudinal cohort of 54,306 participants from the UK Biobank Pharmacological Proteomics Project (UKB-PPP), we performed an exhaustive evaluation of the associations between 2,923 serum proteins and acne risk. Initial multivariate Cox regression analyses assessed the relationship between protein expression levels and acne onset, followed by two-sample Mendelian Randomization (TSMR), Summary-data-based Mendelian Randomization (SMR), and colocalization to identify genetic correlations with potential protein targets.

Results

Within the UKB cohort, we identified 19 proteins significantly associated with the risk of acne. Subsequent analysis using Two-Sample Mendelian Randomization (TSMR) refined this to two specific proteins: FSTL1 and ANXA5. Each one-standard deviation increase in the expression levels of FSTL1 and ANXA5 was associated with a 24% and 32% increase in acne incidence, respectively. These results were further validated by additional Summary-data-based Mendelian Randomization (SMR) and differential expression analyses.

Conclusions

Our comprehensive analysis of proteomic and genetic data from a European adult cohort provides compelling causal evidence that several proteins are promising targets for novel acne treatments.

Dispositional Traits, Characteristic Adaptations, and Narrative Identity Reconstructions in Individuals With Depersonalization and Derealization

INTRODUCTION

Depersonalization and derealization disorder (DPDR) is a debilitating condition. To date, little was known about the role of personality structure and of perceived social support and loneliness in DPDR.

METHODS

Three studies investigated, respectively: (i) broadband personality traits (five-factor model), maladaptive trait domains (PID-5), and perceived support and loneliness in individuals with self-reported DPDR (N = 160) versus a general population sample (N = 303), using network modeling; (ii) structure and interconnectivity of personality, perceived support and loneliness, and DPDR traits (frequency/duration) in individuals with self-reported DPDR (N = 160); (iii) characteristic adaptations and narrative identities in individuals with self-reported DPDR (N = 19), using thematic analysis.

RESULTS

Study 1 found between-samples differences across several traits, especially psychoticism and negative affect. Differences in networks' global centrality, but not structures or edges, were also found. The graphical model in Study 2 showed a community of dissociative tendencies including DPDR traits and psychoticism. Study 3 highlighted the development of DPDR as a key life transition for those experiencing it, with narratives focusing on feelings of poor agency, isolation, and a disrupted sense of self.

CONCLUSIONS

Individual differences in personality characterize DPDR, especially in psychoticism. Implications for theory and research are discussed.

Characterization of ligand-receptor pair in acute myeloid leukemia: a scoring model for prognosis, therapeutic response, and T cell dysfunction

Introduction

The significance of ligand-receptor (LR) pair interactions in the progression of acute myeloid leukemia (AML) has been the focus of numerous studies. However, the relationship between LR pairs and the prognosis of AML, as well as their impact on treatment outcomes, is not fully elucidated.

Methods

Leveraging data from the TCGA-LAML cohort, we mapped out the LR pair interactions and distinguished specific molecular subtypes, with each displaying distinct biological characteristics. These subtypes exhibited varying mutation landscapes, pathway characteristics, and immune infiltration levels. Further insight into the immune microenvironment among these subtypes revealed disparities in immune cell abundance.

Results

Notably, one subtype showed a higher prevalence of CD8 T cells and plasma cells, suggesting increased adaptive immune activities. Leveraging a multivariate Lasso regression, we formulated an LR pair-based scoring model, termed "LR.score," to classify patients based on prognostic risk. Our findings underscored the association between elevated LR scores and T-cell dysfunction in AML. This connection highlights the LR score's potential as both a prognostic marker and a guide for personalized therapeutic interventions. Moreover, our LR.score revealed substantial survival prediction capacities in an independent AML cohort. We highlighted CLEC11A, ICAM4, ITGA4, and AVP as notably AML-specific.

Discussion

qRT-PCR analysis on AML versus normal bone marrow samples confirmed the significant downregulation of CLEC11A, ITGA4, ICAM4, and AVP in AML, suggesting their inverse biomarker potential in AML. In summary, this study illuminates the significance of the LR pair network in predicting AML prognosis, offering avenues for more precise treatment strategies tailored to individual patient profiles.

Modelling the spatial variability and uncertainty for under-vaccination and zero-dose children in fragile settings

Universal access to childhood vaccination is important to child health and sustainable development. Here we identify, at a fine spatial scale, under-immunized children and zero-dose children. Using Chad, as an example, the most recent nationally representative household survey that included recommended vaccine antigens was assembled. Age-disaggregated population (12-23 months) and vaccination coverage were modelled at a fine spatial resolution scale (1km × 1 km) using a Bayesian geostatistical framework adjusting for a set of parsimonious covariates. There was a variation at fine spatial scale in the population 12-23 months a national mean of 18.6% (CrI 15.8%-22.6%) with the highest proportion in the South-East district of Laremanaye 20.0% (14.8-25.0). Modelled coverage at birth was 49.0% (31.2%-75.3%) for BCG, 44.8% (27.1-74.3) for DTP1, 24.7% (12.5-46.3) for DTP3 and 47.0% (30.6-71.0) for measles (MCV1). Combining coverage estimates with the modelled population at a fine spatial scale yielded 312,723 (Lower estimate 156055-409266) zero-dose children based on DTP1. Improving routine immunization will require investment in the health system as part of enhancing primary health care. The uncertainties in our estimates highlight areas that require further investigation and higher quality data to gain a better understanding of vaccination coverage.

Assessment of sensitivity and specificity of bacterial culture and the VetMAX™ MastiType Multi Kit in detecting Streptococcus uberis and Escherichia coli in milk samples from dairy cows with clinical mastitis in subtropical Australia

Mastitis, a prevalent and economically important disease in the dairy industry, poses substantial challenges to dairy cow health, milk quality, and farm profitability worldwide. Mastitis is predominantly caused by bacterial infections. The objective of this study was to estimate the sensitivity (Se) and specificity (Sp) of bacterial culture and the VetMAX™ MastiType Multi Kit PCR in identified clinical mastitis pathogens. A total of 396 quarter-level milk samples were collected from 396 cows with clinical mastitis on 29 farms in the subtropical dairy region of Australia between March and December 2021. These samples were cultured and tested by PCR, and analysed using Bayesian latent class analysis under the assumption of one population two tests and also of three populations two tests, by dividing the population into subpopulations based on regions. Informative priors used in the analysis were calculated from published evidence. Models were compared using the Deviance Information Criterion (DIC). Sensitivity analysis was performed to evaluate the impact of changes in priors. The most common isolates cultured and detected by PCR were Streptococcus uberis (17.4 % and 27.3 %, respectively) and Escherichia coli (12.6 % and 25.0 %, respectively). Under the assumption of one population two tests, the Se of PCR (at cycle threshold (Ct) ≤ 37) was higher than that of bacterial culture for both pathogens: for E. coli, the Se was 50.2 % (95 % posterior probability interval (PPI): 37.4; 74.1) for bacterial culture, and 93.7 % (95 % PPI: 85.5; 98.4) for PCR. For S. uberis, the Se was 50.4 % (95 % PPI: 40.9; 61.3) for bacterial culture, and 81.5 % (73.0; 88.9) for PCR. Conversely, the Sp of bacterial culture was higher than that of PCR for both pathogens: for E. coli, the Sp was 99.2 % (97.8; 100) for bacterial culture, and 95.1 % (87.8; 99.4) for PCR. For S. uberis, the Sp was 99.2 % (95 % PPI: 97.6; 100) for bacterial culture, and 96.7 % (95 % PPI: 92.1; 99.2) for PCR. Bayesian latent class analysis with three populations two tests was only performed for S. uberis. For E. coli, this could not be performed because there were no PCR-positive results in one subpopulation. Under the assumption of three populations two tests, for S. uberis, the Se was 49.6 % (40.6; 59.4) for bacterial culture, and 81.1 % (72.6; 88.6) for PCR; and the Sp for bacterial culture was 99.1 % (97.7; 100), and for PCR was 96.9 % (93.0; 99.3). The DIC for the one population two tests model was lower than the DIC for the three populations two tests model. The sensitivity analysis for the one population two tests model demonstrated that a 10 % reduction in priors led to substantial changes in Se of both bacterial culture and PCR tests for E. coli and S. uberis, with overlap percentages ranging from 80.6 % to 92.2 %. In contrast, the Sp of bacterial culture and PCR tests remained relatively stable despite changes in priors, except for the Sp of PCR test for E. coli. In summary, the VetMAX™ MastiType Multi Kit demonstrated higher Se compared to bacterial culture, suggesting its potential as a routine test for identifying mastitis pathogens in milk samples from cows with clinical mastitis. While the bacterial culture method offered higher Sp in pathogen detection; results obtained following bacterial culture and subsequent susceptibility testing remain valuable, particularly in guiding antimicrobial treatment for mastitis.

