Lasso Regression Methods Research Articles

Research on the Driving Factors and Prediction Model of Urban Underground Space Demand in China

The development and utilization of urban underground space (UUS) have emerged as critical strategies to address the challenges posed by urban population growth and land resource depletion. Accurate prediction of UUS demand serves as the cornerstone for scientifically planning underground space and promoting sustainable urban development. In this study, statistical analysis methods were used to investigate the relationship between potential driving factors and UUS demand based on collected data from 16 cities in China. The identification of primary driving factors involves correlation, path, and determination coefficient analyses. Subsequently, univariate regression, multiple linear regression, and LASSO regression methods are employed to construct prediction models for UUS demand. Additionally, the link between historical data and UUS demand in each city was studied separately. The findings reveal that GDP per km2 and GDP per capita comprehensively capture the influence of urban population, economy, and transportation on UUS demand. Notably, GDP per km2 makes the most significant contribution to the proposed regression models, followed by GDP per capita. The application of LASSO regression proves effective in selecting potential factors while maximizing data utilization, presenting itself as a valuable auxiliary tool for UUS planning.

AIIB Investment and Economic Development of India: The Case of the Gujarat Road Project

The purpose of this study is to verify whether the transportation infrastructure investment carried out by the Asian Infrastructure Investment Bank (AIIB) has promoted the economic development of its recipient countries. Since the establishment of the AIIB, its investments in infrastructure development, aimed at promoting economic growth in Asian developing countries, have garnered considerable attention. This study selects India, the largest recipient country of the AIIB, as the research object and chooses the Gujarat Road Project as the research case, since it is a completed infrastructure construction investment project in the transportation field. This paper provides an overview of the project’s operation and summarizes key factors in the project’s implementation. In the data analysis section, the per capita GDP is selected as the explained variable to measure economic development, and the LASSO regression method is used to select several variables that affect economic development. Moreover, the random forest model is used to obtain the causal relationship between road construction and the per capita GDP from 2001 to 2022. The results indicate that road construction in India has a significant positive effect on per capita GDP growth, the Gujarat Road Project supported by the AIIB also has a positive effect on per capita GDP growth, and this effect is stronger than that at the national level. The main contribution of this work is the validation of the investment strategy of the AIIB and the quantification of the economic contribution of AIIB investment projects to the local area.

Identification and verification of a prognostic autophagy-related gene signature in hepatocellular carcinoma

This study aimed to investigate the potential of autophagy-related genes (ATGs) as a prognostic signature for HCC and explore their relationships with immune cells and immune checkpoint molecules. A total of 483 samples were collected from the GEO database (n = 115) and The Cancer Genome Atlas (TCGA) database (n = 368). The GEO dataset was used as the training set, while the TCGA dataset was used for validation. The list of ATGs was obtained from the human autophagy database (HADB). Using Cox regression and LASSO regression methods, a prognostic signature based on ATGs was established. The independent use of this prognostic signature was tested through subgroup analysis. Additionally, the predictive value of this signature for immune-related profiles was explored. Following selection through univariate Cox regression analysis and iterative LASSO Cox analysis, a total of 11 ATGs were used in the GEO dataset to establish a prognostic signature that stratified patients into high- and low-risk groups based on survival. The robustness of this prognostic signature was validated using an external dataset. This signature remained a prognostic factor even in subgroups with different clinical features. Analysis of immune profiles revealed that patients in the high-risk group exhibited immunosuppressive states characterized by lower immune scores and ESTIMATE scores, greater tumour purity, and increased expression of immune checkpoint molecules. Furthermore, this signature was found to be correlated with the infiltration of different immune cell subpopulations. The results suggest that the ATG-based signature can be utilized to evaluate the prognosis of HCC patients and predict the immune status within the tumour microenvironment (TME). However, it is important to note that this study represents a preliminary attempt to use ATGs as prognostic indicators for HCC, and further validation is necessary to determine the predictive power of this signature.

A nomogram to predict stoma creation in elective surgical resection for penetrating Crohn's disease.

