Principal Component Analysis Research Articles

Stock structure delineation of Kawakawa Euthynnus affinis (Cantor, 1849) from Malaysian Borneo using multivariate morphometric analysis

Euthynnus affinis, also known as Kawakawa, is a species of fish that is very significant to the fishing industry, particularly in the Indo-Pacific area. Regrettably, its stock structure and regional species’ management are poorly understood. This research aimed to employ a morphometric approach and determine if different stocks exist among 188 specimens of E. affinis obtained from three different areas: the Celebes Sea, the South China Sea, and the Sulu Sea, all surrounding Malaysian Borneo. Multivariate studies, including PCA (Principal Component Analysis) and DFA (Discriminant Function Analysis), were utilized to differentiate eight E. affinis populations based on 12 morphometric parameters. The findings divulged that E. affinis populations of Malaysian Borneo (Sabah and Sarawak) were relatively homogenous, as indicated by the DFA and the principal component’s scatterplot. Moreover, hierarchical cluster analysis focused on the Mahalanobis distances demonstrated that eight populations of E. affinis were grouped into three separate groups. The first group comprises Kota Kinabalu (Sabah), Mukah, Bintawa, and Pulau Bruit (Sarawak) populations. Euthynnus affinis from Lahad Datu and Kudat (Sabah) were clustered in the second group, while the third group consisted of Tawau and Sandakan (Sabah) populations. This study’s findings will give important preliminary data on E. affinis for effectively managing and protecting Malaysian Borneo’s regional waterways.

Development and validation of a pictographic assessment embodiment scale.

Embodiment is a complex concept related to the subjective perception of an object as it belongs to its own body. In general, this construct has been evaluated by means of questionnaires, but validation studies in other cultures and limitations related with barriers of language received little attention. The purpose of the present investigation was twofold: to validate the factorial structure of embodiment questionnaire (EQ) and to construct a pictographic scale (PAE) to measure embodiment without relapse verbal representations. In the first experiment, 136 participants underwent a Rubber Hand Illusion (RHI) procedure following both congruent and incongruent (control) visuo-tactile stimulations. Then, they evaluated embodiment illusion in EQ using a Likert-type scale to rate their agreement or disagreement with 27 statements and with a pictographic scale designed to assess their subjective experience of the illusion. Principal components analysis in EQ scores identified four components that emerged in both conditions: Embodiment, Disembodiment, Affect and Deafference. PAE scale was highly correlated with embodiment factor and can differentiate between conditions. In a second experiment, 30 participants underwent the RHI procedure, and they were assessed using PAE and proprioceptive drift. Results indicate a high positive correlation between PAE and post-illusion drift score. These results provide evidence about the consistency of the factorial structure of EQ across cultures, and we also provide a new pictographic tool that allows quick measurement of embodiment overcoming language barriers.

Determination of geometric properties of cocoa beans (Theobroma cacao L.)

Objective: To determine the linear physical dimensions of dried Mexican cocoa beans to estimate their geometric properties and differentiate them through a principal component analysis. Design/Methodology/Approach: For the research, 51 dry samples of cocoa beans (Theobroma cacao L.) were collected from three producer states in Mexico. The physical properties of cocoa beans were determined based on their linear dimensions: length, width, thickness, geometric diameter, sphericity, volume, and shape factor. Moisture, average weight, and ether extract were also determined. Results: The results revealed the relation between linear and geometric properties, particularly the fact that bean weight is significant (p<0.001) regarding all the properties evaluated, except moisture. Said relation explained 98.8% of the total variation in the first two components observed in the cocoa samples from the three states (Tabasco, Chiapas, and Oaxaca). Average bean weight, sphericity, and volume contributed the most to the total variation. Findings/Conclusions: The only quantitative variable that showed significance was bean weight. The other measurements—length, width, and thickness—did not. However, there was significance when coupling the measurements in the expressions of surface area and volume.

