Missing value estimation methods for DNA microarrays

drexml: A command line tool and Python package for drug repurposing

We introduce drexml, a command line tool and Python package for rational data-driven drug repurposing. The package employs machine learning and mechanistic signal transduction modeling to identify drug targets capable of regulating a particular disease. In addition, it employs explainability tools to contextualize potential drug targets within the functional landscape of the disease. The methodology is validated in Fanconi Anemia and Familial Melanoma, two distinct rare diseases where there is a pressing need for solutions. In the Fanconi Anemia case, the model successfully predicts previously validated repurposed drugs, while in the Familial Melanoma case, it identifies a promising set of drugs for further investigation.

Exploration of Variables Predicting Sense of School Belonging Using the Machine Learning Method-Group Mnet

The purpose of this study was to explore variables related to school belonging from a holistic perspective, including a large number of variables in one model, different to the traditional analytical method. Using 2015 data from the Program for International Student Assessment (PISA), we sought to identify variables related to school belonging by searching for hundreds of predictors in one model using the group Mnet machine learning technique. The study repeated 100 rounds of model building after random data splitting. After exploring 504 variables (384 student and 99 parent), 32 variables were finally selected after selection counts. Variables predicting a sense of school belonging were categorized as individual/parent variables (e.g. motivation to achieve, tendency to cooperative learning, parental support) and school-related variables (e.g. school satisfaction, peer/teacher relationship, learning/physical activities). The significance and implications of the study as well as future research topics were discussed.

Two-step interpretable modeling of ICU-AIs

We present a novel methodology for integrating high resolution longitudinal data with the dynamic prediction capabilities of survival models. The aim is two-fold: to improve the predictive power while maintaining the interpretability of the models. To go beyond the black box paradigm of artificial neural networks, we propose a parsimonious and robust semi-parametric approach (i.e., a landmarking competing risks model) that combines routinely collected low-resolution data with predictive features extracted from a convolutional neural network, that was trained on high resolution time-dependent information. We then use saliency maps to analyze and explain the extra predictive power of this model. To illustrate our methodology, we focus on healthcare-associated infections in patients admitted to an intensive care unit.

Pretreating and normalizing metabolomics data for statistical analysis

Metabolomics as a research field and a set of techniques is to study the entire small molecules in biological samples. Metabolomics is emerging as a powerful tool generally for precision medicine. Particularly, integration of microbiome and metabolome has revealed the mechanism and functionality of microbiome in human health and disease. However, metabolomics data are very complicated. Preprocessing/pretreating and normalizing procedures on metabolomics data are usually required before statistical analysis. In this review article, we comprehensively review various methods that are used to preprocess and pretreat metabolomics data, including MS-based data and NMR -based data preprocessing, dealing with zero and/or missing values and detecting outliers, data normalization, data centering and scaling, data transformation. We discuss the advantages and limitations of each method. The choice for a suitable preprocessing method is determined by the biological hypothesis, the characteristics of the data set, and the selected statistical data analysis method. We then provide the perspective of their applications in the microbiome and metabolome research.

Multiomics of early epileptogenesis in mice reveals phosphorylation and dephosphorylation-directed growth and synaptic weakening

To investigate the phosphorylation-based signaling and protein changes occurring early in epileptogenesis, the hippocampi of mice treated with pilocarpine were examined by quantitative mass spectrometry at 4 and 24 h post-status epilepticus at vast depth. Hundreds of posttranscriptional regulatory proteins were the major early targets of increased phosphorylation. At 24 h, many protein level changes were detected and the phosphoproteome continued to be perturbed. The major targets of decreased phosphorylation at 4 and 24 h were a subset of postsynaptic density scaffold proteins, ion channels, and neurotransmitter receptors. Many proteins targeted by dephosphorylation at 4 h also had decreased protein abundance at 24 h, indicating a phosphatase-mediated weakening of synapses. Increased translation was indicated by protein changes at 24 h. These observations, and many additional indicators within this multiomic resource, suggest that early epileptogenesis is characterized by signaling that stimulates both growth and a homeostatic response that weakens excitability.

Dual-space disentangled-multimodal network (DDM-net) for glioma diagnosis and prognosis with incomplete pathology and genomic data

Objective. Effective fusion of histology slides and molecular profiles from genomic data has shown great potential in the diagnosis and prognosis of gliomas. However, it remains challenging to explicitly utilize the consistent-complementary information among different modalities and create comprehensive representations of patients. Additionally, existing researches mainly focus on complete multi-modality data and usually fail to construct robust models for incomplete samples.Approach. In this paper, we propose adual-space disentangled-multimodal network (DDM-net)for glioma diagnosis and prognosis. DDM-net disentangles the latent features generated by two separate variational autoencoders (VAEs) into common and specific components through a dual-space disentangled approach, facilitating the construction of comprehensive representations of patients. More importantly, DDM-net imputes the unavailable modality in the latent feature space, making it robust to incomplete samples.Main results. We evaluated our approach on the TCGA-GBMLGG dataset for glioma grading and survival analysis tasks. Experimental results demonstrate that the proposed method achieves superior performance compared to state-of-the-art methods, with a competitive AUC of 0.952 and a C-index of 0.768.Significance. The proposed model may help the clinical understanding of gliomas and can serve as an effective fusion model with multimodal data. Additionally, it is capable of handling incomplete samples, making it less constrained by clinical limitations.

Recalibrating differential gene expression by genetic dosage variance prioritizes functionally relevant genes

Differential expression (DE) analysis is a widely used method for identifying genes that are functionally relevant for an observed phenotype or biological response. However, typical DE analysis includes selection of genes based on a threshold of fold change in expression under the implicit assumption that all genes are equally sensitive to dosage changes of their transcripts. This tends to favor highly variable genes over more constrained genes where even small changes in expression may be biologically relevant. To address this limitation, we have developed a method to recalibrate each gene's differential expression fold change based on genetic expression variance observed in the human population. The newly established metric ranks statistically differentially expressed genes not by nominal change of expression, but by relative change in comparison to natural dosage variation for each gene. We apply our method to RNA sequencing datasets from rare disease and in-vitro stimulus response experiments. Compared to the standard approach, our method adjusts the bias in discovery towards highly variable genes, and enriches for pathways and biological processes related to metabolic and regulatory activity, indicating a prioritization of functionally relevant driver genes. With that, our method provides a novel view on DE and contributes towards bridging the existing gap between statistical and biological significance. We believe that this approach will simplify the identification of disease causing genes and enhance the discovery of therapeutic targets.

Missing Value Imputation of Wireless Sensor Data for Environmental Monitoring

Over the past few years, the scale of sensor networks has greatly expanded. This generates extended spatiotemporal datasets, which form a crucial information resource in numerous fields, ranging from sports and healthcare to environmental science and surveillance. Unfortunately, these datasets often contain missing values due to systematic or inadvertent sensor misoperation. This incompleteness hampers the subsequent data analysis, yet addressing these missing observations forms a challenging problem. This is especially the case when both the temporal correlation of timestamps within a single sensor and the spatial correlation between sensors are important. Here, we apply and evaluate 12 imputation methods to complete the missing values in a dataset originating from large-scale environmental monitoring. As part of a large citizen science project, IoT-based microclimate sensors were deployed for six months in 4400 gardens across the region of Flanders, generating 15-min recordings of temperature and soil moisture. Methods based on spatial recovery as well as time-based imputation were evaluated, including Spline Interpolation, MissForest, MICE, MCMC, M-RNN, BRITS, and others. The performance of these imputation methods was evaluated for different proportions of missing data (ranging from 10% to 50%), as well as a realistic missing value scenario. Techniques leveraging the spatial features of the data tend to outperform the time-based methods, with matrix completion techniques providing the best performance. Our results therefore provide a tool to maximize the benefit from costly, large-scale environmental monitoring efforts.