Compressed sensing based approach identifies modular neural circuitry driving learned pathogen avoidance

An animal's survival hinges on its ability to integrate past information to modify future behavior. The nematode C. elegans adapts its behavior based on prior experiences with pathogen exposure, transitioning from attraction to avoidance of the pathogen. A systematic screen for the neural circuits that integrate the information of previous pathogen exposure to modify behavior has not been feasible because of the lack of tools for neuron type specific perturbations. We overcame this challenge using methods based on compressed sensing to efficiently determine the roles of individual neuron types in learned avoidance behavior. Our screen revealed that distinct sets of neurons drive exit from lawns of pathogenic bacteria and prevent lawn re-entry. Using calcium imaging of freely behaving animals and optogenetic perturbations, we determined the neural dynamics that regulate one key behavioral transition after infection: stalled re-entry into bacterial lawns. We find that key neuron types govern pathogen lawn specific stalling but allow the animal to enter nonpathogenic E. coli lawns. Our study shows that learned pathogen avoidance requires coordinated transitions in discrete neural circuits and reveals the modular structure of this complex adaptive behavioral response to infection.

Who gets better after surgery for degenerative cervical myelopathy? A responder analysis from the multicenter Canadian spine outcomes and research network

BACKGROUND CONTEXT

Degenerative cervical myelopathy (DCM) is the most common cause of acquired nontraumatic spinal cord injury worldwide. Surgery is a common treatment for DCM; however, outcomes often vary across patients.

PURPOSE

To inform preoperative education and counseling, we performed a responder analysis to identify factors associated with treatment response.

STUDY DESIGN/SETTING

An observational cohort study was conducted utilizing prospectively collected data from the Canadian Spine Outcomes Research Network (CSORN) registry collected between 2015-2022.

PATIENT SAMPLE

We included all surgically treated DCM patients with complete 12-month follow-up and patient-reported outcomes (PROs) available at 1-year.

OUTCOME MEASURES

Treatment response was measured using the minimal clinically important difference (MCID) in PROs including the Neck Disability Index (NDI) and EuroQol-5D (EQ-5D) at 12 months postsurgery.

METHODS

A Least Absolute Shrinkage and Selection Operator (LASSO) machine learning model was used to identify significant associations between 14 preoperative patient factors and likelihood of treatment response measured by achievement of the MCID in NDI, and EQ-5D. Variable importance was measured using standardized coefficients. To test robustness of findings we trained a separate XGBOOST model, with variable importance measured using SHAP values.

RESULTS

Among the 554 DCM patients included, 229 (41.3%) and 330 (59.6%) patients responded to treatment by meeting or surpassing MCID thresholds for NDI and EQ-5D at 1-year, respectively. LASSO regression for likelihood of treatment response measured through NDI found the variable importance rank order to be baseline NDI (OR 1.06 per 1 point increase; 95% CI 1.04-1.07), then symptom duration (OR 0.65; 95% CI 0.44-0.97). For EQ-5D, the variable importance rank order was baseline EQ-5D (OR 0.16 per 0.1-point increase; 95% CI 0.03-0.78), living independently (OR 2.17; 95% CI 1.22-3.85), symptom duration (OR 0.62; 95% CI 0.40-0.97), then number of levels affected (OR 0.80 per additional level; 95% CI 0.67-0.96). A separate XGBoost model of treatment response measured through NDI, corroborated findings that patients with higher baseline NDI, and shorter symptom duration were more likely to respond to treatment, and additionally found older patients, and those with kyphosis on baseline upright X-ray were less likely to respond. Similarly, an XGBoost model for treatment response measured through EQ-5D corroborated findings that patients with higher baseline EQ-5D, shorter symptom duration, living independently, with fewer affected levels were more likely to respond to treatment, and additionally found older patients were less likely to respond.

CONCLUSIONS

Our findings suggest patients with shorter symptom duration, higher baseline patient NDI, lower EQ-5D, younger age, living independently, without kyphosis on preoperative X-ray, and fewer affected levels are more likely to respond to treatment. Timing of surgery with respect to patient symptoms is underscored as a crucial and modifiable patient factor associated with improved surgical outcomes in DCM.

Causal Inference for Continuous Multiple Time Point Interventions

There are limited options to estimate the treatment effects of variables which are continuous and measured at multiple time points, particularly if the true dose-response curve should be estimated as closely as possible. However, these situations may be of relevance: in pharmacology, one may be interested in how outcomes of people living with-and treated for-HIV, such as viral failure, would vary for time-varying interventions such as different drug concentration trajectories. A challenge for doing causal inference with continuous interventions is that the positivity assumption is typically violated. To address positivity violations, we develop projection functions, which reweigh and redefine the estimand of interest based on functions of the conditional support for the respective interventions. With these functions, we obtain the desired dose-response curve in areas of enough support, and otherwise a meaningful estimand that does not require the positivity assumption. We developg $$ g $$ -computation type plug-in estimators for this case. Those are contrasted with g-computation estimators which are applied to continuous interventions without specifically addressing positivity violations, which we propose to be presented with diagnostics. The ideas are illustrated with longitudinal data from HIV positive children treated with an efavirenz-based regimen as part of the CHAPAS-3 trial, which enrolled children< 13 $$ <13 $$ years in Zambia/Uganda. Simulations show in which situations a standard g-computation approach is appropriate, and in which it leads to bias and how the proposed weighted estimation approach then recovers the alternative estimand of interest.

Neural parameter calibration and uncertainty quantification for epidemic forecasting

The recent COVID-19 pandemic has thrown the importance of accurately forecasting contagion dynamics and learning infection parameters into sharp focus. At the same time, effective policy-making requires knowledge of the uncertainty on such predictions, in order, for instance, to be able to ready hospitals and intensive care units for a worst-case scenario without needlessly wasting resources. In this work, we apply a novel and powerful computational method to the problem of learning probability densities on contagion parameters and providing uncertainty quantification for pandemic projections. Using a neural network, we calibrate an ODE model to data of the spread of COVID-19 in Berlin in 2020, achieving both a significantly more accurate calibration and prediction than Markov-Chain Monte Carlo (MCMC)-based sampling schemes. The uncertainties on our predictions provide meaningful confidence intervals e.g. on infection figures and hospitalisation rates, while training and running the neural scheme takes minutes where MCMC takes hours. We show convergence of our method to the true posterior on a simplified SIR model of epidemics, and also demonstrate our method's learning capabilities on a reduced dataset, where a complex model is learned from a small number of compartments for which data is available.

Two-stage group-sequential designs with delayed responses - what is the point of applying corresponding methods?

BACKGROUND

In group-sequential designs, it is typically assumed that there is no time gap between patient enrollment and outcome measurement in clinical trials. However, in practice, there is usually a lag between the two time points. This can affect the statistical analysis of the data, especially in trials with interim analyses. One approach to address delayed responses has been introduced by Hampson and Jennison (J R Stat Soc Ser B Stat Methodol 75:3-54, 2013), who proposed the use of error-spending stopping boundaries for patient enrollment, followed by critical values to reject the null hypothesis if the stopping boundaries are crossed beforehand. Regarding the choice of a trial design, it is important to consider the efficiency of trial designs, e.g. in terms of the probability of trial success (power) and required resources (sample size and time).

METHODS

This article aims to shed more light on the performance comparison of group sequential clinical trial designs that account for delayed responses and designs that do not. Suitable performance measures are described and designs are evaluated using the R package rpact. By doing so, we provide insight into global performance measures, discuss the applicability of conditional performance characteristics, and finally whether performance gain justifies the use of complex trial designs that incorporate delayed responses.

RESULTS

We investigated how the delayed response group sequential test (DR-GSD) design proposed by Hampson and Jennison (J R Stat Soc Ser B Stat Methodol 75:3-54, 2013) can be extended to include nonbinding lower recruitment stopping boundaries, illustrating that their original design framework can accommodate both binding and nonbinding rules when additional constraints are imposed. Our findings indicate that the performance enhancements from methods incorporating delayed responses heavily rely on the sample size at interim and the volume of data in the pipeline, with overall performance gains being limited.

CONCLUSION

This research extends existing literature on group-sequential designs by offering insights into differences in performance. We conclude that, given the overall marginal differences, discussions regarding appropriate trial designs can pivot towards practical considerations of operational feasibility.