Detection of Signal Integrity Issues in Vibration Monitoring Using One-Class Support Vector Machine

AbstractThis paper presents an analysis of the common signal integrity issues in vibration monitoring caused by sensor saturation and signal distortion, or sensor loosening and detachment, and the development of a method of detecting the occurrence of vibration signal integrity issues using a one-class support vector machine. For this, vibration signals with distortions due to sensor saturation and/or sensor detachment are analysed to determine parameters sensitive to common integrity issues. These features are then extracted from training data of good quality vibration signals. Principal Component Analysis is used to dimensionally reduce the parameters which are then used in the training of the one-class support vector machine. The proposed method was validated on trained and untrained health conditions. Additionally, model sensitivity was also tested on trained health conditions with varying defect sizes. The method successfully detected all signal integrity issues tested proving to be an effective pre-processing step in condition monitoring systems for increased reliability and accuracy.

University Information System’s Impact on Academic Performance: A Comprehensive Logistic Regression Analysis with Principal Component Analysis and Performance Metrics

University Information System’s Impact on Academic Performance: A Comprehensive Logistic Regression Analysis with Principal Component Analysis and Performance Metrics

Does institutional quality complement the relationship between ownership structure and corporate social responsibility?

The key purpose of the Study is to examine if institutional quality complements the relationship between Ownership Structure and Corporate Social Responsibility disclosure and performance in the light of legitimacy and agency theory. To the best of my knowledge, it is the first study in literature of finance. The sample comprises of 112 top-performing listed firms (based on market capitalization) at Pakistan Stock Exchange from 2010 to 2019. Institutional quality comprised of world governance indicators which is developed via principal component analysis, an instrumental variable approach and content analysis are used for CSR Disclosure Index to demonstrate the relationship between ownership structure and CSR. The resources complementary phenomenon is adopted to examine the institutional quality's role. Our results show significantly positive impact of Institutional and Foreign Ownerships on CSR while negative significant influence of CEO Duality and Family Ownership on CSR, suggesting that well governed firms will be more socially responsible. In addition, the findings suggest the institutional quality's positive moderating role on the relationship between ownership structure and CSR, signifying the institutional quality's complementary role for the weak corporate environment in Pakistan. Our findings are robust to a series of tests by using Generalized Method of Moment (GMM).

Multi-omics comprehensive analysis reveals the predictive value of N6-methyladenosine- related genes in prognosis and immune escape of bladder cancer.

N6-methyladenosine (m6A) is the most frequent RNA modification in mammals, and its role in bladder cancer (BC) remains rarely revealed. To predict the value of m6A-related genes in prognosis and immunity in BC. We performed multiple omics analysis of 618 TCGA and GEO patients and used principal component analysis (PCA) to calculate the m6A score for BC patients. We described the multiple omics status of 23 m6A methylation-related genes (MRGs), and four m6A clusters were identified, which showed significant differences in immune infiltration and biological pathways. Next, we intersected the differential genes among m6A clusters, and 11 survival-related genes were identified, which were used to calculate the m6A score for the patients. We found that the high-score (HS) group showed lower tumor mutation burden (TMB) and TP53 mutations and better prognosis than the low-score (LS) group. Lower immune infiltration, higher expression of PD-L1, PD-1, and CTLA4, and higher immune dysfunction and immune exclusion scores were identified in the LS group, suggesting a higher possibility of immune escape. Finally, the experimental verification shows that the m6A related genes, such as IGFBP1, plays an important role in the growth and metastasis of bladder cancer. These findings revealed the important roles of m6A MRGs in predicting prognosis, TMB status, TP53 mutation, immune functions and immunotherapeutic response in BC.

Systematic Implications of Palynomorph Diversity Using Microscopic Trends Among Asteraceous Flora From the Drylands of Baluchistan, Pakistan.