Proteomic Profiling of Endothelial Cells Exposed to Mitomycin C: Key Proteins and Pathways Underlying Genotoxic Stress-Induced Endothelial Dysfunction

Mitomycin C (MMC)-induced genotoxic stress can be considered to be a novel trigger of endothelial dysfunction and atherosclerosis-a leading cause of cardiovascular morbidity and mortality worldwide. Given the increasing genotoxic load on the human organism, the decryption of the molecular pathways underlying genotoxic stress-induced endothelial dysfunction could improve our understanding of the role of genotoxic stress in atherogenesis. Here, we performed a proteomic profiling of human coronary artery endothelial cells (HCAECs) and human internal thoracic endothelial cells (HITAECs) in vitro that were exposed to MMC to identify the biochemical pathways and proteins underlying genotoxic stress-induced endothelial dysfunction. We denoted 198 and 71 unique, differentially expressed proteins (DEPs) in the MMC-treated HCAECs and HITAECs, respectively; only 4 DEPs were identified in both the HCAECs and HITAECs. In the MMC-treated HCAECs, 44.5% of the DEPs were upregulated and 55.5% of the DEPs were downregulated, while in HITAECs, these percentages were 72% and 28%, respectively. The denoted DEPs are involved in the processes of nucleotides and RNA metabolism, vesicle-mediated transport, post-translation protein modification, cell cycle control, the transport of small molecules, transcription and signal transduction. The obtained results could improve our understanding of the fundamental basis of atherogenesis and help in the justification of genotoxic stress as a risk factor for atherosclerosis.

wMKL: multi-omics data integration enables novel cancer subtype identification via weight-boosted multi-kernel learning

BACKGROUND

Cancer is a heterogeneous disease driven by complex molecular alterations. Cancer subtypes determined from multi-omics data can provide novel insight into personalised precision treatment. It is recognised that incorporating prior weight knowledge into multi-omics data integration can improve disease subtyping.

METHODS

We develop a weighted method, termed weight-boosted Multi-Kernel Learning (wMKL) which incorporates heterogeneous data types as well as flexible weight functions, to boost subtype identification. Given a series of weight functions, we propose an omnibus combination strategy to integrate different weight-related P-values to improve subtyping precision.

RESULTS

wMKL models each data type with multiple kernel choices, thus alleviating the sensitivity and robustness issue due to selecting kernel parameters. Furthermore, wMKL integrates different data types by learning weights of different kernels derived from each data type, recognising the heterogeneous contribution of different data types to the final subtyping performance. The proposed wMKL outperforms existing weighted and non-weighted methods. The utility and advantage of wMKL are illustrated through extensive simulations and applications to two TCGA datasets. Novel subtypes are identified followed by extensive downstream bioinformatics analysis to understand the molecular mechanisms differentiating different subtypes.

CONCLUSIONS

The proposed wMKL method provides a novel strategy for disease subtyping. The wMKL is freely available at https://github.com/biostatcao/wMKL .

Late-onset and long-lasting neutropenias in the young: A new entity anticipating immune-dysregulation disorders

This study identifies a new chronic form of immune neutropenia in the young with or without detectable indirect anti-neutrophil antibodies, characterized by mild/moderate neutropenia low risk of severe infection (14%), tendency to develop autoimmune phenomena over the course of the disease (cumulative incidence of 58.6% after 20 years of disease duration), leukopenia, progressive reduction of absolute lymphocyte count and a T- and B-cell profile similar to autoimmune disorders like Sjogren syndrome, rheumatoid arthritis, and systemic lupus erythematosus (increased HLADR+ and CD3 + TCRγδ cells, reduced T regulatory cells, increased double-negative B and a tendency to reduced B memory cells). In a minority of patients, P/LP variants related to primary immuno-regulatory disorders were found. This new form may fit the group of "Likely acquired neutropenia," a provisional category included in the recent International Guidelines on Diagnosis and Management of Neutropenia of EHA and EUNET INNOCHRON ACTION 18233. The early recognition of this form of neutropenia would help clinicians to delineate better specific monitoring plans, genetic counseling, and potentially targeted therapies.

A prospective study of smoking-related white blood cell DNA methylation markers and risk of bladder cancer

Bladder cancer, a common neoplasm, is primarily caused by tobacco smoking. Epigenetic alterations including DNA methylation have the potential to be used as prospective markers of increased risk, particularly in at-risk populations such as smokers. We aimed to investigate the potential of smoking-related white blood cell (WBC) methylation markers to contribute to an increase in bladder cancer risk prediction over classical questionnaire-based smoking metrics (i.e., duration, intensity, packyears) in a nested case-control study within the prospective prostate, lung, colorectal, and ovarian (PLCO) Cancer Screening Trial and the alpha-tocopherol, beta-carotene cancer (ATBC) Prevention Study (789 cases; 849 controls). We identified 200 differentially methylated sites associated with smoking status and 28 significantly associated (after correction for multiple testing) with bladder cancer risk among 2670 previously reported smoking-related cytosine-phosphate-guanines sites (CpGs). Similar patterns were observed across cohorts. Receiver operating characteristic (ROC) analyses indicated that cg05575921 (AHHR), the strongest smoking-related association we identified for bladder cancer risk, alone yielded similar predictive performance (AUC: 0.60) than classical smoking metrics (AUC: 0.59-0.62). Best prediction was achieved by including the first principal component (PC1) from the 200 smoking-related CpGs alongside smoking metrics (AUC: 0.63-0.65). Further, PC1 remained significantly associated with elevated bladder cancer risk after adjusting for smoking metrics. These findings suggest DNA methylation profiles reflect aspects of tobacco smoke exposure in addition to those captured by smoking duration, intensity and packyears, and/or individual susceptibility relevant to bladder cancer etiology, warranting further investigation.

Exploiting nearest neighbor data and fuzzy membership function to address missing values in classification

The accuracy of most classification methods is significantly affected by missing values. Therefore, this study aimed to propose a data imputation method to handle missing values through the application of nearest neighbor data and fuzzy membership function as well as to compare the results with standard methods. A total of five datasets related to classification problems obtained from the UCI Machine Learning Repository were used. The results showed that the proposed method had higher accuracy than standard imputation methods. Moreover, triangular method performed better than Gaussian fuzzy membership function. This showed that the combination of nearest neighbor data and fuzzy membership function was more effective in handling missing values and improving classification accuracy.

Association of adrenal steroids with metabolomic profiles in patients with primary and endocrine hypertension

Introduction

Endocrine hypertension (EHT) due to pheochromocytoma/paraganglioma (PPGL), Cushing's syndrome (CS), or primary aldosteronism (PA) is linked to a variety of metabolic alterations and comorbidities. Accordingly, patients with EHT and primary hypertension (PHT) are characterized by distinct metabolic profiles. However, it remains unclear whether the metabolomic differences relate solely to the disease-defining hormonal parameters. Therefore, our objective was to study the association of disease defining hormonal excess and concomitant adrenal steroids with metabolomic alterations in patients with EHT.

Methods

Retrospective European multicenter study of 263 patients (mean age 49 years, 50% females; 58 PHT, 69 PPGL, 37 CS, 99 PA) in whom targeted metabolomic and adrenal steroid profiling was available. The association of 13 adrenal steroids with differences in 79 metabolites between PPGL, CS, PA and PHT was examined after correction for age, sex, BMI, and presence of diabetes mellitus.

Results

After adjustment for BMI and diabetes mellitus significant association between adrenal steroids and metabolites - 18 in PPGL, 15 in CS, and 23 in PA - were revealed. In PPGL, the majority of metabolite associations were linked to catecholamine excess, whereas in PA, only one metabolite was associated with aldosterone. In contrast, cortisone (16 metabolites), cortisol (6 metabolites), and DHEA (8 metabolites) had the highest number of associated metabolites in PA. In CS, 18-hydroxycortisol significantly influenced 5 metabolites, cortisol affected 4, and cortisone, 11-deoxycortisol, and DHEA each were linked to 3 metabolites.

Discussions

Our study indicates cortisol, cortisone, and catecholamine excess are significantly associated with metabolomic variances in EHT versus PHT patients. Notably, catecholamine excess is key to PPGL's metabolomic changes, whereas in PA, other non-defining adrenal steroids mainly account for metabolomic differences. In CS, cortisol, alongside other non-defining adrenal hormones, contributes to these differences, suggesting that metabolic disorders and cardiovascular morbidity in these conditions could also be affected by various adrenal steroids.

Interrogations of single-cell RNA splicing landscapes with SCASL define new cell identities with physiological relevance

RNA splicing shapes the gene regulatory programs that underlie various physiological and disease processes. Here, we present the SCASL (single-cell clustering based on alternative splicing landscapes) method for interrogating the heterogeneity of RNA splicing with single-cell RNA-seq data. SCASL resolves the issue of biased and sparse data coverage on single-cell RNA splicing and provides a new scheme for classifications of cell identities. With previously published datasets as examples, SCASL identifies new cell clusters indicating potentially precancerous and early-tumor stages in triple-negative breast cancer, illustrates cell lineages of embryonic liver development, and provides fine clusters of highly heterogeneous tumor-associated CD4 and CD8 T cells with functional and physiological relevance. Most of these findings are not readily available via conventional cell clustering based on single-cell gene expression data. Our study shows the potential of SCASL in revealing the intrinsic RNA splicing heterogeneity and generating biological insights into the dynamic and functional cell landscapes in complex tissues.