Development of a metabolomic risk score for exposure to traffic-related air pollution: A multi-cohort study

To synthesize vast amounts of high-throughput biological information, omics-fields like epigenetics have applied risk scores to develop biomarkers for environmental exposures. Extending the risk score analytic tool to the metabolomic data would be highly beneficial. This research aimed to develop and evaluate metabolomic risk score (metRS) approaches reflecting the biological response to traffic-related air pollution (TRAP) exposure (fine particulate matter, black carbon, and nitrogen dioxide). A simulation study compared three metRS methodologies: elastic net regression, which uses penalized regression to select metabolites, and two variations of thresholding, where a p-value cutoff is used to select metabolites. The methods performance was compared to assess 1) ability to correctly select metabolites associated with daily TRAP and 2) ability of the risk score to predict daily TRAP exposure. Power calculations and false discovery rates (FDR) were calculated for each approach. This metRS was applied to two real cohorts, the Center for Health Discovery and Wellbeing (CHDWB, n = 180) and Environment and Reproductive Health (EARTH, n = 200). In simulations, elastic net regression consistently presented inflated FDR for both high and low effect sizes and across all three sample sizes (n = 200; 500; 1000). Power to detect correct metabolites exceeded 0.8 for all three sample sizes in all three methods. In the real data application assessing associations of metabolomics risk scores and TRAP, associations were largely null. While we did not identify strong associations between the risk scores and TRAP in the real data application, metabolites selected by the risk score approaches were enriched in pathways that are well-known for their association with TRAP. These results demonstrate that certain methodologies to construct metabolomics risk scores are statistically robust and valid; however, standardized metabolic profiling and large sample sizes are required.

Increased listening effort and cochlear neural degeneration underlie behavioral deficits in speech perception in noise in normal hearing middle-aged adults

Middle-age is a critical period of rapid changes in brain function that presents an opportunity for early diagnostics and intervention for neurodegenerative conditions later in life. Hearing loss is one such early indicator linked to many comorbidities later in life. However, current clinical tests fail to capture hearing difficulties for ∼10% of middle-aged adults seeking help at hearing clinics. Cochlear neural degeneration (CND) could play a role in these hearing deficits, but our current understanding is limited by the lack of objective diagnostics and uncertainty regarding its perceptual consequences. Here, using a cross-species approach, we measured envelope following responses (EFRs) - neural ensemble responses to sound originating from the peripheral auditory pathway - in young and middle-aged adults with normal audiometric thresholds, and compared these responses to young and middle-aged Mongolian gerbils, where CND was histologically confirmed. We observed near identical changes in EFRs across species that were associated with CND. Perceptual effects measured as behavioral readouts showed deficits in the most challenging listening conditions and were associated with CND. Additionally, pupil-indexed listening effort increased even at moderate task difficulties where behavioral outcomes were matched. Our results reveal perceptual deficits in middle-aged adults driven by CND and increases in listening effort, which may result in increased listening fatigue and conversational disengagement.

Development of a multifactorial prediction model for commute mode choice in 10 983 Finnish public sector employees: a cross-sectional study

OBJECTIVE

The objective of this study is to examine the feasibility of using survey data to identify factors that predict commute mode choice.

DESIGN

The study design is cross-sectional.

SETTING

Survey data from the Finnish Public Sector study (2020) were used.

PARTICIPANTS

42 574 public sector employees, of whom 10 983 were selected for the final sample. These included employees with 5 km or less commuting distances and those working full-time onsite or partly remotely. The mean age was 46 (SD 11) years, and 84% were women.

PRIMARY OUTCOMES

Commute by (1) bike or foot (an active mode) during summer and winter weather and (2) by car (a passive mode) during summer and winter weather.

METHODS

Using logistic Lasso (least-absolute-shrinkage-and-selection-operator) regression, we developed and tested a prediction model for short commutes of 5 km or less to identify the characteristics of employees most likely to commute actively during summer and winter weather and passively during summer and winter weather.

RESULTS

All models had a good predictive ability with a C-index of 0.82, 0.77, 0.72 and 0.71. Cycling and walking during summer weather were predicted by shorter commutes, higher physical activity, lower body mass index (BMI), female sex and higher team psychological safety. Predictors of cycling and walking during winter weather were shorter commute length, higher physical activity, lower BMI and higher age. Commuting by car during summer weather was predicted by longer journey length, higher BMI, lower physical activity, male sex and having children 7-18 years old living at home. Predictors of driving during winter weather were almost identical, but the male sex was replaced by having a spouse.

CONCLUSIONS

We identified the correlates of active and passive commute choice in different weather conditions with eight variables. This information can be used to develop and target interventions to promote sustainable and healthy commuting modes.

Multi-omics data integration and analysis pipeline for precision medicine: Systematic review

Precision medicine has gained considerable popularity since the "one-size-fits-all" approach did not seem very effective or reflective of the complexity of the human body. Subsequently, since single-omics does not reflect the complexity of the human body's inner workings, it did not result in the expected advancement in the medical field. Therefore, the multi-omics approach has emerged. The multi-omics approach involves integrating data from different omics technologies, such as DNA sequencing, RNA sequencing, mass spectrometry, and others, using computational methods and then analyzing the integrated result for different downstream analysis applications such as survival analysis, cancer classification, or biomarker identification. Most of the recent reviews were constrained to discussing one aspect of the multi-omics analysis pipeline, such as the dimensionality reduction step, the integration methods, or the interpretability aspect; however, very few provide a comprehensive review of every step of the analysis. This study aims to give an overview of the multi-omics analysis pipeline, starting with the most popular multi-omics databases used in recent literature, dimensionality reduction techniques, details the different types of data integration techniques and their downstream analysis applications, describes the most commonly used evaluation metrics, highlights the importance of model interpretability, and lastly discusses the challenges and potential future work for multi-omics data integration in precision medicine.

Characterization of Volatile Organic Compounds and Aroma Sensory Properties in Yunnan Cigar

To characterize volatile organic compounds (VOCs) and aromatic sensory properties in Yun cigar, 27 samples from four origins were analyzed using SPME-HS-GC/MS and sensory analysis. The investigation results were analyzed using principal component analysis (PCA), Fisher linear discriminant analysis (LDA), and Pearson correlation analysis. In Yunnan cigars, the content of nicotine and neophytadiene accounted for over 90% of the total VOC content. Nicotine was significantly positively correlated with neophytadiene and phytol. The cigars from four origins were clearly classified by the PCA of VOCs. Four region discrimination functions were established through the LDA of 14 compounds, and the validation accuracy was 100%. The sensory descriptors with the highest geometric mean were woody, roasted, fresh-sweet, bean, and scorched. Acetophenone, megastigmatrienone A, and thunbergene were positively correlated with multiple aroma descriptors, while nicotine was negatively correlated with multiple aroma descriptors.

Predicting 30-Day and 1-Year Mortality in Heart Failure with Preserved Ejection Fraction (HFpEF)

Objectives

To develop and compare prediction models for 30-day and 1-year mortality in Heart failure with preserved ejection fraction (HFpEF) using EHR data, utilizing both traditional and machine learning (ML) techniques.

Background

HFpEF represents 1 in 2 heart failure patients. Predictive models in HFpEF, specifically those derived from electronic health record (EHR) data, are less established.

Methods

Using MIMIC-IV EHR data from 2008-2019, patients aged ≥ 18 years admitted with a primary diagnosis of HFpEF were identified using ICD-9 and 10 codes. Demographics, vital signs, prior diagnoses, and lab data were extracted. Data was partitioned into 80% training, 20% test sets. Prediction models from seven model classes (Support Vector Classifier (SVC), Logistic Regression, Lasso Regression, Elastic Net, Random Forest, Histogram-based Gradient Boosting Classifier (HGBC), and XGBoost) were developed using various imputation and oversampling techniques with 5-fold cross-validation. Model performance was compared using several metrics, and individual feature importance assessed using SHapley Additive exPlanations (SHAP) analysis.

Results

Among 3910 hospitalizations for HFpEF, 30-day mortality was 6.3%, and 1-year mortality was 29.2%. Logistic regression performed well for 30-day mortality (Area Under the Receiver operating characteristic curve (AUC) 0.83), whereas Random Forest (AUC 0.79) and HGBC (AUC 0.78) for 1-year mortality. Age and NT-proBNP were the strongest predictors in SHAP analyses for both outcomes.

Conclusion

Models derived from EHR data can predict mortality after HFpEF hospitalization with comparable performance to models derived from registry or trial data, highlighting the potential for clinical implementation.