The decision to perform a stoma during surgical resection of penetrating Crohn's disease (CD) is a critical consideration. The objective of this study was to identify factors that influence stoma creation and develop a predictive nomogram model to assist surgeons in making optimal surgical decisions. A retrospective study was conducted at a tertiary center from December 1, 2012, to December 1, 2021, involving consecutive patients with penetrating CD who underwent elective surgical resection. The LASSO regression method was used to select preoperative predictors, and a nomogram was constructed using multivariate logistic regression. The performance of the nomogram was validated in an internal cohort by assessing its discrimination, calibration, and clinical usefulness. The study included 405 cases in the training group and 135 cases in the validation group. Nine risk factors for stoma formation were identified, including disease location, fistula resulted from previous anastomosis, absence of preoperative exclusive enteral nutrition support, albumin levels below 35 g/L, C-reactive protein levels above 10 mg/L, hemoglobin levels below 100 g/L, perianal disease, internal fistula, and surgical approach. These risk factors were selected using the LASSO regression method, and a nomogram was developed based on them. The area under the curve and the coefficient of determination (R2) of the nomogram were 0.821 and 0.394, respectively. And the nomogram demonstrated good performance in the validation cohort. The nomogram exhibited good predictive ability for stoma formation during elective surgical resection for penetrating CD, which can assist surgeons in making informed clinical decisions.

Development of a stemness-related prognostic index to provide therapeutic strategies for bladder cancer

Bladder cancer (BC) is a heterogeneous disease with varying clinical outcomes. Recent evidence suggests that cancer progression involves the acquisition of stem-like signatures, and assessing stemness indices help uncover patterns of intra-tumor molecular heterogeneity. We used the one-class logistic regression algorithm to compute the mRNAsi for each sample in BLCA cohort. We subsequently classified BC patients into two subtypes based on 189 mRNAsi-related genes, using the unsupervised consensus clustering. Then, we identified nine hub genes to construct a stemness-related prognostic index (SRPI) using Cox regression, LASSO regression and Random Forest methods. We further validated SRPI using two independent datasets. Afterwards, we examined the molecular and immune characterized of SRPI. Finally, we conducted multiply drug screening and experimental approaches to identify and confirm the most proper agents for patients with high SRPI. Based on the mRNAsi-related genes, BC patients were classified into two stemness subtypes with distinct prognosis, functional annotations, genomic variations and immune profiles. Using the SRPI, we identified a specific subgroup of BC patients with high SRPI, who had a poor response to immunotherapy, and were less sensitive to commonly used chemotherapeutic agents, FGFR inhibitors, and EGFR inhibitors. We further identified that dasatinib was the most promising therapeutic agent for this subgroup of patients. This study provides further insights into the stemness classification of BC, and demonstrates that SRPI is a promising tool for predicting prognosis and therapeutic opportunities for BC patients.

Developing a prognosis and chemotherapy evaluating model for colon adenocarcinoma based on mitotic catastrophe-related genes

Mitotic catastrophe (MC) is a novel form of cell death that plays an important role in the treatment and drug resistance of colon adenocarcinoma (COAD). However, MC related genes in COAD treatment and prognosis evaluation are rarely studied. In this study, the transcriptome data, somatic mutation and copy number variation data were obtained from The Cancer Genome Atlas (TCGA) database. The mitotic catastrophe related genes (MCRGs) were obtained from GENCARDS website. Differential gene analysis was conducted with LIMMA package. Univariate Cox regression analysis was used to identify prognostic related genes. Mutation analysis was performed and displayed by maftools package. RCircos package was used for localizing the position of genes on chromosomes. “Glmnet” R package was applied for constructing a risk model via the LASSO regression method. Consensus clustering analyses was implemented for clustering different subtypes. Functional enrichment analysis through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) methods, immune infiltration analysis via single sample gene set enrichment analysis (ssGSEA), tumor mutation burden and drug sensitivity analysis by pRRophetic R package were also carried out for risk model or molecular subtype’s assessment. Additionally, the connections between the expression of hub genes and overall survival (OS) were obtained from online Human Protein Atlas (HPA) website. Real-Time Quantitative Polymerase Chain Reaction (RT‑qPCR) further validated the expression of hub genes. A total of 207 differentially expressed MCRGs were selected in the TCGA cohort, 23 of which were significantly associated with OS in COAD patients. Subsequently, we constructed risk score prognostic models with 5 hub MCRGs, including SYCE2, SERPINE1, TRIP6, LIMK1, and EEPD1. The high-risk patients suffered from poorer prognosis. Furthermore, we developed a nomogram that gathered age, sex, staging, and risk score to accurately forecast the clinical survival outcomes in 1, 3, and 5 years. The results of functional enrichment suggested a significant correlation between MCRGs characteristics and cancer progression, with important implications for the immune microenvironment. Moreover, patients who displayed high TMB and high risk score showed worse prognosis, and risk characteristics were associated with different chemotherapeutic agents. Finally, RT‑qPCR verified the increased expression of the five MCRGs in clinical samples. The five MCRGs in the prognostic signature were associated with prognosis, and could be treated as reliable prognostic biomarkers and therapeutic targets for COAD patients with distinct clinicopathological characteristics, thereby providing a foundation for the precise application of pertinent drugs in COAD patients.

