Assess Prediction Accuracy Research Articles

A Comprehensive Evaluation of Machine Learning Algorithms for Digital Soil Organic Carbon Mapping on a National Scale

The aim of this study was to narrow the research gap of ambiguity in which machine learning algorithms should be selected for evaluation in digital soil organic carbon (SOC) mapping. This was performed by providing a comprehensive assessment of prediction accuracy for 15 frequently used machine learning algorithms in digital SOC mapping based on studies indexed in the Web of Science Core Collection (WoSCC), providing a basis for algorithm selection in future studies. Two study areas, including mainland France and the Czech Republic, were used in the study based on 2514 and 400 soil samples from the LUCAS 2018 dataset. Random Forest was first ranked for France (mainland) and then ranked for the Czech Republic regarding prediction accuracy; the coefficients of determination were 0.411 and 0.249, respectively, which was in accordance with its dominant appearance in previous studies indexed in the WoSCC. Additionally, the K-Nearest Neighbors and Gradient Boosting Machine regression algorithms indicated, relative to their frequency in studies indexed in the WoSCC, that they are underrated and should be more frequently considered in future digital SOC studies. Future studies should consider study areas not strictly related to human-made administrative borders, as well as more interpretable machine learning and ensemble machine learning approaches.

An efficient computational investigation on accurate daily soil temperature prediction using boosting ensemble methods explanation based on SHAP importance analysis

Accurately predicting soil temperature (Ts) serves as the foundation of geothermal applications, modern irrigation strategies in arid agricultural landscapes, and understanding ecosystem changes. Also, this parameter is crucial for estimating crop water requirements, thereby enabling efficient management of scarce water resources in these moisture-limited environments. Therefore, the development of a sophisticated intelligent algorithm based on boosting ensemble including XGBoost, CatBoost, LightGBM, and AdaBoost models which incorporates nine different scenarios from meteorological variables as input parameters, offers valuable insights into subsurface thermal dynamics at various depths for accurately predicting thermal gradients within the soil profile, is imperative for soil science investigations. To unravel the underlying mechanisms influencing the models' predictions of Ts, the SHapley Additive exPlanations (SHAP) methodology was employed. This algorithm quantifies the contributory significance of each input variable and facilitates the explication of input-output dependencies. In this study, the model's performance was evaluated at two meteorological monitoring sites, denoted as Penjwen and Bazian, situated within the geopolitical boundaries of the Kurdistan region in Iraq using a suite of statistical indicators, including the correlation coefficient (R), the root mean square error (RMSE), the Nash-Sutcliffe efficiency (NSE), and the mean absolute error (MAE). These metrics provided a comprehensive assessment of the model's predictive accuracy and reliability. Moreover, this localized analysis provided insights into the model's efficacy under specific regional climatic conditions. At station Penjwen, the results based on the RMSE values indicated that the LightGBM and CatBoost methods performed better than other models in Ts estimation at depths of 5 cm and 10 cm, with RMSE values of 2.502 °C and 2.164 °C, respectively. Also, at station Bazian, CatBoost and LightGBM approaches showed the best results at depths of 5 cm and 10 cm, with RMSE values of 2.069 °C and 1.786 °C, respectively. The study's findings suggest that meteorological variables can serve as effective inputs for predicting Ts using the proposed algorithms.

Genomic selection in western redcedar: from proof of concept to operational application.

Forests face many threats. While traditional breeding may be too slow to deliver well-adapted trees, genomic selection (GS) can accelerate the process. We describe a comprehensive study of GS from proof of concept to operational application in western redcedar (WRC, Thuja plicata). Using genomic data, we developed models on a training population (TrP) of trees to predict breeding values (BVs) in a target seedling population (TaP) for growth, heartwood chemistry, and foliar chemistry traits. We used cross-validation to assess prediction accuracy (PACC) in the TrP; we also validated models for early-expressed foliar traits in the TaP. Prediction accuracy was high across generations, environments, and ages. PACC was not reduced to zero among unrelated individuals in TrP and was only slightly reduced in the TaP, confirming strong linkage disequilibrium and the ability of the model to generate accurate predictions across breeding generations. Genomic BV predictions were correlated with those from pedigree but displayed a wider range of within-family variation due to the ability of GS to capture the Mendelian sampling term. Using predicted TaP BVs in multi-trait selection, we functionally implemented and integrated GS into an operational tree-breeding program.