Expanding drug targets for 112 chronic diseases using a machine learning-assisted genetic priority score

Identifying genetic drivers of chronic diseases is necessary for drug discovery. Here, we develop a machine learning-assisted genetic priority score, which we call ML-GPS, that incorporates genetic associations with predicted disease phenotypes to enhance target discovery. First, we construct gradient boosting models to predict 112 chronic disease phecodes in the UK Biobank and analyze associations of predicted and observed phenotypes with common, rare, and ultra-rare variants to model the allelic series. We integrate these associations with existing evidence using gradient boosting with continuous feature encoding to construct ML-GPS, training it to predict drug indications in Open Targets and externally testing it in SIDER. We then generate ML-GPS predictions for 2,362,636 gene-phecode pairs. We find that the use of predicted phenotypes, which identify substantially more genetic associations than observed phenotypes across the allele frequency spectrum, significantly improves the performance of ML-GPS. ML-GPS increases coverage of drug targets, with the top 1% of all scores providing support for 15,077 gene-phecode pairs that previously had no support. ML-GPS can also identify well-known target-disease relationships, promising targets without indicated drugs, and targets for several drugs in clinical trials, including LRRK2 inhibitors for Parkinson's disease and olpasiran for cardiovascular disease.

Proteo-genomic analyses in relatively lean Chinese adults identify proteins and pathways that affect general and central adiposity levels

Adiposity is an established risk factor for multiple diseases, but the causal relationships of different adiposity types with circulating protein biomarkers have not been systematically investigated. We examine the causal associations of general and central adiposity with 2923 plasma proteins among 3977 Chinese adults (mean BMI = 23.9 kg/m²). Genetically-predicted body mass index (BMI), body fat percentage (BF%), waist circumference (WC), and waist-to-hip ratio (WHR) are significantly (FDR < 0.05) associated with 399, 239, 436, and 283 proteins, respectively, with 80 proteins associated with all four and 275 with only one adiposity trait. WHR is associated with the most proteins (n = 90) after adjusting for other adiposity traits. These associations are largely replicated in Europeans (mean BMI = 27.4 kg/m²). Two-sample Mendelian randomisation (MR) analyses in East Asians using cis-protein quantitative trait locus (cis-pQTLs) identified in GWAS find 30/2 proteins significantly affect levels of BMI/WC, respectively, with 10 showing evidence of colocalisation, and seven (inter-alpha-trypsin inhibitor heavy chain H3, complement factor B, EGF-containing fibulin-like extracellular matrix protein 1, thioredoxin domain-containing protein 15, alpha-2-antiplasmin, fibronectin, mimecan) are replicated in separate MR using different cis-pQTLs identified in Europeans. These findings identified potential novel mechanisms and targets, to our knowledge, for improved treatment and prevention of obesity and associated diseases.

A novel DNA methylation-based surrogate biomarker for chronic systemic inflammation (InfLaMeS): results from the Health and Retirement Study

Chronic low-grade systemic inflammation is a risk factor for chronic diseases and mortality and is an important biomarker in health research. DNA methylation (DNAm) surrogate biomarkers are valuable exposure, risk factor and health outcome predictors in studies where the measures cannot be measured directly and often perform as well or better than direct measure. We generated a DNAm surrogate biomarker for chronic, systemic inflammation from a systemic inflammation latent variable of seven inflammatory markers and evaluated its performance relative to measured inflammatory biomarkers in predicting several age-associated outcomes of interest, including mortality, activities of daily living and multimorbidity in the Health and Retirement Study (HRS). The DNAm surrogate, Inflammation Latent Variable Methylation Surrogate (InfLaMeS), correlated with seven individual inflammation markers (r= -0.2-0.6) and performed as well or better to the systemic inflammation latent variable measure when predicting multimorbidity, disability, and 4-year mortality in HRS. Findings were validated in an external cohort, The Irish Longitudinal Study of Ageing. These results suggest that InfLaMeS provides a robust alternative to measured blood-chemistry measures of inflammation with broad applicability in instances where values of inflammatory markers are not measured but DNAm data is available.

Potential association between PSCA rs2976395 functional variant and pancreatic cancer risk

Correlated regions of systemic interindividual variation (CoRSIV) represent a small proportion of the human genome showing DNA methylation patterns that are the same in all human tissues, are different among individuals, and are partially regulated by genetic variants in cis. In this study we aimed at investigating single-nucleotide polymorphisms (SNPs) within CoRSIVs and their involvement with pancreatic ductal adenocarcinoma (PDAC) risk. We analyzed 29,099 CoRSIV-SNPs and 133,615 CoRSIV-mQTLs in 14,394 cases and 247,022 controls of European and Asian descent. We observed that the A allele of the rs2976395 SNP was associated with increased PDAC risk in Europeans (p = 2.81 × 10-5). This SNP lies in the prostate stem cell antigen gene and is in perfect linkage disequilibrium with a variant (rs2294008) that has been reported to be associated with risk of many other cancer types. The A allele is associated with the DNA methylation level of the gene according to the PanCan-meQTL database and with overexpression according to QTLbase. The expression of the gene has been observed to be deregulated in many tumors of the gastrointestinal tract including pancreatic cancer; however, functional studies are needed to elucidate the function relevance of the association.

Development of a predictive model for the U-shaped relationship between the triglyceride glycemic index and depression using machine learning (NHANES 2009-2018)

Background

At present, the relationship between depression and the triglyceride glycemic (TyG) index remains a topic of debate. This study sought to elucidate the relationship between depression and the TyG index to create a predictive model that would help doctors diagnose patients.

Methods

We conducted a cross-sectional study utilizing the National Health and Nutrition Examination Survey (NHANES) dataset, which comprises data from 2009 to 2018. The analysis involved 11,222 adults with a Patient Health Questionnaire-9 (PHQ-9) score of 5 or higher, indicating the presence of depression. As part of the analysis, multiple regression models were used to test whether a linear relationship existed between the TyG index and depression. A threshold effects analysis was used to generate smoothed curves and detect nonlinear correlations. Additionally, the Least Absolute Shrinkage and Selection Operator (LASSO) regression were employed to identify the key risk factors associated with depression. The factors identified were then used to construct the risk prediction nomogram. Finally, Receiver Operating Characteristic (ROC) curves were used to evaluate the discriminative performance of the model.

Results

Multivariable linear regression analysis indicated a strong positive correlation between depression and the TyG index (β: 0.38, 95 % CI: 0.16-0.60, p = 0.0008). A U-shaped relationship with an inflection point was observed at a TyG index of 8.16. The nomogram model, constructed using risk factors identified by LASSO, exhibited a significant predictive value (AUC = 0.888).

Conclusions

The results of this investigation point to a U-shaped association between depression risk and the TyG index among Americans. Those with a TyG index of over 8.16 are significantly more likely to develop depression. These results suggest a possible causal relationship and emphasize the importance of monitoring the TyG index in depression risk assessment.

Analysis of factors associated with public attitudes towards salt reduction: a multicountry cross-sectional survey

OBJECTIVE

This paper provides an in-depth examination of public attitudes towards salt reduction across seven culturally diverse countries: the USA, the UK, France, Japan, Indonesia, Thailand, and Brazil.

DESIGN

Cross-sectional regression analysis with questionnaire data.

SETTING

An analysis of questionnaire study in seven countries.

PARTICIPANTS

The study's questionnaire collected responses from 7090 participants across seven countries with the mean age of respondents being 46.06 years (SD 16.96). The gender distribution encompassed 3473 men (49.12%), 3582 women (50.66%), 24 non-binary individuals (0.34%) and 11 who identified as 'other' (0.16%).

PRIMARY AND SECONDARY OUTCOME MEASURES

Attitudes toward sodium reduction were measured on a seven-point Likert scale.

RESULTS

Regression analysis revealed significant associations between attitudes towards sodium reduction and various factors across countries. Gender was a significant factor in France, with women showing less awareness than men (coefficient -0.123, 95% CI -0.237 to -0.008). Age was a significant factor in Japan and Thailand, with older generations exhibiting stronger awareness. Occupation was a significant factor in France (grocery, 0.678, 0.229 to 1.127) and Japan (food service, 0.792, 0.300 to 1.283). In France (0.090, 0.033 to 0.146) and Brazil (0.092, 0.040 to 0.144), attitudes towards reducing sugar intake were positively associated with sodium reduction attitudes. Government interventions showed varying impacts, with positive associations in Thailand (0.004, 0.001 to 0.008) and negative associations in France (-0.003 -0.005 to -0.000).

CONCLUSION

Our study reveals a complex array of factors shaping attitudes towards sodium reduction across seven countries. These findings support the need for nuanced, country-specific approaches in formulating sodium reduction strategies. Future research should validate these findings, explore further determinants and understand how attitudes translate into dietary behaviours.

Air pollution's numerical, spatial, and temporal heterogeneous impacts on childhood hand, foot and mouth disease: a multi-model county-level study from China

BACKGROUND

While stationary links between childhood hand, foot and mouth disease (HFMD) and air pollution are known, a comprehensive study on their heterogeneous relationships (nonstationarity), jointly considering numerical, temporal and spatial dimensions, has not been reported.