An interactive atlas of genomic, proteomic, and metabolomic biomarkers promotes the potential of proteins to predict complex diseases

Multiomics analyses have identified multiple potential biomarkers of the incidence and prevalence of complex diseases. However, it is not known which type of biomarker is optimal for clinical purposes. Here, we make a systematic comparison of 90 million genetic variants, 1,453 proteins, and 325 metabolites from 500,000 individuals with complex diseases from the UK Biobank. A machine learning pipeline consisting of data cleaning, data imputation, feature selection, and model training using cross-validation and comparison of the results on holdout test sets showed that proteins were most predictive, followed by metabolites, and genetic variants. Only five proteins per disease resulted in median (min-max) areas under the receiver operating characteristic curves for incidence of 0.79 (0.65-0.86) and 0.84 (0.70-0.91) for prevalence. In summary, our work suggests the potential of predicting complex diseases based on a limited number of proteins. We provide an interactive atlas (macd.shinyapps.io/ShinyApp/) to find genomic, proteomic, or metabolomic biomarkers for different complex diseases.

Prediction of dementia based on older adults' sleep disturbances using machine learning

BACKGROUND

The most common degenerative condition in older adults is dementia, which can be predicted using a number of indicators and whose progression can be slowed down. One of the indicators of an increased risk of dementia is sleep disturbances. This study aims to examine if machine learning can predict dementia and which sleep disturbance factors impact dementia.

METHODS

This study uses five machine learning algorithms (gradient boosting, logistic regression, gaussian naive Bayes, random forest and support vector machine) and data on the older population (60+) in Sweden from the Swedish National Study on Ageing and Care - Blekinge (n=4175). Each algorithm uses 10-fold stratified cross-validation to obtain the results, which consist of the Brier score for checking accuracy and the feature importance for examining the factors which impact dementia. The algorithms use 16 features which are on personal and sleep disturbance factors.

RESULTS

Logistic regression found an association between dementia and sleep disturbances. However, it is slight for the features in the study. Gradient boosting was the most accurate algorithm with 92.9% accuracy, 0.926 f1-score, 0.974 ROC AUC and 0.056 Brier score. The significant factors were different in each machine learning algorithm. If the person sleeps more than two hours during the day, their sex, education level, age, waking up during the night and if the person snores are the variables that most consistently have the highest feature importance in all algorithms.

CONCLUSION

There is an association between sleep disturbances and dementia, which machine learning algorithms can predict. Furthermore, the risk factors for dementia are different across the algorithms, but sleep disturbances can predict dementia.

Predicting effluent quality parameters for wastewater treatment plant: A machine learning-based methodology

Wastewater Treatment Plants (WWTPs) present complex biochemical processes of high variability and difficult prediction. This study presents an innovative approach using Machine Learning (ML) models to predict wastewater quality parameters. In particular, the models are applied to datasets from both a simulated wastewater treatment plant (WWTP), using DHI WEST software (WEST WWTP), and a real-world WWTP database from Santa Catarina Brewery AMBEV, located in Lages/SC - Brazil (AMBEV WWTP). A distinctive aspect is the evaluation of predictive performance in continuous data scenarios and the impact of changes in WWTP operations on predictive model performance, including changes in plant layout. For both plants, three different scenarios were addressed, and the quality of predictions by random forest (RF), support vector machine (SVM), and multilayer perceptron (MLP) models were evaluated. The prediction quality by the MLP model reached an R2 of 0.72 for TN prediction in the WEST WWTP output, and the RF model better adapted to the real data of the AMBEV WWTP, despite the significant discrepancy observed between the real and the predicted data. Techniques such as Partial Dependence Plots (PDP) and Permutation Importance (PI) were used to assess the importance of features, particularly in the simulated WEST tool scenario, showing a strong correlation of prediction results with influent parameters related to nitrogen content. The results of this study highlight the importance of collecting and storing high-quality data and the need for information on changes in WWTP operation for predictive model performance. These contributions advance the understanding of predictive modeling for wastewater quality and provide valuable insights for future practice in wastewater treatment.

Evidence of reduced gestational age in response to in utero arsenic exposure and implications for aging trajectories of the newborn

Arsenic exposure is associated with a plethora of age-related health outcomes of disparate etiology. However, evidence of the impact of arsenic on aging remains limited. Here, we investigated the utility of epigenetic clocks in two different populations and the impact of maternal arsenic exposure during pregnancy on epigenetic gestational age at birth. To do this, we examined publicly available DNA methylation data and estimated gestational age across five gestational clocks in two unrelated human populations. These populations also differ in the extent of arsenic exposure and the targeted tissue of analysis (cord blood and placental tissue). Our results indicate that same-tissue clocks produce gestational age estimates that are more highly correlated with clinical gestational age. Interestingly, our results also indicate that arsenic exposure is associated with gestational age, with higher arsenic exposures associated with decreased gestational age. We also applied two pediatric clocks to evaluate infant biological age in the same samples. The data is suggestive of higher pediatric age in infants exposed to higher arsenic levels during gestation. Taken altogether, our findings are consistent with past work indicating that that in utero arsenic exposure is associated with decreased gestational maturity as characterized by infant outcomes such as low birthweight and lung underdevelopment and dysfunction in arsenic exposed infants. The findings are also consistent with arsenic exposure setting infants on a trajectory of accelerated epigenetic aging that starts at birth.

Innovative approach for predicting biogas production from large-scale anaerobic digester using long-short term memory (LSTM) coupled with genetic algorithm (GA)

An artificial neural network (ANN) model called long-short term memory (LSTM), coupled with a genetic algorithm (GA) for feature selection, was used to predict biogas production of large-scale anaerobic digesters (ADs) of Tehran South Wastewater Treatment Plant (Iran), with a biogas production of approximately 30,000 Nm3/d. In order to employ the real conditions, the hydraulic retention time (HRT) of the ADs (21 days) was considered as the LSTM look-back window. To evaluate the model predictions, three different scenarios were defined. In the first scenario, the model predicted the produced biogas by using raw wastewater characteristics and reached the coefficient of determination of R2 = 0.84. The GA selected four out of eleven parameters of raw wastewater, including loads of BOD5, COD, TSS, and TN (kg/d), as the most informative data for the model. In the second scenario, the model predicted the produced biogas by employing the data of the thickened sludge streams entering the ADs and yielded a higher accuracy (R2 = 0.89). In this scenario, GA selected two out of six parameters of the sludge streams, including total flow rate (m3/d) and average solids content (w/w%). Finally, in the third scenario, by putting the parameters of the two previous scenarios together, the model's prediction accuracy increased slightly (R2 = 0.90). The results demonstrated that the GA-LSTM modeling technique could achieve reliable performance in predicting biogas production of large-scale ADs by including HRT in modeling procedure. It was also found that the raw wastewater characteristics severely affect AD behavior and can be successfully used as the input data of the AD models.

Predicting the effectiveness of drugs used for treating cardiovascular conditions in newborn infants

BACKGROUND

Cardiovascular support (CVS) treatment failure (TF) is associated with a poor prognosis in preterm infants.

METHODS

Medical charts of infants with a birth weight <1500 g who received either dopamine (Dp) or dobutamine (Db), were reviewed. Treatment response (TR) occurred if blood pressure increased >3rd centile for gestational age or superior vena cava flow was maintained >55 ml/kg/min, with decreased lactate or less negative base excess, without additional CVS. A predictive model of Dp and Db on TR was designed and the impact of TR on survival was analyzed.

RESULTS

Sixty-six infants (median gestational age 27.3 weeks, median birth weight 864 g) received Dp (n = 44) or Db (n = 22). TR occurred in 59% of the cases treated with Dp and 31% with Db, p = 0.04. Machine learning identified a model that correctly labeled Db response in 90% of the cases and Dp response in 61.4%. Sixteen infants died (9% of the TR group, 39% of the TF group; p = 0.004). Brain or gut morbidity-free survival was observed in 52% vs 30% in the TR and TF groups, respectively (p = 0.08).

CONCLUSIONS

New predictive models can anticipate Db but not Dp effectiveness in preterm infants. These algorithms may help the clinicians in the decision-making process.