A Sustainable Development for Building Energy Consumption Based on Improved Rafflesia Optimization Algorithm with Feature Selection and Ensemble Deep Learning

Buildings emit a great deal of carbon dioxide and use a lot of energy. The study of building energy consumption is useful for the sustainable development of multi-energy planning and energy-saving strategies. Therefore, a sustainable development for building energy consumption based on the improved rafflesia optimization algorithm (ROA) with feature selection and ensemble deep learning is proposed in this paper. This method can explore data on building energy usage, assess prediction accuracy, and address concerns that building energy usage research must address. The proposed model first uses an improved self-organizing map with a new neighborhood function to select important features. After that, it uses ensemble deep learning to accurately anticipate the building’s energy usage. In addition, the improved ROA is used to fine-tune parameters for feature selection and ensemble deep learning. This research uses the dataset of the American Society of Heating, Refrigerating, and Air Conditioning Engineers (ASHRAE) to compare the performance of several modeling approaches. It identifies the top five most important features based on the model’s results. Furthermore, the proposed model can be successfully applied to a real-world application. They both have the lowest root mean squared errors among the approaches examined. The proposed model indeed provides the benefits of feature selection and ensemble deep learning with the improved ROA for the prediction of building energy consumption.

Utilizing a CNN-RNN machine learning approach for forecasting time-series outlet fluid temperature monitoring by long-term operation of BHEs system

The Borehole Heat Exchanger (BHE) plays a pivotal role in enhancing heat exchange efficiency within Ground Source Heat Pump (GSHP) systems. The accurate prediction of the BHE's outlet fluid temperature is crucial for optimizing GSHP performance, energy storage, and resource conservation. However, conventional machine learning methods encounter challenges in manual feature extraction, learning complex nonlinear relationships, and adapting to real-world scenarios. To address these limitations, this research proposes a crossbreed model integrating Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) architectures to forecast long-term outlet fluid temperature in BHE systems. The model framework encompasses data preprocessing, utilizing refined data in the CNN module for temporal feature extraction, subsequently passed to the RNN module to capture sequential and temporal patterns from each dataset. Specifically, the advanced CNN-RNN architecture is designed to establish a comprehensive input-output mapping, leveraging essential input features such as inlet fluid, ambient air, and subsurface temperatures at varying depths (0, 10, and 20 m). Performance evaluation metrics, including R2, RMSE, MAE, and AARE, are employed to compare and assess prediction accuracy across various models, including LSTM, CNN, and SimpleRNN. The obtained results demonstrate the superior performance of the proposed model, achieving an RSME of 0.818, MAE of 0.642, AARE of 0.0305, and an R2 value of 98.75 %. This surpasses the performance of traditional prediction models (LSTM, CNN, and SimpleRNN) by 3.01 %, 5.80 %, and 19.52 %, respectively. Notably, the remarkably low MAE of 0.642 exhibited by a CNN-RNN model underscores its capability to outperform traditional approaches, especially when handling large datasets. These findings emphasize the significance of the developed model in facilitating efficient operation, positioning it as a valuable tool for advancing the long-term sustainability of BHE systems.