A simple nomogram for predicting aspiration associated with dysphagia in hospitalized patients after stroke.

Aspiration is a common complication of poststroke dysphagia (PSD) and is associated with poor prognosis and mortality. There is no uniform criterion for determining aspiration associated with dysphagia. The aim of this study was to identify early predictors of aspiration, leading to the development of a simple nomogram for identifying aspiration risk associated with dysphagia in hospitalized patients after stroke. Demographic information and clinical characteristics of 330 patients with PSD in the training cohort were utilized to develop a nomogram. The LASSO regression method was used to screen variables, and logistic regression was used to construct the nomogram. Internal validation was performed with bootstrap in the training cohort, and external validation was performed in the validation cohort of another 82 patients. The area under the curve (AUC), calibration curves, and decision curve analysis (DCA) were used to evaluate the performance of the nomogram. Seven variables were selected based on LASSO and multivariate logistic regression. The AUC of the nomogram was 0.834 (95% CI, 0.790-0.878) in the training cohort, 0.806 (95% CI, 0.791-0.880) in the internal validation cohort, and 0.882 (95% CI, 0.810-0.954) in the external validation cohort, which indicated that the model had good discrimination. The calibration and DCA curves showed that the nomogram had good accuracy and clinical utility. In this study, we established a nomogram that can be used to identify the risk of aspiration associated with dysphagia after stroke, and patients may benefit from early screening and preventive care.

Glutamine metabolism-related genes and immunotherapy in nonspecific orbital inflammation were validated using bioinformatics and machine learning

BackgroundNonspecific orbital inflammation (NSOI) is an idiopathic, persistent, and proliferative inflammatory condition affecting the orbit, characterized by polymorphous lymphoid infiltration. Its pathogenesis and progression have been linked to imbalances in tumor metabolic pathways, with glutamine (Gln) metabolism emerging as a critical aspect in cancer. Metabolic reprogramming is known to influence clinical outcomes in various malignancies. However, comprehensive research on glutamine metabolism's significance in NSOI is lacking.MethodsThis study conducted a bioinformatics analysis to identify and validate potential glutamine-related molecules (GlnMgs) associated with NSOI. The discovery of GlnMgs involved the intersection of differential expression analysis with a set of 42 candidate GlnMgs. The biological functions and pathways of the identified GlnMgs were analyzed using GSEA and GSVA. Lasso regression and SVM-RFE methods identified hub genes and assessed the diagnostic efficacy of fourteen GlnMgs in NSOI. The correlation between hub GlnMgs and clinical characteristics was also examined. The expression levels of the fourteen GlnMgs were validated using datasets GSE58331 and GSE105149.ResultsFourteen GlnMgs related to NSOI were identified, including FTCD, CPS1, CTPS1, NAGS, DDAH2, PHGDH, GGT1, GCLM, GLUD1, ART4, AADAT, ASNSD1, SLC38A1, and GFPT2. Biological function analysis indicated their involvement in responses to extracellular stimulus, mitochondrial matrix, and lipid transport. The diagnostic performance of these GlnMgs in distinguishing NSOI showed promising results.ConclusionsThis study successfully identified fourteen GlnMgs associated with NSOI, providing insights into potential novel biomarkers for NSOI and avenues for monitoring disease progression.