IMPACT

Failure of cardiovascular support treatment increases the risk of mortality in very low birth weight infants. A predictive model built with machine learning techniques can help anticipate treatment response to dobutamine with high accuracy. Predictive models based on artificial intelligence may guide the clinicians in the decision-making process.

One Month Prediction of Pressure Ulcers in Nursing Home Residents with Bayesian Networks

OBJECTIVES

Pressure ulcers (PUs) are a common and avoidable condition among residents of nursing homes, and their consequences are severe. Reliable and simple identification of high-risk residents is a major challenge for prevention. Available tools like the Braden and Norton scale have imperfect predictive performance. The objective is to predict the occurrence of PUs in nursing home residents from electronic health record (EHR) data.

DESIGN

Longitudinal retrospective nested case-control study.

SETTING AND PARTICIPANTS

EHR database of French nursing homes from 2013 to 2022.

METHODS

Residents who suffered from PUs were cases and those who did not were controls. For cases, we analyzed the data available in their EHR 1 month before the occurrence of the first PU. For controls, we used available data 1 month before an index date adjusted on the delays of PU onset. We conducted a Bayesian network (BN) analysis, an explainable machine learning method, using 136 input variables of potential medical interest determined with experts. To validate the model, we used scores, features selection, and explainability tools such as Shapley values.

RESULTS

Among 58,368 residents analyzed, 29% suffered from PUs during their stay. The obtained BN model predicts the occurrence of a PU at a 1-month horizon with a sensitivity of 0.94 (±0.01), a precision of 0.32 (±0.01) and an area under the curve of 0.69 (±0.02). It selects 3 variables: length of stay, delay since last hospitalization, and dependence for transfer. This BN model is suitable and simpler than models provided by other machine learning methods.

CONCLUSIONS AND IMPLICATIONS

One-month prediction for incident PU is possible in nursing home residents from their EHR data. The study paves the way for the development of a predictive tool fueled by routinely collected data that do not require additional work from health care professionals, thereby opening a new preventive strategy for PUs.

Disease-Associated Neurotoxic Astrocyte Markers in Alzheimer Disease Based on Integrative Single-Nucleus RNA Sequencing

Alzheimer disease (AD) is an irreversible neurodegenerative disease, and astrocytes play a key role in its onset and progression. The aim of this study is to analyze the characteristics of neurotoxic astrocytes and identify novel molecular targets for slowing down the progression of AD. Single-nucleus RNA sequencing (snRNA-seq) data were analyzed from various AD cohorts comprising about 210,654 cells from 53 brain tissue. By integrating snRNA-seq data with bulk RNA-seq data, crucial astrocyte types and genes associated with the prognosis of patients with AD were identified. The expression of neurotoxic astrocyte markers was validated using 5 × FAD and wild-type (WT) mouse models, combined with experiments such as western blot, quantitative real-time PCR (qRT-PCR), and immunofluorescence. A group of neurotoxic astrocytes closely related to AD pathology was identified, which were involved in inflammatory responses and pathways related to neuron survival. Combining snRNA and bulk tissue data, ZEP36L, AEBP1, WWTR1, PHYHD1, DST and RASL12 were identified as toxic astrocyte markers closely related to disease severity, significantly elevated in brain tissues of 5 × FAD mice and primary astrocytes treated with Aβ. Among them, WWTR1 was significantly increased in astrocytes of 5 × FAD mice, driving astrocyte inflammatory responses, and has been identified as an important marker of neurotoxic astrocytes. snRNA-seq analysis reveals the biological functions of neurotoxic astrocytes. Six genes related to AD pathology were identified and validated, among which WWTR1 may be a novel marker of neurotoxic astrocytes.

Physiological response to fetal intravenous lipid emulsion

In preterm neonates unable to obtain sufficient oral nutrition, intravenous lipid emulsion is life-saving. The contribution of post-conceptional level of maturation to pathology that some neonates experience is difficult to untangle from the global pathophysiology of premature birth. In the present study, we determined fetal physiological responses to intravenous lipid emulsion. Fetal sheep were given intravenous Intralipid 20® (n = 4 females, 7 males) or Lactated Ringer's Solution (n = 7 females, 4 males) between 125 ± 1 and 133 ± 1 d of gestation (term = 147 d). Manufacturer's recommendation for premature human infants was followed: 0.5-1 g/kg/d initial rate, increased by 0.5-1 to 3 g/kg/d. Hemodynamic parameters and arterial blood chemistry were measured, and organs were studied postmortem. Red blood cell lipidomics were analyzed by LC-MS. Intravenous Intralipid did not alter hemodynamic or most blood parameters. Compared with controls, Intralipid infusion increased final day plasma protein (P=0.004; 3.5 ± 0.3 vs. 3.9 ± 0.2 g/dL), albumin (P = 0.031; 2.2 ± 0.1 vs. 2.4 ± 0.2 g/dL), and bilirubin (P<0.001; conjugated: 0.2 ± 0.1 vs. 0.6 ± 0.2 mg/dL; unconjugated: 0.2 ± 0.1 vs. 1.1 ± 0.4 mg/dL). Circulating IGF-1 decreased following Intralipid infusion (P<0.001; 66 ± 24 vs. 46 ± 24 ng/mL). Compared with control Oil Red O liver stains (median score 0), Intralipid-infused fetuses scored 108 (P=0.0009). Lipidomic analysis revealed uptake and processing of infused lipids into red blood cells, increasing abundance of saturated fatty acids. The near-term fetal sheep tolerates intravenous lipid emulsion well, although lipid accumulates in the liver. Increased levels of unconjugated bilirubin may reflect increased red blood cell turnover or impaired placental clearance. Whether Intralipid is less well tolerated earlier in gestation remains to be determined.

Application of SWATH Mass Spectrometry and Machine Learning in the Diagnosis of Inflammatory Bowel Disease Based on the Stool Proteome

Inflammatory bowel disease (IBD) flare-ups exhibit symptoms that are similar to other diseases and conditions, making diagnosis and treatment complicated. Currently, the gold standard for diagnosing and monitoring IBD is colonoscopy and biopsy, which are invasive and uncomfortable procedures, and the fecal calprotectin test, which is not sufficiently accurate. Therefore, it is necessary to develop an alternative method. In this study, our aim was to provide proof of concept for the application of Sequential Window Acquisition of All Theoretical Mass Spectra-Mass spectrometry (SWATH-MS) and machine learning to develop a non-invasive and accurate predictive model using the stool proteome to distinguish between active IBD patients and symptomatic non-IBD patients. Proteome profiles of 123 samples were obtained and data processing procedures were optimized to select an appropriate pipeline. The differentially abundant analysis identified 48 proteins. Utilizing correlation-based feature selection (Cfs), 7 proteins were selected for proceeding steps. To identify the most appropriate predictive machine learning model, five of the most popular methods, including support vector machines (SVMs), random forests, logistic regression, naive Bayes, and k-nearest neighbors (KNN), were assessed. The generated model was validated by implementing the algorithm on 45 prospective unseen datasets; the results showed a sensitivity of 96% and a specificity of 76%, indicating its performance. In conclusion, this study illustrates the effectiveness of utilizing the stool proteome obtained through SWATH-MS in accurately diagnosing active IBD via a machine learning model.

Leveraging synthetic lethality to uncover potential therapeutic target in gastric cancer

Since trastuzumab was approved in 2012 for the first-line treatment of gastric cancer (GC), no significant advancement in GC targeted therapies has occurred. Synthetic lethality refers to the concept that simultaneous dysfunction of a pair of genes results in a lethal effect on cells, while the loss of an individual gene does not cause this effect. Through exploiting synthetic lethality, novel targeted therapies can be developed for the individualized treatment of GC. In this study, we proposed a computational strategy named Gastric cancer Specific Synthetic Lethality inference (GSSL) to identify synthetic lethal interactions in GC. GSSL analysis was used to infer probable synthetic lethality in GC using four accessible clinical datasets. In addition, prediction results were confirmed by experiments. GSSL analysis identified a total of 34 candidate synthetic lethal pairs, which included 33 unique targets. Among the synthetic lethal gene pairs, TP53-CHEK1 was selected for further experimental validation. Both computational and experimental results indicated that inhibiting CHEK1 could be a potential therapeutic strategy for GC patients with TP53 mutation. Meanwhile, in vitro experimental validation of two novel synthetic lethal pairs TP53-AURKB and ARID1A-EP300 further proved the universality and reliability of GSSL. Collectively, GSSL has been shown to be a reliable and feasible method for comprehensive analysis of inferring synthetic lethal interactions of GC, which may offer novel insight into the precision medicine and individualized treatment of GC.