METHODS

Monthly HFMD incidence and air pollution data were collected at the county level from Sichuan-Chongqing, China (2009-2011), alongside meteorological and social environmental covariates. Key influential factors were identified using random forest (RF) under the stationary assumption. Factors' numerically, temporally, and spatially heterogeneous relationships with HFMD were assessed using generalized additive model (GAM) and geographically and temporally weighted regression (GTWR).

RESULTS

Our findings highlighted the relatively higher stationary contributions of fine particulate matter (PM2.5) and ozone (O3) to HFMD incidence across Sichuan-Chongqing counties. We further uncovered heterogeneous impacts of PM2.5 and O3 from three nonstationary perspectives. Numerically, PM2.5 showed an inverse 'V'-shaped relationship with HFMD incidence, while O3 exhibited a complex pattern, with increased HFMD incidence at low PM2.5 and moderate O3 concentrations. Temporally, PM2.5's impact peaked in autumn and was weakest in spring, whereas O3's effect was strongest in summer. Spatially, hotspot mapping revealed high-risk clusters for PM2.5 impact across all seasons, with notable geographical variations, and for O3 in spring, summer, and autumn, concentrated in specific regions of Sichuan-Chongqing.

CONCLUSIONS

This study underscores the nuanced and three-perspective heterogeneous influences of air pollution on HFMD in small areas, emphasizing the need for differentiated, localized, and time-sensitive prevention and control strategies to enhance the precision of dynamic early warnings and predictive models for HFMD and other infectious diseases, particularly in the fields of environmental and spatial epidemiology.

Developmental patterns of inhibition and fronto-basal-ganglia white matter organisation in healthy children and children with attention-deficit/hyperactivity disorder

There is robust evidence implicating inhibitory deficits as a fundamental behavioural phenotype in children with attention-deficit/hyperactivity disorder (ADHD). However, prior studies have not directly investigated the role in which white matter properties within the fronto-basal-ganglia circuit may play in the development of inhibitory control deficits in this group. Combining recent advancements in brain-behavioural modelling, we mapped the development of stop-signal task (SST) performance and fronto-basal-ganglia maturation in a longitudinal sample of children aged 9-14 with and without ADHD. In a large sample of 135 ADHD and 138 non-ADHD children, we found that the ADHD group had poorer inhibitory control (i.e., longer stop-signal reaction times) across age compared to non-ADHD controls. When applying the novel parametric race model, this group effect was driven by higher within-subject variability (sigma) and higher number of extreme responses (tau) on stop trials. The ADHD group also displayed higher within-subject variability on correct responses to go stimuli. Moreover, we observed the ADHD group committing more task-based failures such as responding on stop trials (trigger failures) and omissions on go trials (go failures) compared to non-ADHD controls, suggesting the contribution of attentional lapses to poorer response inhibition performance. In contrast, longitudinal modelling of fixel-based analysis measures revealed no significant group differences in the maturation of fronto-basal-ganglia fibre cross-section in a subsample (74 ADHD and 73 non-ADHD children). Finally, brain-behavioural models revealed that age-related changes in fronto-basal-ganglia morphology (fibre cross-section) were significantly associated with reductions in the variability of the correct go-trial responses (sigma.true) and skew of the stop-trial distribution (tauS). However, this effect did not differ between ADHD and typically developing children. Overall, our findings support the growing consensus suggesting that attentional deficits subserve ADHD-related inhibitory dysfunction. Furthermore, we show novel evidence suggesting that while children with ADHD are consistently performing worse on the SST than their non-affected peers, they appear to have comparable rates of neurocognitive maturation across this period.

Development of novel computational models based on artificial intelligence technique to predict liquids mixtures separation via vacuum membrane distillation

The fundamental objective of this paper is to use Machine Learning (ML) methods for building models on temperature (T) prediction using input features r and z for a membrane separation process. A hybrid model was developed based on computational fluid dynamics (CFD) to simulate the separation process and integrate the results into machine learning models. The CFD simulations were performed to estimate temperature distribution in a vacuum membrane distillation (VMD) process for separation of liquid mixtures. The evaluated ML models include Support Vector Machine (SVM), Elastic Net Regression (ENR), Extremely Randomized Trees (ERT), and Bayesian Ridge Regression (BRR). Performance was improved using Differential Evolution (DE) for hyper-parameter tuning, and model validation was performed using Monte Carlo Cross-Validation. The results clearly indicated the models' effectiveness in temperature prediction, with SVM outperforming other models in terms of accuracy. The SVM model had a mean R2 value of 0.9969 and a standard deviation of 0.0001, indicating a strong and consistent fit to the membrane data. Furthermore, it exhibited the lowest mean squared error, mean absolute error, and mean absolute percentage error, signifying superior predictive accuracy and reliability. These outcomes highlight the importance of selecting a suitable model and optimizing hyperparameters to guarantee accurate predictions in ML tasks. It demonstrates that using SVM, optimized with DE improves accuracy and consistency for this specific predictive task in membrane separation context.

Machine vision based frailty assessment for genetically diverse mice

Frailty indexes (FIs) capture health status in humans and model organisms. To accelerate our understanding of biological aging and carry out scalable interventional studies, high-throughput approaches are necessary. We previously introduced a machine vision-based visual frailty index (vFI) that uses mouse behavior in the open field to assess frailty using C57BL/6J (B6J) data. Aging trajectories are highly genetic and are frequently modeled in genetically diverse animals. In order to extend the vFI to genetically diverse mouse populations, we collect frailty and behavior data on a large cohort of aged Diversity Outbred (DO) mice. Combined with previous data, this represents one of the largest video-based aging behavior datasets to date. Using these data, we build accurate predictive models of frailty, chronological age, and even the proportion of life lived. The extension of automated and objective frailty assessment tools to genetically diverse mice will enable better modeling of aging mechanisms and enable high-throughput interventional aging studies.

Associations of "Weekend Warrior" Physical Activity With Incident Disease and Cardiometabolic Health

BACKGROUND

Achievement of guideline-recommended levels of physical activity (≥150 minutes of moderate-to-vigorous physical activity per week) is associated with lower risk of adverse cardiovascular events and represents an important public health priority. Although physical activity commonly follows a "weekend warrior" pattern, in which most moderate-to-vigorous physical activity is concentrated in 1 or 2 days rather than spread more evenly across the week (regular), the effects of physical activity pattern across a range of incident diseases, including cardiometabolic conditions, are unknown.

METHODS

We tested associations between physical activity pattern and incidence of 678 conditions in 89 573 participants (62±8 years of age; 56% women) of the UK Biobank prospective cohort study who wore an accelerometer for 1 week between June 2013 and December 2015. Models were adjusted for multiple baseline clinical factors, and P value thresholds were corrected for multiplicity.

RESULTS

When compared to inactive (<150 minutes moderate-to-vigorous physical activity/week), both weekend warrior (267 total associations; 264 [99%] with lower disease risk; hazard ratio [HR] range, 0.35-0.89) and regular activity (209 associations; 205 [98%] with lower disease risk; HR range, 0.41-0.88) were broadly associated with lower risk of incident disease. The strongest associations were observed for cardiometabolic conditions such as incident hypertension (weekend warrior: HR, 0.77 [95% CI, 0.73-0.80]; P=1.2×10-27; regular: HR, 0.72 [95% CI, 0.68-0.77]; P=4.5×10-28), diabetes (weekend warrior: HR, 0.57 [95% CI, 0.51-0.62]; P=3.9×10-32; regular: HR, 0.54 [95% CI, 0.48-0.60]; P=8.7×10-26), obesity (weekend warrior: HR, 0.55 [95% CI, 0.50-0.60]; P=2.4×10-43, regular: HR, 0.44 [95% CI, 0.40-0.50]; P=9.6×10-47), and sleep apnea (weekend warrior: HR, 0.57 [95% CI, 0.48-0.69]; P=1.6×10-9; regular: HR, 0.49 [95% CI, 0.39-0.62]; P=7.4×10-10). When weekend warrior and regular activity were compared directly, there were no conditions for which effects differed significantly. Observations were similar when activity was thresholded at the sample median (≥230.4 minutes of moderate-to-vigorous physical activity/week).

CONCLUSIONS

Achievement of measured physical activity volumes consistent with guideline recommendations is associated with lower risk for >200 diseases, with prominent effects on cardiometabolic conditions. Associations appear similar whether physical activity follows a weekend warrior pattern or is spread more evenly throughout the week.