Exploring the intestinal ecosystem: from gut microbiota to associations with subtypes of inflammatory bowel disease.

Significant differences have been discovered between subtypes of Crohn's disease (CD) and ulcerative colitis (UC). The role of gut microbiota in promoting the onset of UC and CD is established, but conclusions regarding subtype-specific analyses remain limited. This study aims to explore the influence of gut microbiota on subtypes of UC and CD, offering novel insights into the pathogenesis and treatment of UC and CD.Two-sample Mendelian randomization (MR) analysis was employed to examine the causal relationship between subtypes of UC and CD and gut microbiota composition. Gut microbiota data were sourced from the International Consortium MiBioGen, while UC and CD data were obtained from FINNGEN. Eligible single nucleotide polymorphisms (SNPs) were selected as instrumental variables. Multiple analytical approaches such as inverse variance-weighted (IVW), MR-Egger regression, weighted median, weighted mode, and MR-RAPS were utilized. Sensitivity analyses including MR-Egger intercept test, Cochran's Q test, and leave-one-out analysis were conducted for quality control. Subsequently, we employed multivariable IVW, MR-Egger, weighted median, and LASSO regression methods to identify independently significant genera or families and conducted sensitivity analyses. We have determined that Hungatella, Acidaminococcaceae, and 15 other microbial taxa act as protective factors for various CD and UC subtypes, while Terrisporobacter, Anaerostipes, and 23 other microbial taxa are associated with increased risk for different CD and UC subtypes. Furthermore, through multivariable MR analysis, we have identified significant genera or families with independent effects. Our study confirms a causal relationship between dysbiosis of gut microbiota and the occurrence of CD and UC subtypes. Furthermore, it validates etiological distinctions among different subtypes of CD and UC. A novel approach to adjunctive therapy involving distinct UC or CD subtypes may involve the use of probiotics and represents a potential avenue for future treatments.

Security and Risk Control of Enterprise Supply Chain Finance in the Context of Digital Economy

Abstract The widespread application of new-generation information technologies, such as the Internet of Things (IoT), has brought about significant changes in the operating environments of enterprises and the supply chains in which they operate. At the same time, enterprises aim to maximize profits while minimizing enterprise risks. This paper examines the risk assessment of enterprises in supply chain finance based on the relevant literature review. First of all, after combing and defining the relevant applications of digital technology in enterprise risk prevention and control, a Logistic enterprise risk assessment model under supply chain finance is constructed, and the parameters of the model are optimally selected and estimated for testing. The main factors affecting enterprise risk are analyzed from four dimensions: enterprise, core enterprise, financing asset risk and supply chain-related aspects, and a set of index systems suitable for enterprise risk assessment under supply chain finance is established by using correlation analysis, RF and Lasso regression methods. x3, x4, x5, x6, x7, x9 and x15 are negatively correlated with the enterprise risk, and the gross profit margin and the top five customers’ gross profit margin are negatively correlated with the enterprise risk, while sales and profit margin are negatively correlated with the top five customers’ gross profit margin. The correlation between enterprise credit risk and gross margin and the proportion of sales revenue from the top five customers is positive. The accuracy of prediction of the evaluation model is 93.3%. It can be seen that the proposed logistic assessment model is reasonable and effective, and can be practically applied in the security and risk control of enterprise supply chain finance.

Prognostic analysis of patients with gastric cancer based on N6-methyladenosine modification patterns and tumor microenvironment characterization.