Multiple augmented reduced rank regression for pan-cancer analysis

Statistical approaches that successfully combine multiple datasets are more powerful, efficient, and scientifically informative than separate analyses. To address variation architectures correctly and comprehensively for high-dimensional data across multiple sample sets (ie, cohorts), we propose multiple augmented reduced rank regression (maRRR), a flexible matrix regression and factorization method to concurrently learn both covariate-driven and auxiliary structured variations. We consider a structured nuclear norm objective that is motivated by random matrix theory, in which the regression or factorization terms may be shared or specific to any number of cohorts. Our framework subsumes several existing methods, such as reduced rank regression and unsupervised multimatrix factorization approaches, and includes a promising novel approach to regression and factorization of a single dataset (aRRR) as a special case. Simulations demonstrate substantial gains in power from combining multiple datasets, and from parsimoniously accounting for all structured variations. We apply maRRR to gene expression data from multiple cancer types (ie, pan-cancer) from The Cancer Genome Atlas, with somatic mutations as covariates. The method performs well with respect to prediction and imputation of held-out data, and provides new insights into mutation-driven and auxiliary variations that are shared or specific to certain cancer types.

Effects of Preanalytical Sample Collection and Handling on Comprehensive Metabolite Measurements in Human Urine Biospecimens

Epidemiology studies evaluate associations between the metabolome and disease risk. Urine is a common biospecimen used for such studies due to its wide availability and non-invasive collection. Evaluating the robustness of urinary metabolomic profiles under varying preanalytical conditions is thus of interest. Here we evaluate the impact of sample handling conditions on urine metabolome profiles relative to the gold standard condition (no preservative, no refrigeration storage, single freeze thaw). Conditions tested included the use of borate or chlorhexidine preservatives, various storage and freeze/thaw cycles. We demonstrate that sample handling conditions impact metabolite levels, with borate showing the largest impact with 125 of 1,048 altered metabolites (adjusted P < 0.05). When simulating a case-control study with expected inconsistencies in sample handling, we predicted the occurrence of false positive altered metabolites to be low (< 11). Predicted false positives increased substantially (³63) when cases were simulated to undergo alternate handling. Finally, we demonstrate that sample handling impacts on the urinary metabolome were markedly smaller than those in serum. While changes in urine metabolites incurred by sample handling are generally small, we recommend implementing consistent handling conditions and evaluating robustness of metabolite measurements for those showing significant associations with disease outcomes.

Predicting lung cancer survival prognosis based on the conditional survival bayesian network

Lung cancer is a leading cause of cancer deaths and imposes an enormous economic burden on patients. It is important to develop an accurate risk assessment model to determine the appropriate treatment for patients after an initial lung cancer diagnosis. The Cox proportional hazards model is mainly employed in survival analysis. However, real-world medical data are usually incomplete, posing a great challenge to the application of this model. Commonly used imputation methods cannot achieve sufficient accuracy when data are missing, so we investigated novel methods for the development of clinical prediction models. In this article, we present a novel model for survival prediction in missing scenarios. We collected data from 5,240 patients diagnosed with lung cancer at the Weihai Municipal Hospital, China. Then, we applied a joint model that combined a BN and a Cox model to predict mortality risk in individual patients with lung cancer. The established prognostic model achieved good predictive performance in discrimination and calibration. We showed that combining the BN with the Cox proportional hazards model is highly beneficial and provides a more efficient tool for risk prediction.

A multivariate cognitive approach to predict social functioning in recent onset psychosis in response to computerized cognitive training

Clinical and neuroimaging data has been increasingly used in recent years to disentangle heterogeneity of treatment response to cognitive training (CT) and predict which individuals may achieve the highest benefits. CT has small to medium effects on improving cognitive and social functioning in recent onset psychosis (ROP) patients, who show the most profound cognitive and social functioning deficits among psychiatric patients. We employed multivariate pattern analysis (MVPA) to investigate the potential of cognitive data to predict social functioning improvement in response to 10 h of CT in patients with ROP. A support vector machine (SVM) classifier was trained on the naturalistic data of the Personalized Prognostic Tools for Early Psychosis Management (PRONIA) study sample to predict functioning in an independent sample of 70 ROP patients using baseline cognitive data. PRONIA is a part of a FP7 EU grant program that involved 7 sites across 5 European countries, designed and conducted with the main aim of identifying (bio)markers associated with an enhanced risk of developing psychosis in order to improve early detection and prognosis. Social functioning was predicted with a balanced accuracy (BAC) of 66.4% (Sensitivity 78.8%; Specificity 54.1%; PPV 60.5%; NPV 74.1%; AUC 0.64; P = 0.01). The most frequently selected cognitive features (mean feature weights > ± 0.2) included the (1) correct number of symbol matchings within the Digit Symbol Substitution Test, (2) the number of distracting stimuli leading to an error within 300 and 200 trials in the Continuous Performance Test and (3) the dynamics of verbal fluency between 15 and 30 s within the Verbal Fluency Test, phonetic part. Next, the SVM classifier generated on the PRONIA sample was applied to the intervention sample, that obtained 54 ROP patients who were randomly assigned to a social cognitive training (SCT) or treatment as usual (TAU) group and dichotomized into good (GF-S ≥ 7) and poor (GF-S < 7) functioning patients based on their level of Global Functioning-Social (GF-S) score at follow-up (FU). By applying the initial PRONIA classifier, using out-of-sample cross-validation (OOCV) to the sample of ROP patients who have undergone the CT intervention, a BAC of 59.3% (Sensitivity 70.4%; Specificity 48.1%; PPV 57.6%; NPV 61.9%; AUC 0.63) was achieved at T0 and a BAC of 64.8% (Sensitivity 66.7%; Specificity 63.0%; PPV 64.3%; NPV 65.4%; AUC 0.66) at FU. After SCT intervention, a significant improvement in predicted social functioning values was observed in the SCT compared to TAU group (P ≤0.05; ES[Cohens' d] = 0.18). Due to a small sample size and modest variance of social functioning of the intervention sample it was not feasible to predict individual response to SCT in the current study. Our findings suggest that the use of baseline cognitive data could provide a robust individual estimate of future social functioning, while prediction of individual response to SCT using cognitive data that can be generated in the routine patient care remains to be addressed in large-scale cognitive training trials.

Climatic zone effects of non-native plant invasion on CH4 and N2O emissions from natural wetland ecosystems

Plant invasion can significantly alter the carbon and nitrogen cycles of wetlands, which potentially affects the emission of greenhouse gases (GHGs). The extent of these effects can vary depending on several factors, including the species of invasive plants, their growth patterns, and the climatic conditions prevailing in the wetland. Understanding the global effects of plant invasion on the emission of methane (CH4) and nitrous oxide (N2O) is crucial for the climate-smart management of wetlands. Here, we performed a global meta-analysis of 207 paired case studies that quantified the effect of non-native plant invasion on CH4 and N2O emissions in tropical/sub-tropical (TS) and temperate (TE) wetlands. The average emission rate of CH4 from the TS wetlands increased significantly from 337 to 577 kg CH4 ha-1 yr-1 in areas where native plants had been displaced by invasive plants. Similarly, in TE wetlands, the emission rates increased from 211 to 299 kg CH4 ha-1 yr-1 following the invasion of alien plant species. The increase in CH4 emissions at invaded sites was attributed to the increase in plant biomass, soil organic carbon (SOC), and soil moisture (SM). The effects of plant invasion on N2O emissions differed between TS and TE wetlands in that there was no significant effect in TS wetlands, whereas the N2O emissions reduced in TE wetlands. This difference in N2O emissions between climate zones was attributed to the depletion of NH4+ and NO3- in soils and the lower soil temperature in temperate regions. Overall, plant invasion increased the global net CH4 emissions from natural wetlands by 10.54 Tg CH4 yr-1. However, there were variations in CH4 emissions across different climatic zones, indicated by a net increase in CH4 emissions, of 9.97 and 0.57 Tg CH4 yr-1 in TS and TE wetlands, respectively. These findings highlight that plant invasion not only strongly stimulates the emission of CH4 from TS wetlands, but also suppresses N2O emissions from TE wetlands. These novel insights immensely improve our current understanding of the effects of climatic zones on biogeochemical controlling factors that influence the production of greenhouse gases (GHGs) from wetlands following plant invasion. By analyzing the specific mechanisms by which invasive plants affect GHG emissions in different climatic zones, effective strategies can be devised to reduce GHG emissions and preserve wetland ecosystems.