Evaluation of electrical load demand forecasting using various machine learning algorithms

The energy sector heavily relies on a diverse array of machine learning algorithms for power load prediction, which plays a pivotal role in shaping policies for power generation and distribution. The precision of power load prediction depends on numerous factors that reflect nonlinear traits within the data. Notably, machine learning algorithms and artificial neural networks have emerged as indispensable components in contemporary power load forecasting. This study focuses specifically on machine learning algorithms, encompassing support vector machines (SVMs), long short-term memory (LSTM), ensemble classifiers, recurrent neural networks, and deep learning methods. The research meticulously examines short-term power load prediction by leveraging Chandigarh UT electricity utility data spanning the last 5 years. The assessment of prediction accuracy utilizes metrics such as normalized mean square error (NMSE), root mean squared error (RMSE), mean absolute error (MAE), and mutual information (MI). The prediction results demonstrate superior performance in LSTM compared to other algorithms, with the prediction error being the lowest in LSTM and 13.51% higher in SVMs. These findings provide valuable insights into the strengths and limitations of different machine learning algorithms. Validation experiments for the proposed method are conducted using MATLAB R2018 software.

Fast Decision-Tree-Based Series Partitioning and Mode Prediction Termination Algorithm for H.266/VVC

With the advancement of network technology, multimedia videos have emerged as a crucial channel for individuals to access external information, owing to their realistic and intuitive effects. In the presence of high frame rate and high dynamic range videos, the coding efficiency of high-efficiency video coding (HEVC) falls short of meeting the storage and transmission demands of the video content. Therefore, versatile video coding (VVC) introduces a nested quadtree plus multi-type tree (QTMT) segmentation structure based on the HEVC standard, while also expanding the intra-prediction modes from 35 to 67. While the new technology introduced by VVC has enhanced compression performance, it concurrently introduces a higher level of computational complexity. To enhance coding efficiency and diminish computational complexity, this paper explores two key aspects: coding unit (CU) partition decision-making and intra-frame mode selection. Firstly, to address the flexible partitioning structure of QTMT, we propose a decision-tree-based series partitioning decision algorithm for partitioning decisions. Through concatenating the quadtree (QT) partition division decision with the multi-type tree (MT) division decision, a strategy is implemented to determine whether to skip the MT division decision based on texture characteristics. If the MT partition decision is used, four decision tree classifiers are used to judge different partition types. Secondly, for intra-frame mode selection, this paper proposes an ensemble-learning-based algorithm for mode prediction termination. Through the reordering of complete candidate modes and the assessment of prediction accuracy, the termination of redundant candidate modes is accomplished. Experimental results show that compared with the VVC test model (VTM), the algorithm proposed in this paper achieves an average time saving of 54.74%, while the BDBR only increases by 1.61%.

Development of a methodology for empirical scientific research using standard libraries of automatic machine learning on the example of predicting oil and gas reservoirs in the Tomsk Region

A methodology for applying machine learning to predict oil and gas reservoirs with a specified accuracy has been developed and illustrated using examples. The methodology includes principles of working with datasets, methods for interpreting the obtained models, methods for filtering local data noise, empirical methods for assessing prediction accuracy, and a method for calculating prediction accuracy based on Bayes' theorem. This work may be of interest to specialists in the field of logic and methodology of natural sciences, petroleum and gas geology, design of geological exploration works, design of oil and gas field development, as well as specialists in data analytics and artificial intelligence in the oil and gas industry.