Cancers arise from genetic and epigenetic abnormalities that affect oncogenes and tumor suppressor genes, compounded by gene mutations. The N6-methyladenosine (m6A) RNA modification, regulated by methylation regulators, has been implicated in tumor proliferation, differentiation, tumorigenesis, invasion, and metastasis. However, the role of m6A modification patterns in the tumor microenvironment of gastric cancer (GC) remains poorly understood. In this study, we analyzed m6A modification patterns in 267 GC samples utilizing 31 m6A regulators. Using consensus clustering, we identified two unique subgroups of GC. Patients with GC were segregated into high- and low-infiltration cohorts to evaluate the infiltration proportions of the five prognostically significant immune cell types. Leveraging the differential genes in GC, we identified a "green" module via Weighted Gene Co-expression Network Analysis. A risk prediction model was established using the LASSO regression method. The "green" module was connected to both the m6A RNA methylation cluster and immune infiltration patterns. Based on "Module Membership" and "Gene Significance", 37 hub genes were identified, and a risk prediction model incorporating nine hub genes was established. Furthermore, methylated RNA immunoprecipitation and RNA Immunoprecipitation assays revealed that YTHDF1 elevated the expression of DNMT3B, which synergistically promoted the initiation and development of GC. We elucidated the molecular mechanism underlying the regulation of DNMT3B by YTHDF1 and explored the crosstalk between m6A and 5mC modification. m6A RNA methylation regulators are instrumental in malignant progression and the dynamics of tumor microenvironment infiltration of GC. Assessing m6A modification patterns and tumor microenvironment infiltration characteristics in patients with GC holds promise as a valuable prognostic biomarker.

A Prediction Model for Prolonged Hospital Length of Stay (ProLOS) in Coronavirus Disease 2019 (COVID-19) Patients.

Coronavirus disease 2019 (COVID-19) is an acute respiratory infectious disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). With the normalization of COVID-19 globally, it is crucial to construct a prediction model that enables clinicians to identify patients at risk for ProLOS based on demographics and serum inflammatory biomarkers. The study included hospitalized patients with a confirmed diagnosis of COVID-19. These patients were randomly grouped into a training (80%) and a test (20%) cohort. The LASSO regression and ten-fold cross-validation method were applied to filter variables. The training cohort utilized multifactorial logistic regression analyses to identify the independent factors of ProLOS in COVID-19 patients. A 4-variable nomogram was created for clinical use. ROC curves were plotted, and the area under the curve (AUC) was calculated to evaluate the model's discrimination; calibration analysis was planned to assess the validity of the nomogram, and decision curve analysis (DCA) was used to evaluate the clinical usefulness of the model. The results showed that among 310 patients with COVID-19, 80 had extended hospitalization (80/310). Four independent risk factors for COVID-19 patients were identified: age, coexisting chronic respiratory diseases, white blood cell count (WBC), and serum albumin (ALB). A nomogram based on these variables was created. The AUC in the training cohort was 0.808 (95% CI: 0.75 - 0.8671), and the AUC in the test cohort was 0.815 (95% CI: 0.7031 - 0.9282). The model demonstrates good calibration and can be used with threshold probabilities ranging from 0% to 100% to obtain clinical net benefits. A predictive model has been created to accurately predict whether the hospitalization duration of COVID-19 patients will be prolonged. This model incorporates serum WBC, ALB levels, age, and the presence of chronic respiratory system diseases.

Penerapan Metode Regresi Least Absolute Shrinkage and Selection Operator (LASSO) dan Regresi Linier untuk Memprediksi Tingkat Kemiskinan di Indonesia

Poverty is defined as a situation where a person or family is unable to meet the basic needs for survival, such as clothing, food, shelter, and education. This study aims to compare the accuracy of the Least Absolute Shrinkage and Selection Operator (Lasso) regression method and linear regression in predicting poverty levels in Indonesia and to choose the best model from the two methods used. The results of this study indicate that of the two algorithms used, namely the linear regression algorithm and Lasso regression, the algorithm that has a higher level of accuracy for predicting poverty rates in each province in Indonesia is the linear regression algorithm because it has a lower MSE value and has the value of R^2 is closer to 1 than the Lasso regression algorithm. In addition, the results of the analysis show that the variables that have the highest influence on poverty rates in provinces in Indonesia are education, as well as the Human Development Index (IPM).Keywords: Poverty Rate Prediction, Lasso Regression, Linear Regression

Single-cell transcriptomics reveals heterogeneity in esophageal squamous epithelial cells and constructs models for predicting patient prognosis and immunotherapy.