Deep learning based decision tree ensembles for incomplete medical datasets

BACKGROUND

In practice, the collected datasets for data analysis are usually incomplete as some data contain missing attribute values. Many related works focus on constructing specific models to produce estimations to replace the missing values, to make the original incomplete datasets become complete. Another type of solution is to directly handle the incomplete datasets without missing value imputation, with decision trees being the major technique for this purpose.

OBJECTIVE

To introduce a novel approach, namely Deep Learning-based Decision Tree Ensembles (DLDTE), which borrows the bounding box and sliding window strategies used in deep learning techniques to divide an incomplete dataset into a number of subsets and learning from each subset by a decision tree, resulting in decision tree ensembles.

METHOD

Two medical domain problem datasets contain several hundred feature dimensions with the missing rates of 10% to 50% are used for performance comparison.

RESULTS

The proposed DLDTE provides the highest rate of classification accuracy when compared with the baseline decision tree method, as well as two missing value imputation methods (mean and k-nearest neighbor), and the case deletion method.

CONCLUSION

The results demonstrate the effectiveness of DLDTE for handling incomplete medical datasets with different missing rates.

Label-Free Quantitation of Endogenous Peptides

Liquid chromatography-mass spectrometry (LC-MS)-based peptidomics methods allow for the detection and identification of many peptides in a complex biological mixture in an untargeted manner. Quantitative peptidomics approaches allow for comparisons of peptide abundance between different samples, allowing one to draw conclusions about peptide differences as a function of experimental treatment or physiology. While stable isotope labeling is a powerful approach for quantitative proteomics and peptidomics, advances in mass spectrometry instrumentation and analysis tools have allowed label-free methods to gain popularity in recent years. In a general label-free quantitative peptidomics experiment, peak intensity information for each peptide is compared across multiple LC-MS runs. Here, we outline a general approach for label-free quantitative peptidomics experiments, including steps for sample preparation, LC-MS data acquisition, data processing, and statistical analysis. Special attention is paid to address run-to-run variability, which can lead to several major problems in label-free experiments. Overall, our method provides researchers with a framework for the development of their own quantitative peptidomics workflows applicable to quantitation of peptides from a wide variety of different biological sources.

Towards Personalised Mood Prediction and Explanation for Depression from Biophysical Data

Digital health applications using Artificial Intelligence (AI) are a promising opportunity to address the widening gap between available resources and mental health needs globally. Increasingly, passively acquired data from wearables are augmented with carefully selected active data from depressed individuals to develop Machine Learning (ML) models of depression based on mood scores. However, most ML models are black box in nature, and hence the outputs are not explainable. Depression is also multimodal, and the reasons for depression may vary significantly between individuals. Explainable and personalised models will thus be beneficial to clinicians to determine the main features that lead to a decline in the mood state of a depressed individual, thus enabling suitable personalised therapy. This is currently lacking. Therefore, this study presents a methodology for developing personalised and accurate Deep Learning (DL)-based predictive mood models for depression, along with novel methods for identifying the key facets that lead to the exacerbation of depressive symptoms. We illustrate our approach by using an existing multimodal dataset containing longitudinal Ecological Momentary Assessments of depression, lifestyle data from wearables and neurocognitive assessments for 14 mild to moderately depressed participants over one month. We develop classification- and regression-based DL models to predict participants' current mood scores-a discrete score given to a participant based on the severity of their depressive symptoms. The models are trained inside eight different evolutionary-algorithm-based optimisation schemes that optimise the model parameters for a maximum predictive performance. A five-fold cross-validation scheme is used to verify the DL model's predictive performance against 10 classical ML-based models, with a model error as low as 6% for some participants. We use the best model from the optimisation process to extract indicators, using SHAP, ALE and Anchors from explainable AI literature to explain why certain predictions are made and how they affect mood. These feature insights can assist health professionals in incorporating personalised interventions into a depressed individual's treatment regimen.

Evaluating Machine Learning Methods of Analyzing Multiclass Metabolomics

Multiclass metabolomic studies have become popular for revealing the differences in multiple stages of complex diseases, various lifestyles, or the effects of specific treatments. In multiclass metabolomics, there are multiple data manipulation steps for analyzing raw data, which consist of data filtering, the imputation of missing values, data normalization, marker identification, sample separation, classification, and so on. In each step, several to dozens of machine learning methods can be chosen for the given data set, with potentially hundreds or thousands of method combinations in the whole data processing chain. Therefore, a clear understanding of these machine learning methods is helpful for selecting an appropriate method combination for obtaining stable and reliable analytical results of specific data. However, there has rarely been an overall introduction or evaluation of these methods based on multiclass metabolomic data. Herein, detailed descriptions of these machine learning methods in multiple data manipulation steps are reviewed. Moreover, an assessment of these methods was performed using a benchmark data set for multiclass metabolomics. First, 12 imputation methods for imputing missing values were evaluated based on the PSS (Procrustes statistical shape analysis) and NRMSE (normalized root-mean-square error) values. Second, 17 normalization methods for processing multiclass metabolomic data were evaluated by applying the PMAD (pooled median absolute deviation) value. Third, different methods of identifying markers of multiclass metabolomics were evaluated based on the CWrel (relative weighted consistency) value. Fourth, nine classification methods for constructing multiclass models were assessed using the AUC (area under the curve) value. Performance evaluations of machine learning methods are highly recommended to select the most appropriate method combination before performing the final analysis of the given data. Overall, detailed descriptions and evaluation of various machine learning methods are expected to improve analyses of multiclass metabolomic data.

Pneumonia in the first week after polytrauma is associated with reduced blood levels of soluble herpes virus entry mediator

Background

Pneumonia develops frequently after major surgery and polytrauma and thus in the presence of systemic inflammatory response syndrome (SIRS) and organ dysfunction. Immune checkpoints balance self-tolerance and immune activation. Altered checkpoint blood levels were reported for sepsis. We analyzed associations of pneumonia incidence in the presence of SIRS during the first week of critical illness and trends in checkpoint blood levels.

Materials and methods

Patients were studied from day two to six after admission to a surgical intensive care unit (ICU). Blood was sampled and physician experts retrospectively adjudicated upon the presence of SIRS and Sepsis-1/2 every eight hours. We measured the daily levels of immune checkpoints and inflammatory markers by bead arrays for polytrauma patients developing pneumonia. Immune checkpoint time series were additionally determined for clinically highly similar polytrauma controls remaining infection-free during follow-up. We performed cluster analyses. Immune checkpoint time trends in cases and controls were compared with hierarchical linear models. For patients with surgical trauma and with and without sepsis, selected immune checkpoints were determined in study baseline samples.

Results

In polytrauma patients with post-injury pneumonia, eleven immune checkpoints dominated subcluster 3 that separated subclusters 1 and 2 of myeloid markers from subcluster 4 of endothelial activation, tissue inflammation, and adaptive immunity markers. Immune checkpoint blood levels were more stable in polytrauma cases than controls, where they trended towards an increase in subcluster A and a decrease in subcluster B. Herpes virus entry mediator (HVEM) levels (subcluster A) were lower in cases throughout. In unselected surgical patients, sepsis was not associated with altered HVEM levels at the study baseline.

Conclusion

Pneumonia development after polytrauma until ICU-day six was associated with decreased blood levels of HVEM. HVEM signaling may reduce pneumonia risk by strengthening myeloid antimicrobial defense and dampening lymphoid-mediated tissue damage. Future investigations into the role of HVEM in pneumonia and sepsis development and as a predictive biomarker should consider the etiology of critical illness and the site of infection.

Machine learning-based meta-analysis of colorectal cancer and inflammatory bowel disease

Colorectal cancer (CRC) is a major global health concern, resulting in numerous cancer-related deaths. CRC detection, treatment, and prevention can be improved by identifying genes and biomarkers. Despite extensive research, the underlying mechanisms of CRC remain elusive, and previously identified biomarkers have not yielded satisfactory insights. This shortfall may be attributed to the predominance of univariate analysis methods, which overlook potential combinations of variants and genes contributing to disease development. Here, we address this knowledge gap by presenting a novel multivariate machine-learning strategy to pinpoint genes associated with CRC. Additionally, we applied our analysis pipeline to Inflammatory Bowel Disease (IBD), as IBD patients face substantial CRC risk. The importance of the identified genes was substantiated by rigorous validation across numerous independent datasets. Several of the discovered genes have been previously linked to CRC, while others represent novel findings warranting further investigation. A Python implementation of our pipeline can be accessed publicly at https://github.com/AriaSar/CRCIBD-ML.