ДИАГНОСТИЧЕСКАЯ ЦЕННОСТЬ МОДЕЛИ ПРОГНОЗИРОВАНИЯ ПРИ ОБСТРУКЦИИ ВЕРХНИХ МОЧЕВЫВОДЯЩИХ ПУТЕЙ

Patients with urological pathology often develop supravesical obstruction (SVO), which affects the renal parenchyma and causes disruption of the normal renal function and various complications, such as chronic renal disease and urosepsis, which can lead to death. Prediction accuracy assessment for this group of patients is challenging, making it difficult to draw firm conclusions Objective: To develop predictive models for choosing rational treatment tactics in SVO. Methods: Data from 655 patients aged 4 to 86 (average age 39.56±17.23 years) admitted to the Republican Specialized Scientific and Practical Medical Center of Urology in 2021-2023 with SVO were analyzed. There were 350 men (53.4%) and 305 women (46.6%) enrolled in the study. Patients were divided into groups according to the primary diagnoses: urolithiasis (UL) – 231 (35.3%), anomalies of the upper urinary tract (UUT) – 332 (50.7%), acquired diseases of the ureter – 92 (14.0%). Results: Critical Z-values (cut-off points) were calculated for each group, determining the complicated course of SVO. In the UL group, the Z-value was 1.910; in the group of patients with anomalies of the UUT, ureteropelvic stricture (UPS) – 1.998, ureteral stricture – 1.239, ureterocele – 1.894; in the group of patients with acquired diseases of the ureter, secondary ureteral strictures – 1.209, ureteral obliteration – 1.713. Conclusion: Discriminant prediction models showed high sensitivity and specificity for choosing the optimal tactical approach in patients with complicated SVO. Keywords: Supravesical obstruction, hydronephrosis, prediction model, management tactics.

External Validation of a Dynamic Prediction Model for Upper Limb Function After Stroke

ObjectiveTo externally validate the dynamic prediction model for prediction of upper limb (UL) function 6 months after stroke. The dynamic prediction model has been developed and cross-validated on data from 4 Dutch studies. DesignData from a prospective Danish cohort study were used to assess prediction accuracy. SettingA Danish neurorehabilitation hospital. ParticipantsIn this external validation study, follow-up data for 80 patients in the subacute phase after stroke (N=80), mean age 64 (SD11), 43% women, could be obtained. They were assessed at 2 weeks, 3 months, and 6 months after stroke with the Action Research Arm Test (ARAT), Fugl-Meyer Motor Assessment upper limb (FMA), and Shoulder Abduction (SA) Finger Extension (FE), (SAFE) test. InterventionNot applicable. Main Outcome MeasuresPrediction accuracy at 6 months was examined for 3 categories of ARAT (0-57 points): mild (48-57), moderate (23-47), and severe (0-22). Two individual predictions of ARAT scores at ±6 months post-stroke were computed based on, respectively, baseline (2 weeks) and 3 months ARAT, FE, SA values. The absolute individual differences between observed and predicted ARAT scores were summarized. ResultsThe prediction model performed best for patients with relatively good UL motor function, with an absolute error median (IQR) of 3 (2-9), and worst for patients with severe UL impairment, with a median (IQR) of 30 (3-39) at baseline. In general, prediction accuracy substantially improved when data obtained 3 months after stroke was included compared with baseline at 2 weeks after stroke. ConclusionWe found limited clinical usability due to the lack of prediction accuracy 2 weeks after stroke and for patients with severe UL impairments. The dynamic prediction model could probably be refined with data from biomarkers.

A multiscale accuracy assessment of moisture content predictions using time-lapse electrical resistivity tomography in mine tailings

Accurate and large-scale assessment of volumetric water content (VWC) plays a critical role in mining waste monitoring to mitigate potential geotechnical and environmental risks. In recent years, time-lapse electrical resistivity tomography (TL-ERT) has emerged as a promising monitoring approach that can be used in combination with traditional invasive and point-measurements techniques to estimate VWC in mine tailings. Moreover, the bulk electrical conductivity (EC) imaged using TL-ERT can be converted into VWC in the field using petrophysical relationships calibrated in the laboratory. This study is the first to assess the scale effect on the accuracy of ERT-predicted VWC in tailings. Simultaneous and co-located monitoring of bulk EC and VWC are carried out in tailings at five different scales, in the laboratory and in the field. The hydrogeophysical datasets are used to calibrate a petrophysical model used to predict VWC from TL-ERT data. Overall, the accuracy of ERT-predicted VWC is pm 0.03~textrm{m}^3/textrm{m}^3, and the petrophysical models determined at sample-scale in the laboratory remain valid at larger scales. Notably, the impact of temperature and pore water EC evolution plays a major role in VWC predictions at the field scale (tenfold reduction of accuracy) and, therefore, must be properly taken into account during the TL-ERT data processing using complementary hydrogeological sensors. Based on these results, we suggest that future studies using TL-ERT to predict VWC in mine tailings could use sample-scale laboratory apparatus similar to the electrical resistivity Tempe cell presented here to calibrate petrophysical models and carefully upscale them to field applications.