Pathologic subtyping of Alzheimer's disease brain tissue reveals disease heterogeneity

In recent years, multiple groups have shown that what is currently thought of as "Alzheimer's Disease" (AD) may be usefully viewed as several related disease subtypes. As these efforts have continued, a related issue is how common co-pathologies and ethnicity intersect with AD subtypes. The goal of this study was to use a dataset constituting 153 pathologic variables recorded on 666 AD brain autopsies to better define how co-pathologies and ethnicity relate to established AD subtypes. Pathologic clustering suggests 8 subtypes within this cohort, and further analysis reveals that the previously described continuum from hippocampal predominant to hippocampal sparing is well represented in our data. Small vessel disease is overall highest in a cluster with a low hippocampal/cortical tau ratio, and across all clusters small vessel disease segregates separately from Lewy body disease. Two AD clusters are identified with extensive Lewy bodies outside amygdala (one with a high hippocampal/cortical tau ratio and one with a low ratio), and we find an inverse relationship between cortical tau and Lewy body pathology across these two clusters. Finally, we find that brains from persons of Hispanic descent have significantly more AD pathology in multiple neuroanatomic areas. We find that Hispanic ethnicity is not uniformly distributed across clusters, and this is particularly pronounced in clusters with significant Lewy body pathology, where Hispanic donors are only found in a cluster with a low hippocampal/cortical tau ratio. In summary, our analysis of recorded pathologic data across two decades of banked brains reveals new relationships in the patterns of AD-related proteinopathy, co-pathology, and ethnicity, and highlights the utility of pathologic subtyping to classify AD pathology.

Prediction of Treatment Recommendations Via Ensemble Machine Learning Algorithms for Non-Small Cell Lung Cancer Patients in Personalized Medicine

Objectives

The primary goal of this research is to develop treatment-related genomic predictive markers for non-small cell lung cancer by integrating various machine learning algorithms that recommends near-optimal individualized patient treatment for chemotherapy in an effort to maximize efficacy or minimize treatment-related toxicity. This research can contribute toward developing a more refined, accurate and effective therapy accounting for specific patient needs.

Methods

To accomplish our research goal, we implement ensemble learning algorithms, bagging with regularized Cox regression models and nonparametric tree-based models via Random Survival Forests. A comprehensive meta-database was compiled from the NCBI Gene Expression Omnibus data repository for lung cancer patients to capture and utilize complex genomic patterns that can predict treatment outcomes more accurately.

Results

The developed novel prediction algorithm demonstrates the ability to support complex clinical decision-making processes in the treatment of NSCLC. It effectively addresses patient heterogeneity, offering predictions that are both refined and personalized in improving the precision of chemotherapy regimens prescribed to the eligible patients.

Conclusion

This research should contribute substantial advancement of cancer treatments by improving the accuracy and efficacy of chemotherapy treatments for a targeted group of patients who need the right treatment. The integration of complex machine learning techniques with genomic data holds substantial potential to transform current cancer treatment paradigms by providing robust support in clinical decision-making.

Quantifying the effects of the mitochondrial genome on milk production traits in dairy cows: empirical results and modelling challenges

Significant advances in livestock traits have been achieved primarily through selection strategies targeting variation in the nuclear genome, with little attention given to mitogenome variation. We analyzed the influence of the mitogenome on milk production traits of Holstein cattle in Croatia based on strategically generated next-generation sequencing data for 109 cows pedigree-linked to 7115 milk production records (milk, fat and protein yield) from 3006 cows (first 5 lactations). Since little is known about the biology of the relationship between mitogenome variation and production traits, our quantitative genetic modeling was complex. Thus, the proportion of total variance explained by mitogenome inheritance was estimated using 5 different models: (1) cytoplasmic model with maternal lineages (CYTO), (2) haplotypic model with mitogenome sequences (HAPLO), (3) amino acid model with unique amino acid sequences (AMINO), (4) evolutionary model based on a phylogenetic analysis using Bayesian Evolutionary Analysis Sampling Trees phylogenetic analysis (EVOL), and (5) mitogenome SNP model (SNPmt). The polygenic autosomal and X chromosome additive genetic effects based on pedigree were modeled, together with the effects of herd-year-season interaction, permanent environment, location, and age at first calving. The estimated proportions of phenotypic variance explained by mitogenome in 4 different models (CYTO, HAPLO, AMINO, and SNPmt) were found to be substantial given the size of mitogenome, ranging from 5% to 7% for all 3 milk traits. At the same time, a negligible proportion of the phenotypic variance was explained by mitogenome with the EVOL model. Similarly, in all models, no proportion of phenotypic variance was explained by the X chromosome. Although our results should be confirmed in other dairy cattle populations, including a large number of sequenced mitogenomes and nuclear genomes, the potential of utilizing mitogenome information in animal breeding is promising, especially as the acquisition of complete genome sequences becomes cost-effective.

Deconstructing the geography of human impacts on species' natural distribution

It remains unknown how species' populations across their geographic range are constrained by multiple coincident natural and anthropogenic environmental gradients. Conservation actions are likely undermined without this knowledge because the relative importance of the multiple anthropogenic threats is not set within the context of the natural determinants of species' distributions. We introduce the concept of a species 'shadow distribution' to address this knowledge gap, using explainable artificial intelligence to deconstruct the environmental building blocks of current species distributions. We assess shadow distributions for multiple threatened freshwater fishes in Switzerland which indicated how and where species respond negatively to threats - with negative threat impacts covering 88% of locations inside species' environmental niches leading to a 25% reduction in environmental suitability. Our findings highlight that conservation of species' geographic distributions is likely insufficient when biodiversity mapping is based on species distribution models, or threat mapping, without also quantifying species' expected or shadow distributions. Overall, we show how priority actions for nature's recovery can be identified and contextualised within the multiple natural constraints on biodiversity to better meet national and international biodiversity targets.

Identification of an additive interaction using parameter regularization and model selection in epidemiology

Background

In epidemiology, indicators such as the relative excess risk due to interaction (RERI), attributable proportion (AP), and synergy index (S) are commonly used to assess additive interactions between two variables. However, the results of these indicators are sometimes inconsistent in real world applications and it may be difficult to draw conclusions from them.

Method

Based on the relationship between the RERI, AP, and S, we propose a method with consistent results, which are achieved by constraining e θ 3 - e θ 1 - e θ 2 + 1 = 0 , and the interpretation of the results is simple and clear. We present two pathways to achieve this end: one is to complete the constraint by adding a regular penalty term to the model likelihood function; the other is to use model selection.

Result

Using simulated and real data, our proposed methods effectively identified additive interactions and proved to be applicable to real-world data. Simulations were used to evaluate the performance of the methods in scenarios with and without additive interactions. The penalty term converged to 0 with increasing λ, and the final models matched the expected interaction status, demonstrating that regularized estimation could effectively identify additive interactions. Model selection was compared with classical methods (delta and bootstrap) across various scenarios with different interaction strengths, and the additive interactions were closely observed and the results aligned closely with bootstrap results. The coefficients in the model without interaction adhered to a simplifying equation, reinforcing that there was no significant interaction between smoking and alcohol use on oral cancer risk.

Conclusion

In summary, the model selection method based on the Hannan-Quinn criterion (HQ) appears to be a competitive alternative to the bootstrap method for identifying additive interactions. Furthermore, when using RERI, AP, and S to assess the additive interaction, the results are more consistent and the results are simple and easy to understand.

Identification of novel biomarkers for atherosclerosis using single-cell RNA sequencing and machine learning

Atherosclerosis (AS) is a predominant etiological factor in numerous cardiovascular diseases, with its associated complications such as myocardial infarction and stroke serving as major contributors to worldwide mortality rates. Here, we devised dependable AS-related biomarkers through the utilization of single-cell RNA sequencing, weighted co-expression network (WGCNA), and differential expression analysis. Furthermore, we employed various machine learning techniques (LASSO and SVM-RFE) to enhance the identification of AS biomarkers, subsequently validating them using the GEO dataset. Following this, CIBERSORT was employed to investigate the correlation between biomarkers and infiltrating immune cells. Consequently, 256 differentially expressed genes (DEGs) were selected in samples of AS and normal. GO and KEGG analyses indicated that these DEGs may be related to the negative regulation of leukocyte-mediated immunity, leukocyte cell-cell adhesion, and immune system processes. Notably, C1QC and COL1A1 were pinpointed as potential diagnostic markers for AS, a finding that was further validated in the GSE21545 dataset. Moreover, the area under the curve (AUC) values for these markers exceeded 0.8, underscoring their diagnostic utility. Analysis of immune cell infiltration revealed that the expression of C1QC was correlated with M0 macrophages, gamma delta T cells, activated mast cells and memory B cells. Similarly, COL1A1 expression was linked to M0 macrophages, memory B cells, activated mast cells, gamma delta T cells, and CD4 native T cells. Finally, these results were validated using mice and human samples through immunofluorescence, immunohistochemistry, and ELISA analysis. Overall, C1QC and COL1A1 would be potential biomarkers for AS diagnosis, and that would provides novel perspectives on the diagnosis and treatment of AS.