Deviation-support based fuzzy ensemble of multi-modal deep learning classifiers for breast cancer prognosis prediction

Breast cancer is the fifth leading cause of death in females worldwide. Early detection and treatment are crucial for improving health outcomes and preventing more serious conditions. Analyzing diverse information from multiple sources without errors, particularly with the growing burden of cancer cases, is a daunting task for humans. In this study, our main objective is to improve the accuracy of breast cancer survival prediction using a novel ensemble approach. It is novel due to the consideration of deviation (closeness between predicted classes and actual classes) and support (sparsity between predicted classes and actual classes) of the predicted class with respect to the actual class, a feature lacking in traditional ensembles. The ensemble uses fuzzy integrals on support and deviation scores from base classifiers to calculate aggregated scores while considering how confident or uncertain each classifier is. The proposed ensemble mechanism has been evaluated on a multi-modal breast cancer dataset of breast tumors collected from participants in the METABRIC trial. The proposed architecture proves its efficiency by achieving the accuracy, sensitivity, F1-score, and balanced accuracy of 82.88%, 58.64%, 62.94%, and 74.75% respectively. The obtained results are superior to the performance of individual classifiers and existing ensemble approaches.

MultiPro: DDA-PASEF and diaPASEF acquired cell line proteomic datasets with deliberate batch effects

Mass spectrometry-based proteomics plays a critical role in current biological and clinical research. Technical issues like data integration, missing value imputation, batch effect correction and the exploration of inter-connections amongst these technical issues, can produce errors but are not well studied. Although proteomic technologies have improved significantly in recent years, this alone cannot resolve these issues. What is needed are better algorithms and data processing knowledge. But to obtain these, we need appropriate proteomics datasets for exploration, investigation, and benchmarking. To meet this need, we developed MultiPro (Multi-purpose Proteome Resource), a resource comprising four comprehensive large-scale proteomics datasets with deliberate batch effects using the latest parallel accumulation-serial fragmentation in both Data-Dependent Acquisition (DDA) and Data Independent Acquisition (DIA) modes. Each dataset contains a balanced two-class design based on well-characterized and widely studied cell lines (A549 vs K562 or HCC1806 vs HS578T) with 48 or 36 biological and technical replicates altogether, allowing for investigation of a multitude of technical issues. These datasets allow for investigation of inter-connections between class and batch factors, or to develop approaches to compare and integrate data from DDA and DIA platforms.

ABTCN: an efficient hybrid deep learning approach for atmospheric temperature prediction

Temperature prediction is an important and significant step for monitoring global warming and the environment to save and protect human lives. The climatology parameters such as temperature, pressure, and wind speed are time-series data and are well predicted with data driven models. However, data-driven models have certain constraints, due to which these models are unable to predict the missing values and erroneous data caused by factors like sensor failure and natural disasters. In order to solve this issue, an efficient hybrid model, i.e., attention-based bidirectional long short term memory temporal convolution network (ABTCN) architecture is proposed. ABTCN uses k-nearest neighbor (KNN) imputation method for handling the missing data. A bidirectional long short term memory (Bi-LSTM) network with self-attention mechanism and temporal convolutional network (TCN) model that aids in the extraction of features from complex data and prediction of long data sequence. The performance of the proposed model is evaluated in comparison to various state-of-the-art deep learning models using error metrics such as MAE, MSE, RMSE, and R2 score. It is observed that our proposed model is superior over other models with high accuracy.

Impact of increasing post-filter ionized calcium target on filter lifespan in renal replacement therapy with regional citrate anticoagulation: A before-and-after study

INTRODUCTION

Regional citrate anticoagulation (RCA) is the recommended method for anticoagulation in continuous renal replacement therapy (CRRT). However, the optimal post-filter ionized calcium (iCa) target level remains unclear. This study aims to assess the effect of increasing the post-filter iCa target level from 0.25-0.35 mmol/L to 0.30-0.40 mmol/L on filter lifespan until clotting during RCA-CRRT.

METHODS

This before-and-after single-center study included patients who underwent RCA-CRRT sessions without systemic anticoagulation during two periods. The first period included patients with a post-filter iCa target between 0.25 and 0.35 mmol/L, while the second period included those with a target between 0.30 and 0.40 mmol/L. The primary outcome was filter lifespan until clotting.

RESULTS

A total of 1037 CRRT sessions were analyzed, with 610 sessions in the first period and 427 sessions in the second period. After adjusting for confounding factors, there was no significant difference in filter lifespan until clotting between the two groups (hazard ratio, 1.020 [0.703; 1.481]; p = 0.92).

CONCLUSION

Increasing the post-filter iCa target level from 0.25-0.35 mmol/L to 0.30-0.40 mmol/L during RCA-CRRT does not reduce filter lifespan until clotting and may decrease unnecessary citrate exposure. However, the optimal post-filter iCa target should be individualized according to the patient's clinical and biological status.

Prediction models of incontinence and sexual function one year after radical prostatectomy based on data from 20 164 prostate cancer patients

BACKGROUND

Incontinence and sexual dysfunction are long-lasting side effects after surgical treatment (radical prostatectomy, RP) of prostate cancer (PC). For an informed treatment decision, physicians and patients should discuss expected impairments. Therefore, this paper firstly aims to develop and validate prognostic models that predict incontinence and sexual function of PC patients one year after RP and secondly to provide an online decision making tool.

METHODS

Observational cohorts of PC patients treated between July 2016 and March 2021 in Germany were used. Models to predict functional outcomes one year after RP measured by the EPIC-26 questionnaire were developed using lasso regression, 80-20 splitting of the data set and 10-fold cross validation. To assess performance, R2, RMSE, analysis of residuals and calibration-in-the-large were applied. Final models were externally temporally validated. Additionally, percentages of functional impairment (pad use for incontinence and firmness of erection for sexual score) per score decile were calculated to be used together with the prediction models.

RESULTS

For model development and internal as well as external validation, samples of 11 355 and 8 809 patients were analysed. Results from the internal validation (incontinence: R2 = 0.12, RMSE = 25.40, sexual function: R2 = 0.23, RMSE = 21.44) were comparable with those of the external validation. Residual analysis and calibration-in-the-large showed good results. The prediction tool is freely accessible: https://nora-tabea.shinyapps.io/EPIC-26-Prediction/.

CONCLUSION

The final models showed appropriate predictive properties and can be used together with the calculated risks for specific functional impairments. Main strengths are the large study sample (> 20 000) and the inclusion of an external validation. The models incorporate meaningful and clinically available predictors ensuring an easy implementation. All predictions are displayed together with risks of frequent impairments such as pad use or erectile dysfunction such that the developed online tool provides a detailed and informative overview for clinicians as well as patients.

Deep learning-based phenotype imputation on population-scale biobank data increases genetic discoveries

Biobanks that collect deep phenotypic and genomic data across many individuals have emerged as a key resource in human genetics. However, phenotypes in biobanks are often missing across many individuals, limiting their utility. We propose AutoComplete, a deep learning-based imputation method to impute or 'fill-in' missing phenotypes in population-scale biobank datasets. When applied to collections of phenotypes measured across ~300,000 individuals from the UK Biobank, AutoComplete substantially improved imputation accuracy over existing methods. On three traits with notable amounts of missingness, we show that AutoComplete yields imputed phenotypes that are genetically similar to the originally observed phenotypes while increasing the effective sample size by about twofold on average. Further, genome-wide association analyses on the resulting imputed phenotypes led to a substantial increase in the number of associated loci. Our results demonstrate the utility of deep learning-based phenotype imputation to increase power for genetic discoveries in existing biobank datasets.

Plasma proteomic profiles predict individual future health risk

Developing a single-domain assay to identify individuals at high risk of future events is a priority for multi-disease and mortality prevention. By training a neural network, we developed a disease/mortality-specific proteomic risk score (ProRS) based on 1461 Olink plasma proteins measured in 52,006 UK Biobank participants. This integrative score markedly stratified the risk for 45 common conditions, including infectious, hematological, endocrine, psychiatric, neurological, sensory, circulatory, respiratory, digestive, cutaneous, musculoskeletal, and genitourinary diseases, cancers, and mortality. The discriminations witnessed high accuracies achieved by ProRS for 10 endpoints (e.g., cancer, dementia, and death), with C-indexes exceeding 0.80. Notably, ProRS produced much better or equivalent predictive performance than established clinical indicators for almost all endpoints. Incorporating clinical predictors with ProRS enhanced predictive power for most endpoints, but this combination only exhibited limited improvement when compared to ProRS alone. Some proteins, e.g., GDF15, exhibited important discriminative values for various diseases. We also showed that the good discriminative performance observed could be largely translated into practical clinical utility. Taken together, proteomic profiles may serve as a replacement for complex laboratory tests or clinical measures to refine the comprehensive risk assessments of multiple diseases and mortalities simultaneously. Our models were internally validated in the UK Biobank; thus, further independent external validations are necessary to confirm our findings before application in clinical settings.