Integral simulation of the first 3 weeks of severe accident at Fukushima Daiichi Unit 1 with SOCRAT code. Part I – Qualification of model and simulation of the initial phase

This paper is the first part of the summary overview which presents the results of IBRAE's work over the last 11 years to study the severe accident at Unit 1 of Fukushima Daiichi Nuclear Power Station. The paper is dedicated to the description of the integral model of the Unit 1 with assumptions made, and to its qualification with the available data. The qualification includes the comparison of simulation results with measurements for the initial phase of the accident which starts with the initiating event (earthquake) and ends with tsunami arrival. The important issues discussed are the direct modeling of isolation condenser performance and verification of the possibility to use an integral approach to estimate the nuclide inventory in the core. These questions are of methodological importance for modeling the main phase of the accident.Direct modeling of isolation condenser as part of integral calculation allows verifying the hypothesis about the isolating valves left open inside containment after tsunami arrival and station blackout. The isolation condenser system could influence the development of severe accident by limiting the pressure peak in the reactor at the time of melt relocation into the lower plenum. Thus, verification of the isolation condenser model at the initial stage of the accident justifies the correctness of results obtained for the core degradation phase of the accident.The integral approach in nuclide inventory estimate presumes the need to mix the fuel with different burn-ups in the effective groups of fuel assemblies, which may bring an uncertainty in the nuclide inventory prediction. Quantification of this uncertainty is important for assessment of prediction accuracy for the source term.

Comparative Assessment of Pose Prediction Accuracy in RNA-Ligand Docking.

Structure-based virtual high-throughput screening is used in early-stage drug discovery. Over the years, docking protocols and scoring functions for protein-ligand complexes have evolved to improve the accuracy in the computation of binding strengths and poses. In the past decade, RNA has also emerged as a target class for new small-molecule drugs. However, most ligand docking programs have been validated and tested for proteins and not RNA. Here, we test the docking power (pose prediction accuracy) of three state-of-the-art docking protocols on 173 RNA-small molecule crystal structures. The programs are AutoDock4 (AD4) and AutoDock Vina (Vina), which were designed for protein targets, and rDock, which was designed for both protein and nucleic acid targets. AD4 performed relatively poorly. For RNA targets for which a crystal structure of a bound ligand used to limit the docking search space is available and for which the goal is to identify new molecules for the same pocket, rDock performs slightly better than Vina, with success rates of 48% and 63%, respectively. However, in the more common type of early-stage drug discovery setting, in which no structure of a ligand-target complex is known and for which a larger search space is defined, rDock performed similarly to Vina, with a low success rate of ∼27%. Vina was found to have bias for ligands with certain physicochemical properties, whereas rDock performs similarly for all ligand properties. Thus, for projects where no ligand-protein structure already exists, Vina and rDock are both applicable. However, the relatively poor performance of all methods relative to protein-target docking illustrates a need for further methods refinement.

Abstract 12423: CAD Polygenic Risk Score, a Novel Screening Tool for Predicting Likelihood of Coronary Artery Disease, Was Not Superior to Coronary Calcium Score in Predicting MACE