Machine learning classification of archaea and bacteria identifies novel predictive genomic features

BACKGROUND

Archaea and Bacteria are distinct domains of life that are adapted to a variety of ecological niches. Several genome-based methods have been developed for their accurate classification, yet many aspects of the specific genomic features that determine these differences are not fully understood. In this study, we used publicly available whole-genome sequences from bacteria ( N = 2546 ) and archaea ( N = 109 ). From these, a set of genomic features (nucleotide frequencies and proportions, coding sequences (CDS), non-coding, ribosomal and transfer RNA genes (ncRNA, rRNA, tRNA), Chargaff's, topological entropy and Shannon's entropy scores) was extracted and used as input data to develop machine learning models for the classification of archaea and bacteria.

RESULTS

The classification accuracy ranged from 0.993 (Random Forest) to 0.998 (Neural Networks). Over the four models, only 11 examples were misclassified, especially those belonging to the minority class (Archaea). From variable importance, tRNA topological and Shannon's entropy, nucleotide frequencies in tRNA, rRNA and ncRNA, CDS, tRNA and rRNA Chargaff's scores have emerged as the top discriminating factors. In particular, tRNA entropy (both topological and Shannon's) was the most important genomic feature for classification, pointing at the complex interactions between the genetic code, tRNAs and the translational machinery.

CONCLUSIONS

tRNA, rRNA and ncRNA genes emerged as the key genomic elements that underpin the classification of archaea and bacteria. In particular, higher nucleotide diversity was found in tRNA from bacteria compared to archaea. The analysis of the few classification errors reflects the complex phylogenetic relationships between bacteria, archaea and eukaryotes.

Cell-Type Resolved Protein Atlas of Brain Lysosomes Identifies SLC45A1-Associated Disease as a Lysosomal Disorder

Mutations in lysosomal genes cause neurodegeneration and neurological lysosomal storage disorders (LSDs). Despite their essential role in brain homeostasis, the cell-type-specific composition and function of lysosomes remain poorly understood. Here, we report a quantitative protein atlas of the lysosome from mouse neurons, astrocytes, oligodendrocytes, and microglia. We identify dozens of novel lysosomal proteins and reveal the diversity of the lysosomal composition across brain cell types. Notably, we discovered SLC45A1, mutations in which cause a monogenic neurological disease, as a neuron-specific lysosomal protein. Loss of SLC45A1 causes lysosomal dysfunction in vitro and in vivo. Mechanistically, SLC45A1 plays a dual role in lysosomal sugar transport and stabilization of V1 subunits of the V-ATPase. SLC45A1 deficiency depletes the V1 subunits, elevates lysosomal pH, and disrupts iron homeostasis causing mitochondrial dysfunction. Altogether, our work redefines SLC45A1-associated disease as a LSD and establishes a comprehensive map to study lysosome biology at cell-type resolution in the brain and its implications for neurodegeneration.

DNA methylation clocks struggle to distinguish inflammaging from healthy aging, but feature rectification improves coherence and enhances detection of inflammaging

Biological age estimation from DNA methylation and determination of relevant biomarkers is an active research problem which has predominantly been tackled with black-box penalized regression. Machine learning is used to select a small subset of features from hundreds of thousands CpG probes and to increase generalizability typically lacking with ordinary least-squares regression. Here, we show that such feature selection lacks biological interpretability and relevance in the clocks of the first- and next-generations, and clarify the logic by which these clocks systematically exclude biomarkers of aging and disease. Moreover, in contrast to the assumption that regularized linear regression is needed to prevent overfitting, we demonstrate that hypothesis-driven selection of biologically relevant features in conjunction with ordinary least squares regression yields accurate, well-calibrated, generalizable clocks with high interpretability. We further demonstrate that the interplay of disease-related shifts of predictor values and their corresponding weights, which we term feature shifts, contributes to the lack of resolution between health and disease in conventional linear models. Lastly, we introduce a method of feature rectification, which aligns these shifts to improve the distinction of age predictions for healthy people vs. patients with various diseases.

Key Findings

There is no apparent biological significance of the CpGs selected by first- and next-generation clocksThe range of residuals for first- and next-generation clock predications on healthy samples is very large; for all models tested, a prediction error of +/-10-20 years is within the 95% range of variation for healthy controls and does not signify age accelerationThere is no significant shift in the mean of residuals for patient populations relative to healthy populations for most studied first- and next-generation clocks. For those with significance, the effect size is very small.Hypothesis-driven feature pre-selection, coupled with modified forward step-wise selection yields age predictors on par with first and next-generation clocks. EN/ML is not needed.Disease-related shifts at different CpG probes, along with learned model weights, can be either positive or negative; their combination leads to de-coherence effect in linear models.Model coherence can be induced by rectifying features to have only positive shifts in patient samples; this provides a better resolution between health and disease in DNAm age models, and expectedly, introduces more non-linearity to the input data.

Limited Biomarker Potential for IgG Autoantibodies Reactive to Linear Epitopes in Systemic Lupus Erythematosus or Spondyloarthropathy

BACKGROUND

Autoantibodies are commonly used as biomarkers in autoimmune diseases, but there are limitations. For example, autoantibody biomarkers have poor sensitivity or specificity in systemic lupus erythematosus and do not exist in the spondyloarthropathies, impairing diagnosis and treatment. While autoantibodies suitable for strong biomarkers may not exist in these conditions, another possibility is that technology has limited their discovery. The purpose of this study was to use a novel high-density peptide array that enables the evaluation of IgG binding to every possible linear antigen in the entire human peptidome, as well as a novel machine learning approach that incorporates ELISA validation predictability in order to discover autoantibodies that could be developed into sensitive and specific markers of lupus or spondyloarthropathy.

METHODS

We used a peptide array containing the human peptidome, several viral peptidomes, and key post-translational modifications (6 million peptides) to quantify IgG binding in lupus, spondyloarthropathy, rheumatoid arthritis, Sjögren's disease, and control sera. Using ELISA data for 70 peptides, we performed a random forest analysis that evaluated multiple array features to predict which peptides might be good biomarkers, as confirmed by ELISA. We validated the peptide prediction methodology in rheumatoid arthritis and COVID-19, conditions for which the antibody repertoire is well-understood, and then evaluated IgG binding by ELISA to peptides that we predicted would be highly bound specifically in lupus or spondyloarthropathy.

RESULTS

Our methodology performed well in validation studies, but peptides predicted to be highly and specifically bound in lupus or spondyloarthropathy could not be confirmed by ELISA.

CONCLUSIONS

In a comprehensive evaluation of the entire human peptidome, highly sensitive and specific IgG autoantibodies were not identified in lupus or spondyloarthropathy. Thus, the pathogenesis of lupus and spondyloarthropathy may not depend upon unique autoantigens, and other types of molecules should be sought as optimal biomarkers in these conditions.

Conditional modeling of recurrent event data with terminal event

Recurrent event data with a terminal event arise in follow-up studies. The current literature has primarily focused on the effect of covariates on the recurrent event process using marginal estimating equation approaches or joint modeling approaches via frailties. In this article, we propose a conditional model for recurrent event data with a terminal event, which provides an intuitive interpretation of the effect of the terminal event: at an early time, the rate of recurrent events is nearly independent of the terminal event, but the dependence gets stronger as time goes close to the terminal event time. A two-stage likelihood-based approach is proposed to estimate parameters of interest. Asymptotic properties of the estimators are established. The finite-sample behavior of the proposed method is examined through simulation studies. A real data of colorectal cancer is analyzed by the proposed method for illustration.

Tandem mobilization of anti-phage defenses alongside SCCmec elements in staphylococci

Recent research has identified multiple immune systems that bacteria use to protect themselves from viral infections. However, little is known about the mechanisms by which these systems horizontally spread, especially among bacterial pathogens. Here, we investigate antiviral defenses in staphylococci, opportunistic pathogens that constitute leading causes of antibiotic-resistant infections. We show that these organisms harbor a variety of anti-phage defenses encoded within or near SCC (staphylococcal cassette chromosome) mec cassettes, mobile genomic islands that confer methicillin resistance. Importantly, we demonstrate that SCCmec-encoded recombinases mobilize not only SCCmec, but also tandem SCC-like cassettes enriched in genes coding for diverse defense systems. Further, we show that phage infection stimulates cassette mobilization (i.e. excision and circularization). Thus, our findings indicate that SCC/SCCmec cassettes not only spread antibiotic resistance but can also play a role in mobilizing anti-phage defenses.