Esophageal squamous cell carcinoma (ESCC), characterized by its high invasiveness and malignant potential, has long been a formidable challenge in terms of treatment. A variety of advanced analytical techniques are employed, including single-cell RNA sequencing (scRNA-seq), cell trajectory inference, transcription factor regulatory network analysis, GSVA enrichment analysis, mutation profile construction, and the inference of potential immunotherapeutic drugs. The purpose is to conduct a more comprehensive exploration of the heterogeneity among malignant squamous epithelial cell subgroups within the ESCC microenvironment and establish a model for predicting the prognosis and immunotherapy outcomes of ESCC patients. An analysis was conducted through scRNA-seq, and three Cluster of malignant epithelial cells were identified using the infer CNV method. Cluster 0 was found to exhibit high invasiveness, whereas Cluster 1 displayed prominent characteristics associated with epithelial-mesenchymal transition. Confirmation of these findings was provided through cell trajectory analysis, which positioned Cluster 0 at the initiation stage of development and Cluster 1 at the final developmental stage. The abundance of Cluster 0-2 groups in TCGA-LUAD samples was assessed using ssGSEA and subsequently categorized into high and low-expression groups. Notably, it was observed that Cluster 0-1 had a significant impact on survival (p<0.05). Furthermore, GSVA enrichment analysis demonstrated heightened activity in hallmark pathways for Cluster 0, whereas Cluster 1 exhibited notable enrichment in pathways related to cell proliferation. It is noteworthy that a prognostic model was established utilizing feature genes from Cluster 0-1, employing the Lasso and stepwise regression methods. The results revealed that in TCGA and GSE53624 cohorts, the low-risk group demonstrated significantly higher overall survival and increased levels of immune infiltration. An examination of four external immunotherapy cohorts unveiled that the low-risk group exhibited improved immunotherapeutic efficacy. Additionally, more meaningful treatment options were identified for the low-risk group. The findings revealed distinct interactions between malignant epithelial cells of ESCC and subgroups within the tumor microenvironment. Two cell clusters, strongly linked to survival, were pinpointed, and a signature was formulated. This signature is expected to play a crucial role in identifying and advancing precision medicine approaches for the treatment of ESCC.

Research on the Implementation Path of Tourism Contributing to the Development of Rural Revitalization Combined with Elastic Regression Network Algorithm

Abstract This paper analyzes the power mechanism of tourism to help rural revitalization by elaborating the core content of rural tourism and rural revitalization. It also combines the advantages of the Lasso regression method and ridge regression method to form a flexible regression network model, which is used to predict the amount of tourists and tourists’ demand for rural tourism so as to put forward the implementation path of the tourism industry to help rural revitalization development. To verify the effectiveness of the proposed algorithm, data analysis is carried out in rural areas of L city as an example. The results show that compared with the Lasso regression model, the average absolute percentage error decreased by 2.41%, the average absolute error decreased by 2,233, and the root-mean-square error decreased by 2,967, which indicates that the elasticity regression network prediction model has a stronger generalization ability and a better prediction ability, and it can predict the number of tourists more accurately, and it can provide data for the tourism industry to contribute to rural revitalization and development. Reference.

Lasso regression method for a compositional covariate regularised by the norm L1 pairwise logratio

Lasso regression methods include a penalty function expressed in terms of a norm defined in the space of model coefficients. The norm plays a key role as regards the way coefficients can become irrelevant in the model. For models with a compositional covariate, the norm should be coherent with the Aitchison geometry. The proposed method is based on a newly-defined compositional norm called L1 pairwise logratio. The novel approach allows one to construct an appropriate basis through a sequential binary partition for discriminating between balances that influence the response variable and those that have no effect. This generalised Lasso regression scheme is illustrated with the analysis of a geochemical data set.

Prediction of acute kidney injury following coronary artery bypass graft surgery in elderly Chinese population