Introduction: Coronary artery calcium (CAC) score is a well-established tool for risk assessment of coronary artery disease (CAD). Alternatively, CAD polygenic risk score (PRS) is an emerging modality for assessing risk of CAD. Our study aims to compare the sensitivity and specificity of the above scores in predicting major adverse cardiovascular events (MACE). Hypothesis: CAD PRS is a better predictor of MACE than CAC score. Methods: Adult patients had CAD PRS, CAC and baseline characteristics documented through the Sanford Heart Screening Program from 2011 to 2021. Incidence of MACE, defined as myocardial infarction, stroke, death, and revascularization, was analyzed. Chi-squared tests were used to examine the impact of stratified risk scores and MACE. A logistic regression model was trained on 70% of the data to assess prediction accuracy on 30% of the data. Results: A total of 1,380 patients were analyzed. Mean age was 63, with female (65%) and Caucasian (99%) predominance. The prevalence of diabetes mellitus, hyperlipidemia, and hypertension was 8%, 32%, and 7%, respectively. The incidence of MACE was 6.6%. Patients with MACE had higher CAC (527.1±856.2 vs. 98.3±294.3) and CAD PRS (15.5±.62 vs. 15.24±.61). Sensitivity and specificity are represented by the receiver operating characteristic curve, which was used to assess prediction accuracy whereby area under the curve (AUC) was used as a measure of fit ranging from 0 to 1. The curves showed that CAC had better AUC than PRS (Figure 1). Conclusions: In our study CAD PRS failed to be a better predictor of MACE compared to CAC score.

Prediction of Covid-19 Multiparametric Biomarkers and Drug Target of Patients for Risk Stratification Using Machine Learning Approach

In the situation of Coronavirus disease 2019 (COVID-19), forecasting disease progression and identifying therapeutic drug targets is critical, especially given the nonattendance of a viable approach for treating severe cases. The preparation cohort revealed promising biomarkers, which were then precisely measured and employed to assess prediction accuracy across validation cohorts. This approach holds significant potential in enhancing understanding of severe COVID-19 and aiding the development of effective treatments. However, ultrasound-guided MRI (US-MRI) is an emerging modality that can noninvasively acquire multi-parametric information on COVID-19 and function without the need for contrast agents. This shows that neural network analysis of US-MRI transports exclusive prognosis data and this significantly improved prognosis performance. Consequently, the research proposed a deep neural network model of an Ensemble Multi-Relational Graph Neural Network (EMR-GNN) to determine the optimal model for predicting vascular biomarkers (CRP, IL-6, ferritin). In the nonappearance of a tailored treatment for this emerging virus, scientists are actively investigating various strategies to curb its replication. This work focuses on identifying potential drug targets, drawing from proteins abundant in lung material and those targeted by FDA-approved drugs as catalogued in HPA. This effort reflects a broader initiative within the methodical unrestricted to develop effective means of limiting virus replication. Accordingly, recognized five lung-improved proteins, comprising MRC1, SG3A1, CCL18, histone H4, and CLEC3B, were annotated as “drug targets”. For this, the researcher proposes a Heterogeneous Graph Structural Attention Neural Network (HGS-ANN) model to learn topological information of composite molecules and a Dilated Causal CNN-LSTM model with U-Net layers for modelling spatial-sequential information in Simplified Molecular-Input Line-Entry System (SMILES) sequences of drug data. The COVID-19 datasets are downloaded from the GEO database. These data are evaluated using Matlab software. The proposed work evaluated that the AUC of the work is 0.995, however, the AUC is measured based on sex, age, and chronic diseases. This model has a 0.933 accuracy in the subgroup of slices thicker than 1mm. However, the AUC curve and the classification outcome of the proposed method are compared with the existing rad model, deeper, and KNN models. In comparison to existing methods, the proposed model demonstrates superior performance. This research not only identifies potential therapeutic targets nonetheless also serves to uncover biomarkers crucial for comprehending the pathogenesis of undecorated COVID-19.

Predictive Modeling and Integrated Risk Assessment of Postoperative Mortality and Pneumonia in Traumatic Brain Injury Patients through Clustering and Machine Learning: Retrospective Study.