Pollen micromorphological traits with taxonomic implications are first reported from the study area for 50 Asteraceous taxa belonging to nine tribes. Cichorieae (21 taxa), Cardueae (11 taxa), Inuleae (six taxa), and Anthemideae (four taxa) are the leading tribes. The research included Cousinia haeckeliae, Himalaiella afghana, Pterachaenia stewartii (endemic to Afghanistan and Pakistan), and Xylanthemum macropodum (endemic to Baluchistan). Light and scanning electron microscopy were employed for the visualization of pollen photomicrographs. The data was analyzed statistically via SPSS, PAST, and Origin. Significant diagnostic qualitative and quantitative palynological traits were explored for discrimination down to the species level within the tribes. All the investigated taxa possessed radial symmetry, isopolarity, and monad form (characters for distinction at the subdivision level). The aperture types were trizonocolporate, tetrazonocolporate, and tricolporate with number position and character (NPC) formulas N3P4C5, N4P4C5, and N3P4C3. Goniotreme, peritreme, and ptychotreme types of amb were recognized. Echinate, echinate lophate, scabrate, and gemmate sculpturing were present with and without perforated surface patterns. Variations in the shapes in polar and equatorial views and lacuna shapes further assisted the separation of taxa. The observed shape classes were perprolate, prolate spheroidal, prolate, subprolate, oblate spheroidal, suboblate, and oblate. Principal component analysis, correlation, standard probability plots, and ridge line paired features plot for quantitative variables determined the positive correlation between the length and width of colpi in equatorial and polar view with polar axis and equatorial diameter and number of spines between colpi with the number of spines per pollen. The number of spines per pollen was negatively correlated with the width and length of colpi in the polar view. Multiple sample analysis of variance (ANOVA) concluded that a high statistically significant difference exists among the means of analyzed traits. The examined qualitative and quantitative palynological traits revealed noticeable variations, thus providing the source for species discrimination in Asteraceous tribes.

Proteo-genomic characterization of cirrhosis-associated liver cancers reveals potential subtypes and therapeutic targets.

This study aimed to identify potential subtypes of hepatocellular carcinoma (HCC) associated with cirrhosis and to investigate key markers using bioinformatic analysis of gene expression datasets-0. Three data sets (GSE17548, GSE56140, and GSE87630) were extracted from the Gene Expression Omnibus (GEO) database and normalized using the Limma package in R. Principal component analysis (PCA) and cluster analysis was performed to examine data distribution and identify subtypes. Differential gene expression analysis was performed using the Limma software package. Protein-protein interaction analysis and functional annotation were performed using the STRING database and Cytoscape software. Important signaling pathways and processes were identified using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway Analysis. The analysis revealed different subtypes of HCC associated with cirrhosis and identified several key genes, including CCNB2, MCM4, and CDC20, with strong binding power and prognostic value. Functional annotation indicated involvement in cell cycle regulation and metabolic pathways. ROC analysis showed high sensitivity and specificity of these genes in predicting HCC prognosis. These results suggest that CCNB2, MCM4, and CDC20 may serve as potential biomarkers for predicting HCC prognosis in patients with cirrhosis and provide insights into the molecular mechanisms of HCC progression.

Untargeted metabolomics of blood plasma samples of patients with hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths in the world. HCC is often diagnosed late because patients with early-stage cancer have no apparent symptoms. Therefore, it is desirable to find a reliable method for an early diagnosis based on the detection of metabolites – biomarkers, that can be detected in the early stages of the disease. Untargeted metabolomics is often used as a tool to find a suitable biomarker for several diseases. In this work, untargeted metabolomics was performed on blood plasma samples of HCC patients and compared with healthy individuals and patients with liver cirrhosis. A combination of liquid chromatography and high-resolution mass spectrometry was used as an analytical method. More than a thousand peaks were detected in the blood plasma samples, from which mainly amino acids, carboxylic acids, lipids, and their derivatives were evaluated as potential biomarkers. The data obtained were statistically processed using the analysis of variance, correlation analysis, and principal component analysis.

Genetic substructure and host-specific natural selection trend across vaccine-candidate ORF-2 capsid protein of hepatitis-E virus.