BackgroundAcute kidney injury (AKI) is a common and serious complication following coronary artery bypass graft (CABG) surgery. Advanced age is an independent risk factor for the development of AKI, and the incidence of AKI in the elderly increases more rapidly than that in younger patients. This study aimed to develop and validate the risk prediction model for AKI after CABG in elderly patients.MethodsPatients were retrospectively recruited from January 2019 to December 2020. AKI after CABG was defined according to the criteria of Kidney Disease Improving Global Outcomes (KDIGO). The entire population was divided into the derivation set and the verification set using random split sampling (ratio: 7:3). Lasso regression method was applied to screen for the variables in the derivation set. Decision curve analysis (DCA) and receiver operating characteristic (ROC) curves were plotted to analyze the predictive ability of the model for AKI risk in the derivation set and the verification set.ResultsA total of 2155 patients were enrolled in this study. They were randomly divided into the derivation set (1509 cases) and the validation set (646 cases). Risk factors associated with AKI were selected by Lasso regression including T2DM, diabetes mellitus type intraoperative use of intra-aortic ballon pump (IABP), cardiopulmonary bypass (CPB), epinephrine, isoprenaline, and so on. The model was established by Lasso logistic regression. The area under the ROC curve (AUC) of the model for the derivation set was 0.754 (95% CI: 0.720 − 0.789), and that for the validation cohort was 0.718 (95% CI: 0.665 − 0.771).ConclusionIn this study, the model with significant preoperative and intraoperative variables showed good prediction performance for AKI following CABG in elderly patients to optimize postoperative treatment strategies and improve early prognosis.

A Predictive Study on the Content of Epigallocatechin Gallate (EGCG) in Yunnan Large Leaf Tea Trees Based on the Nomogram Model

To explore the changes in epigallocatechin gallate (EGCG) content in tea under abiotic stress conditions, we collected tea samples, along with corresponding soil and altitude data, and utilized the measured data for single-factor analysis. At the same time, the LASSO regression method, which is rarely used in agriculture, was employed to screen modeling factors, a prediction model was established, and the Akaike information criterion (AIC) was introduced to compare the goodness of fit. The results show that LASSO screening reduced the AIC value of the model by 13.8%. The average area under the curve of the training set and the validation set was 0.81 and 0.76, respectively, and the calibration curve also showed good consistency. Based on the nomogram model, a visual prediction system was developed, and the content prediction curve was introduced for detailed soil evaluation. The accuracy rate reached 75% after external verification. This study provides a theoretical basis for elucidating the prediction and intervention of Pu’er tea quality under abiotic stress conditions.

Enhancing Feature Selection Optimization for COVID-19 Microarray Data

The utilization of gene selection techniques is crucial when dealing with extensive datasets containing limited cases and numerous genes, as they enhance the learning processes and improve overall outcomes. In this research, we introduce a hybrid method that combines the binary reptile search algorithm (BRSA) with the LASSO regression method to effectively filter and reduce the dimensionality of a gene expression dataset. Our primary objective was to pinpoint genes associated with COVID-19 by examining the GSE149273 dataset, which focuses on respiratory viral (RV) infections in individuals with asthma. This dataset suggested a potential increase in ACE2 expression, a critical receptor for the SARS-CoV-2 virus, along with the activation of cytokine pathways linked to COVID-19. Our proposed BRSA method successfully identified six significant genes, including ACE2, IFIT5, and TRIM14, that are closely related to COVID-19, achieving an impressive maximum classification accuracy of 87.22%. By conducting a comparative analysis against four existing binary feature selection algorithms, we demonstrated the effectiveness of our hybrid approach in reducing the dimensionality of features, while maintaining a high classification accuracy. As a result, our hybrid approach shows great promise for identifying COVID-19-related genes and could be an invaluable tool for other studies dealing with very large gene expression datasets.

Optimizing Linear Regression Models with Lasso and Ridge Regression: A Study on UAE Financial Behavior during COVID-19

This study examines the way individuals in the United Arab Emirates handled their finances during the COVID-19 pandemic, with a focus on optimizing linear regression models utilizing Lasso and Ridge Regression methods. Using the World Bank's enormous Global Findex dataset, we navigated the obstacles of multicollinearity and missing values. Our investigation showed that specific economic indicators and demographic characteristics, including gender and educational attainment, are crucial in affecting financial choices during crises. Our study carefully analyzed the performance of Lasso and Ridge Regression in the context of the Global Findex dataset. Ridge Regression was outperformed by Lasso, which is renowned for its feature selection and complexity reduction abilities. This research provides a comprehensive knowledge of the way advanced regression approaches might improve predictive modeling in such settings, shedding light on the complex dynamics of financial activity during a worldwide crisis.