Comparison of cyclosporine and tacrolimus after liver transplantation for primary biliary cholangitis: A propensity score-matched intention-to-treat registry study

The optimal calcineurin inhibitor after liver transplantation (LT) for primary biliary cholangitis (PBC) remains debated. We compared tacrolimus with cyclosporine in a propensity score-matched intention-to-treat analysis from the Scientific Registry of Transplant Recipients. We included adults with PBC who underwent primary LT from 1995 to 2022. Patients with initial cyclosporine treatment were 1:3 matched with those with initial tacrolimus treatment, ensuring exact calendar-period match. Primary outcomes were patient and graft survival. After matching, 579 patients with PBC and initial cyclosporine and 1348 with tacrolimus were well balanced for baseline characteristics. During a median follow-up of 11.1 years, 1044 (54%) deaths and 124 (6%) re-LTs occurred. In the overall matched sample, no significant survival difference emerged between cyclosporine and tacrolimus. However, tacrolimus conferred a survival advantage in some secondary analysis, such as LT after year 2000 and women, and in a 6-month landmark analysis excluding early postoperative events and calcineurin inhibitor switches. Cyclosporine did not reduce graft loss from PBC recurrence or affect laboratory markers of recurrence. In conclusion, we found no benefit of starting immunosuppression with cyclosporine after LT for PBC.

Temperature and leaf form drive contrasting sensitivity to nitrogen deposition across European forests

Raised emissions of biologically reactive nitrogen (N) have intensified N deposition, enhancing tree productivity globally. Nonetheless, the drivers of forest sensitivity to N deposition remain unknown. We used stem growth data from 62,000 trees across Europe combined with N deposition data to track the effects of air temperature and precipitation on tree growth's sensitivity to N deposition and how it varied depending on leaf form over the past 30 years. Overall, N deposition enhanced conifer growth (until 30 kg N ha-1 yr-1) while decreasing growth for broadleaved angiosperms. Lower temperatures led to higher growth sensitivity to N deposition in conifers potentially exacerbated by N limitation. In contrast, higher temperatures stimulated growth sensitivity to N deposition for broadleaves. Higher precipitation equally increased N deposition sensitivity in all leaf forms. We conclude that air temperature and leaf form are decisive in disentangling the effect of N deposition in European forests, which provides crucial information to better predict the contribution of N deposition to land carbon sink enhancement.

Spatial variability and uncertainty associated with soil moisture content using INLA-SPDE combined with PyMC3 probability programming

Spatial variability and uncertainty associated with soil volumetric moisture content (SVMC) is crucial in moisture prediction accuracy, this paper sets out to address this point of SVMC by developing data-driven model. Grid samples of SVMC covered approximately a 3-ha field during the jointing growth stage of winter wheat, and SVMC were measured by Time Domain Reflectometry (TDR), located in North China Plain, China. Bayesian inference was performed to explore spatial heterogeneity, robustness, transparency, interpretability and uncertainty related to SVMC using python-based PyMC3 combined with Integrated Nested Laplace Approximation with the Stochastic Partial Differential Equation (INLA-SPDE) model. The results showed that the prediction surface of SVMC, the lower and upper limits of 95% credible intervals quantified uncertainty associated with SVMC, cauchy prior of the flexibility and adaptability to obtain state-of-the-art predictive performance is more robust than gaussian prior for SVMC prediction, the transparency and interpretability of SVMC prediction model were revealed by MCMC (Markov-Chain Monte-Carlo) trace plots, KDE (Kernel density estimates), and rank plots. The uncertainty associated with SVMC can explicitly be described using the highest-posterior density interval, the prediction lower and upper limits.

Association of delayed cord clamping with acute kidney injury and two-year kidney outcomes in extremely premature neonates: a secondary analysis of the preterm erythropoietin neuroprotection trial (PENUT)

BACKGROUND

Delayed cord clamping (DCC) occurs in most preterm births.

OBJECTIVE

Evaluate the association of DCC with acute kidney injury (AKI) and two-year kidney outcomes.

METHODS

Secondary analysis of the Preterm Erythropoietin Neuroprotection Trial of neonates born 240/7 to 276/7 weeks gestation. AKI and two year kidney outcomes were compared in neonates with DCC ( ≥ 30 s after delivery) to those with early cord clamping (ECC) (<30 s after delivery).

RESULTS

The incidence and severity of AKI did not differ between the DCC and ECC groups (aOR 1.17 [95%CI 0.76-1.80]). At two years corrected age, DCC was associated with a 4.5-fold increased adjusted odds of estimated glomerular filtration rate (eGFR) <90 mL/min/1.73m2. No significant associations were noted between DCC and albuminuria or elevated blood pressure.

CONCLUSIONS

DCC was not associated with decreased neonatal AKI, but was associated with higher adjusted odds of eGFR <90 mL/min/1.73m2 at two years.

Integrative multi-omic and machine learning approach for prognostic stratification and therapeutic targeting in lung squamous cell carcinoma

The proliferation, metastasis, and drug resistance of cancer cells pose significant challenges to the treatment of lung squamous cell carcinoma (LUSC). However, there is a lack of optimal predictive models that can accurately forecast patient prognosis and guide the selection of targeted therapies. The extensive multi-omic data obtained from multi-level molecular biology provides a unique perspective for understanding the underlying biological characteristics of cancer, offering potential prognostic indicators and drug sensitivity biomarkers for LUSC patients. We integrated diverse datasets encompassing gene expression, DNA methylation, genomic mutations, and clinical data from LUSC patients to achieve consensus clustering using a suite of 10 multi-omics integration algorithms. Subsequently, we employed 10 commonly used machine learning algorithms, combining them into 101 unique configurations to design an optimal performing model. We then explored the characteristics of high- and low-risk LUSC patient groups in terms of the tumor microenvironment and response to immunotherapy, ultimately validating the functional roles of the model genes through in vitro experiments. Through the application of 10 clustering algorithms, we identified two prognostically relevant subtypes, with CS1 exhibiting a more favorable prognosis. We then constructed a subtype-specific machine learning model, LUSC multi-omics signature (LMS) based on seven key hub genes. Compared to previously published LUSC biomarkers, our LMS score demonstrated superior predictive performance. Patients with lower LMS scores had higher overall survival rates and better responses to immunotherapy. Notably, the high LMS group was more inclined toward "cold" tumors, characterized by immune suppression and exclusion, but drugs like dasatinib may represent promising therapeutic options for these patients. Notably, we also validated the model gene SERPINB13 through cell experiments, confirming its role as a potential oncogene influencing the progression of LUSC and as a promising therapeutic target. Our research provides new insights into refining the molecular classification of LUSC and further optimizing immunotherapy strategies.

Spatial Cellular Networks from omics data with SpaCeNet

Advances in omics technologies have allowed spatially resolved molecular profiling of single cells, providing a window not only into the diversity and distribution of cell types within a tissue, but also into the effects of interactions between cells in shaping the transcriptional landscape. Cells send chemical and mechanical signals which are received by other cells, where they can subsequently initiate context-specific gene regulatory responses. These interactions and their responses shape the individual molecular phenotype of a cell in a given microenvironment. RNAs or proteins measured in individual cells, together with the cells' spatial distribution, provide invaluable information about these mechanisms and the regulation of genes beyond processes occurring independently in each individual cell. "SpaCeNet" is a method designed to elucidate both the intracellular molecular networks (how molecular variables affect each other within the cell) and the intercellular molecular networks (how cells affect molecular variables in their neighbors). This is achieved by estimating conditional independence (CI) relations between captured variables within individual cells and by disentangling these from CI relations between variables of different cells.

Combined effects of herbicides and insecticides reduce biomass of sensitive aquatic invertebrates

The structure and biomass of aquatic invertebrate communities play a crucial role in the matter dynamics of streams. However, biomass is rarely quantified in ecological assessments of streams, and little is known about the environmental and anthropogenic factors that influence it. In this study, we aimed to identify environmental factors that are associated with invertebrate structure and biomass through a monitoring of 25 streams across Germany. We identified invertebrates, assigned them to taxonomic and trait-based groups, and quantified biomass using image-based analysis. We found that insecticide pressure generally reduced the abundance of insecticide-vulnerable populations (R2 = 0.43 applying SPEARpesticides indicator), but not invertebrate biomass. In contrast, herbicide pressure reduced the biomass of several biomass aggregations. Especially, insecticide-sensitive populations, that were directly (algae feeder, R2 = 0.39) or indirectly (predators, R2 = 0.29) dependent on algae, were affected. This indicated a combined effect of possible food shortage due to herbicides and direct insecticide pressure. Specifically, all streams with increased herbicide pressure showed a reduced overall biomass share of Trichoptera from 43 % to 3 % and those of Ephemeroptera from 20 % to 3 % compared to streams grouped by low herbicide pressure. In contrast, insecticide-insensitive Gastropoda increased from 10 % to 45 %, and non-vulnerable leaf-shredding Crustacea increased from 10 % to 22 %. In summary, our results indicate that at the community level, the direct effects of insecticides and the indirect, food-mediated effects of herbicides exert a combined effect on the biomass of sensitive insect groups, thus disrupting food chains at ecosystem level.