This study harnessed machine learning to forecast postoperative mortality (POM) and postoperative pneumonia (PPN) among surgical traumatic brain injury (TBI) patients. Our analysis centered on the following key variables: Glasgow Coma Scale (GCS), midline brain shift (MSB), and time from injury to emergency room arrival (TIE). Additionally, we introduced innovative clustered variables to enhance predictive accuracy and risk assessment. Exploring data from 617 patients spanning 2012 to 2022, we observed that 22.9% encountered postoperative mortality, while 30.0% faced postoperative pneumonia (PPN). Sensitivity for POM and PPN prediction, before incorporating clustering, was in the ranges of 0.43-0.82 (POM) and 0.54-0.76 (PPN). Following clustering, sensitivity values were 0.47-0.76 (POM) and 0.61-0.77 (PPN). Accuracy was in the ranges of 0.67-0.76 (POM) and 0.70-0.81 (PPN) prior to clustering and 0.42-0.73 (POM) and 0.55-0.73 (PPN) after clustering. Clusters characterized by low GCS, small MSB, and short TIE exhibited a 3.2-fold higher POM risk compared to clusters with high GCS, small MSB, and short TIE. In summary, leveraging clustered variables offers a novel avenue for predicting POM and PPN in TBI patients. Assessing the amalgamated impact of GCS, MSB, and TIE characteristics provides valuable insights for clinical decision making.

Construction and validation of a nomogram for predicting diabetes remission at 3 months after bariatric surgery in patients with obesity combined with type 2 diabetes mellitus.

Bariatric metabolic surgery (BMS) is a proven treatment option for patients with both obesity and type 2 diabetes mellitus (T2DM). However, there is a lack of comprehensive reporting on the short-term remission rates of diabetes, and the existing data are inadequate. Hence, this study aimed to investigate the factors that may contribute to diabetes remission (DR) in patients with obesity and T2DM, 3 months after undergoing BMS. Furthermore, our objective was to develop a risk-predicting model using a nomogram. In total, 389 patients with obesity and T2DM, who had complete preoperative information and underwent either laparoscopic sleeve gastrectomy or laparoscopic gastric bypass surgery between January 2014 and May 2023, were screened in the Chinese Obesity and Metabolic Surgery Database. The patients were randomly divided into a training set (n = 272) and a validation set (n = 117) in a 7:3 ratio. Potential factors for DR were analysed through univariate and multivariate logistic regression analyses and then modelled using a nomogram. The model's performance was evaluated using receiver operating characteristic curves and the area under the curve (AUC). Calibration plots were used to assess prediction accuracy and decision curve analyses were conducted to evaluate the clinical usefulness of the model. Glycated haemoglobin, triglycerides, duration of diabetes, insulin requirement and hypercholesterolaemia were identified as independent factors influencing DR. We have incorporated these five indicators into a nomogram, which has shown good efficacy in both the training cohort (AUC = 0.930) and validation cohort (AUC = 0.838). The calibration plots indicated that the model fits well in both the training and the validation cohorts, and decision curve analyses showed that the model had good clinical applicability. The prediction model developed in this study holds predictive value for short-term DR following BMS in patients with obesity and T2DM.

Genomic prediction and association mapping of maize grain yield in multi-environment trials based on reaction norm models.

Genotype-by-environment interaction (GEI) is among the greatest challenges for maize breeding programs. Strong GEI limits both the prediction of genotype performance across variable environmental conditions and the identification of genomic regions associated with grain yield. Incorporating GEI into yield prediction models has been shown to improve prediction accuracy of yield; nevertheless, more work is needed to further understand this complex interaction across populations and environments. The main objectives of this study were to: 1) assess GEI in maize grain yield based on reaction norm models and predict hybrid performance across a gradient of environmental (EG) conditions and 2) perform a genome-wide association study (GWAS) and post-GWAS analyses for maize grain yield using data from 2014 to 2017 of the Genomes to Fields initiative hybrid trial. After quality control, 2,126 hybrids with genotypic and phenotypic data were assessed across 86 environments representing combinations of locations and years, although not all hybrids were evaluated in all environments. Heritability was greater in higher-yielding environments due to an increase in genetic variability in these environments in comparison to the low-yielding environments. GWAS was carried out for yield and five single nucleotide polymorphisms (SNPs) with the highest magnitude of effect were selected in each environment for follow-up analyses. Many candidate genes in proximity of selected SNPs have been previously reported with roles in stress response. Genomic prediction was performed to assess prediction accuracy of previously tested or untested hybrids in environments from a new growing season. Prediction accuracy was 0.34 for cross validation across years (CV0-Predicted EG) and 0.21 for cross validation across years with only untested hybrids (CV00-Predicted EG) when compared to Best Linear Unbiased Prediction (BLUPs) that did not utilize genotypic or environmental relationships. Prediction accuracy improved to 0.80 (CV0-Predicted EG) and 0.60 (CV00-Predicted EG) when compared to the whole-dataset model that used the genomic relationships and the environmental gradient of all environments in the study. These results identify regions of the genome for future selection to improve yield and a methodology to increase the number of hybrids evaluated across locations of a multi-environment trial through genomic prediction.