Behavioral representational similarity analysis reveals how episodic learning is influenced by and reshapes semantic memory

While semantic and episodic memory have been shown to influence each other, uncertainty remains as to how this interplay occurs. We introduce a behavioral representational similarity analysis approach to assess whether semantic space can be subtly re-sculpted by episodic learning. Eighty participants learned word pairs that varied in semantic relatedness, and learning was bolstered via either testing or restudying. Next-day recall is superior for semantically related pairs, but there is a larger benefit of testing for unrelated pairs. Analyses of representational change reveal that successful recall is accompanied by a pulling together of paired associates, with cue words in semantically related (but not unrelated) pairs changing more across learning than target words. Our findings show that episodic learning is associated with systematic and asymmetrical distortions of semantic space which improve later recall by making cues more predictive of targets, reducing interference from potential lures, and establishing novel connections within pairs.

Proteomic investigations of adult polyglucosan body disease: insights into the pathobiology of a neurodegenerative disorder

Inadequate glycogen branching enzyme 1 (GBE1) activity results in different forms of glycogen storage disease type IV, including adult polyglucosan body disorder (APBD). APBD is clinically characterized by adult-onset development of progressive spasticity, neuropathy, and neurogenic bladder and is histologically characterized by the accumulation of structurally abnormal glycogen (polyglucosan bodies) in multiple cell types. How insufficient GBE1 activity causes the disease phenotype of APBD is poorly understood. We hypothesized that proteomic analysis of tissue from GBE1-deficient individuals would provide insights into GBE1-mediated pathobiology. In this discovery study, we utilized label-free LC-MS/MS to quantify the proteomes of lymphoblasts from 3 persons with APBD and 15 age- and gender-matched controls, with validation of the findings by targeted MS. There were 531 differentially expressed proteins out of 3,427 detected between APBD subjects vs. controls, including pronounced deficiency of GBE1. Bioinformatic analyses indicated multiple canonical pathways and protein-protein interaction networks to be statistically markedly enriched in APBD subjects, including: RNA processing/transport/translation, cell cycle control/replication, mTOR signaling, protein ubiquitination, unfolded protein and endoplasmic reticulum stress responses, glycolysis and cell death/apoptosis. Dysregulation of these processes, therefore, are primary or secondary factors in APBD pathobiology in this model system. Our findings further suggest that proteomic analysis of GBE1 mutant lymphoblasts can be leveraged as part of the screening for pharmaceutical agents for the treatment of APBD.

Identifying relapse predictors in individual participant data with decision trees

BACKGROUND

Depression is a highly common and recurrent condition. Predicting who is at most risk of relapse or recurrence can inform clinical practice. Applying machine-learning methods to Individual Participant Data (IPD) can be promising to improve the accuracy of risk predictions.

METHODS

Individual data of four Randomized Controlled Trials (RCTs) evaluating antidepressant treatment compared to psychological interventions with tapering ([Formula: see text]) were used to identify predictors of relapse and/or recurrence. Ten baseline predictors were assessed. Decision trees with and without gradient boosting were applied. To study the robustness of decision-tree classifications, we also performed a complementary logistic regression analysis.

RESULTS

The combination of age, age of onset of depression, and depression severity significantly enhances the prediction of relapse risk when compared to classifiers solely based on depression severity. The studied decision trees can (i) identify relapse patients at intake with an accuracy, specificity, and sensitivity of about 55% (without gradient boosting) and 58% (with gradient boosting), and (ii) slightly outperform classifiers that are based on logistic regression.

CONCLUSIONS

Decision tree classifiers based on multiple-rather than single-risk indicators may be useful for developing treatment stratification strategies. These classification models have the potential to contribute to the development of methods aimed at effectively prioritizing treatment for those individuals who require it the most. Our results also underline the existing gaps in understanding how to accurately predict depressive relapse.

Missing data imputation, prediction, and feature selection in diagnosis of vaginal prolapse

BACKGROUND

Data loss often occurs in the collection of clinical data. Directly discarding the incomplete sample may lead to low accuracy of medical diagnosis. A suitable data imputation method can help researchers make better use of valuable medical data.

METHODS

In this paper, five popular imputation methods including mean imputation, expectation-maximization (EM) imputation, K-nearest neighbors (KNN) imputation, denoising autoencoders (DAE) and generative adversarial imputation nets (GAIN) are employed on an incomplete clinical data with 28,274 cases for vaginal prolapse prediction. A comprehensive comparison study for the performance of these methods has been conducted through certain classification criteria. It is shown that the prediction accuracy can be greatly improved by using the imputed data, especially by GAIN. To find out the important risk factors to this disease among a large number of candidate features, three variable selection methods: the least absolute shrinkage and selection operator (LASSO), the smoothly clipped absolute deviation (SCAD) and the broken adaptive ridge (BAR) are implemented in logistic regression for feature selection on the imputed datasets. In pursuit of our primary objective, which is accurate diagnosis, we employed diagnostic accuracy (classification accuracy) as a pivotal metric to assess both imputation and feature selection techniques. This assessment encompassed seven classifiers (logistic regression (LR) classifier, random forest (RF) classifier, support machine classifier (SVC), extreme gradient boosting (XGBoost) , LASSO classifier, SCAD classifier and Elastic Net classifier)enhancing the comprehensiveness of our evaluation.

RESULTS

The proposed framework imputation-variable selection-prediction is quite suitable to the collected vaginal prolapse datasets. It is observed that the original dataset is well imputed by GAIN first, and then 9 most significant features were selected using BAR from the original 67 features in GAIN imputed dataset, with only negligible loss in model prediction. BAR is superior to the other two variable selection methods in our tests.

CONCLUDES

Overall, combining the imputation, classification and variable selection, we achieve good interpretability while maintaining high accuracy in computer-aided medical diagnosis.

Evaluating Proteomics Imputation Methods with Improved Criteria

Quantitative measurements produced by tandem mass spectrometry proteomics experiments typically contain a large proportion of missing values. Missing values hinder reproducibility, reduce statistical power, and make it difficult to compare across samples or experiments. Although many methods exist for imputing missing values, in practice, the most commonly used methods are among the worst performing. Furthermore, previous benchmarking studies have focused on relatively simple measurements of error such as the mean-squared error between imputed and held-out values. Here we evaluate the performance of commonly used imputation methods using three practical, "downstream-centric" criteria. These criteria measure the ability to identify differentially expressed peptides, generate new quantitative peptides, and improve the peptide lower limit of quantification. Our evaluation comprises several experiment types and acquisition strategies, including data-dependent and data-independent acquisition. We find that imputation does not necessarily improve the ability to identify differentially expressed peptides but that it can identify new quantitative peptides and improve the peptide lower limit of quantification. We find that MissForest is generally the best performing method per our downstream-centric criteria. We also argue that existing imputation methods do not properly account for the variance of peptide quantifications and highlight the need for methods that do.

Hepatic lipid-associated macrophages mediate the beneficial effects of bariatric surgery against MASH

For patients with obesity and metabolic syndrome, bariatric procedures such as vertical sleeve gastrectomy (VSG) have a clear benefit in ameliorating metabolic dysfunction-associated steatohepatitis (MASH). While the effects of bariatric surgeries have been mainly attributed to nutrient restriction and malabsorption, whether immuno-modulatory mechanisms are involved remains unclear. Here we report that VSG ameliorates MASH progression in a weight loss-independent manner. Single-cell RNA sequencing revealed that hepatic lipid-associated macrophages (LAMs) expressing the triggering receptor expressed on myeloid cells 2 (TREM2) increase their lysosomal activity and repress inflammation in response to VSG. Remarkably, TREM2 deficiency in mice ablates the reparative effects of VSG, suggesting that TREM2 is required for MASH resolution. Mechanistically, TREM2 prevents the inflammatory activation of macrophages and is required for their efferocytotic function. Overall, our findings indicate that bariatric surgery improves MASH through a reparative process driven by hepatic LAMs, providing insights into the mechanisms of disease reversal that may result in new therapies and improved surgical interventions.