A Ferroptosis-Related lncRNAs Signature Predicts Prognosis of Colon Adenocarcinoma.

(1) Ferroptosis is a type of cellular death caused by lipid-dependent iron peroxide, which plays a major role in cancer. Long noncoding RNAs (lncRNAs) are increasingly recognized as key regulating substances in ferroptosis; (2) RNA sequencing expressions and clinical data of 519 patients with colon adenocarcinoma (COAD) were downloaded from The Cancer Genome Atlas (TCGA) database. The expression levels of lncRNAs related to ferroptosis were screened with Pearson correlation analysis. Differential genes were enriched with Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. LncRNAs related to ferroptosis were determined with univariate Cox regression and multivariate Cox regression analyses, and patients with COAD were classified into high- and low-risk subgroups according to their median risk score. The prognostic value was further examined, and the association between ferroptosis-related lncRNAs (frlncRNAs) and survival in patients with high and low risks of COAD was validated. A TCGA-COAD data set was used for receiver operating characteristic (ROC) analysis and detrended correspondence analysis (DCA) to assess prediction accuracy. Finally, a nomogram was constructed to predict survival probability; (3) We obtained a model consisting of a five-frlncRNAs signature comprising AP003555.1, AP001469.3, ITGB1-DT, AC129492.1, and AC010973.2 for determining the overall survival (OS) of patients with COAD. The survival analysis and ROC curves showed that the model had good robustness and predictive performance on the TCGA training set; (4) We found that a five-frlncRNAs signature may play a potential role in anti-COAD immunity. Risk characteristics based on frlncRNAs can accurately predict the prognosis and immunotherapy response of patients with COAD.

Improving the accuracy of daily solar radiation prediction by climatic data using an efficient hybrid deep learning model: Long short-term memory (LSTM) network coupled with wavelet transform

Accurate daily solar radiation prediction is a crucial task for the management and generation of solar energy as one of the alternatives to fossil fuels. In this study, the prediction accuracy of new machine learning methods, wavelet long short-term memory (WLSTM), wavelet multi-layer perceptron artificial neural network (WMLPANN), long short-term memory (LSTM), multi-layer perceptron artificial neural network (MLPANN), and multivariate adaptive regression splines (MARS), was assessed for modeling daily solar radiation using various input combinations of climatic data of maximum and minimum relative humidity, potential evapotranspiration, maximum and minimum temperature, precipitation and wind speed from two stations, Brownstown and Carbondale located in Illinois, USA. For accurate assessment of prediction accuracy of the proposed models, four reliable statistical metrics, root mean squared error (RMSE), Nash–Sutcliffe efficiency coefficient (NSE), coefficient of determination (R2), and One-Tailed Wilcoxon Signed-Rank Test were employed. Comparison of results, based on the RMSE values, indicated that the WLSTM method performed better than the WMLPANN, LSTM, MLPANN and MARS methods in the estimation of solar radiation values at both stations. The average RMSE values of WMLPANN, LSTM, MLPANN, and MARS approaches was decreased by 6%, 4%, 7.3%, and 13.5% using WLSTM method at Brownstown Station, by 6%, 5.2%, 13.2%, and 14.3% at Carbondale Station, respectively. The overall results during the testing phase of both stations revealed the successful application of hybridization of the LSTM model with the wavelet transform technique for improving the prediction accuracy of solar radiation based on climatic parameters.