Hepatitis E virus is a primary cause of acute hepatitis worldwide. The present study attempts to assess the genetic variability and evolutionary divergence among HEV genotypes. A vaccine promising capsid-protein coding ORF-2 gene sequences of HEV was evaluated using phylogenetics, model-based population genetic methods and principal component analysis. The analyses unveiled nine distinct clusters as subpopulations for six HEV genotypes. HEV-3 genotype samples stratified into four different subgroups, while HEV-4 stratified into three additional subclusters. Rabbit-infectious HEV-3ra samples constitute a distinct cluster. Pairwise analysis identified marked genetic distinction of HEV-4c and HEV-4i subgenotypes compared to other genotypes. Numerous admixed, inter and intragenotype recombinant strains were detected. The MEME method identified several ORF-2 codon sites under positive selection. Some selection signatures lead to amino acid substitutions within ORF-2, resulting in altered physicochemical features. Moreover, a pattern of host-specific adaptive signatures was identified among HEV genotypes. The analyses conclusively depict that recombination and episodic positive selection events have shaped the observed genetic diversity among different HEV genotypes. The significant genetic diversity and stratification of HEV-3 and HEV-4 genotypes into subgroups, as identified in the current study, are noteworthy and may have implications for the efficacy of anti-HEV vaccines.

Gut microbiota composition links to variation in functional domains across psychiatric disorders

ObjectiveChanges in microbial composition are observed in various psychiatric disorders, but their specificity to certain symptoms or processes remains unclear. This study explores the associations between the gut microbiota composition and the Research Domain Criteria (RDoC) domains of functioning, representing symptom domains, specifically focusing on stress-related and neurodevelopmental disorders in patients with and without psychiatric comorbidity. MethodsThe gut microbiota was analyzed in 369 participants, comprising 272 individuals diagnosed with a mood disorder, anxiety disorder, attention deficit/hyperactivity disorder, autism spectrum disorder, and/or substance use disorder, as well as 97 psychiatrically unaffected individuals. The RDoC domains were estimated using principal component analysis (PCA) with oblique rotation on a range of psychiatric, psychological, and personality measures. Associations between the gut microbiota and the functional domains were assessed using multiple linear regression and permanova, adjusted for age, sex, diet, smoking, medication use and comorbidity status. ResultsFour functional domains, aligning with RDoC’s negative valence, social processes, cognitive systems, and arousal/regulatory systems domains, were identified. Significant associations were found between these domains and eight microbial genera, including associations of negative valence with the abundance of the genera Sellimonas, CHKCI001, Clostridium sensu stricto 1, Oscillibacter, and Flavonifractor; social processes with Sellimonas; cognitive systems with Sporobacter and Hungatella; and arousal/regulatory systems with Ruminococcus torques (all pFDR < 0.05). ConclusionOur findings demonstrate associations between the gut microbiota and the domains of functioning across patients and unaffected individuals, potentially mediated by immune-related processes. These results open avenues for microbiota-focused personalized interventions, considering psychiatric comorbidity. However, further research is warranted to establish causality and elucidate mechanistic pathways.

Machine Learning Techniques and Hybrid Feature Selection Methods for Efficient Prediction of Cancer

The high-dimensional genomic data presents significant challenges, and traditional analytical methods often struggle to capture the complex, non-linear relationships within these datasets. This study elaborates into the application of machine learning methods for dimensionality reduction and predictive modeling of binary phenotypes using gene expression data. Various dimensionality reduction techniques are explored, including t-distributed stochastic neighbor embedding (t-SNE), Non-negative matrix factorization (NMF), Principal component analysis (PCA), and manifold learning methods. Additionally, various algorithms such as logistic regression, random forests, support vector machines (SVMs)cand naive Bayes models are evaluated for predicting phenotypes. The study employs rigorous cross-validation, permutation testing, and evaluation metrics like the Matthews Correlation Coefficient (MCC) to assess model performance. The study rigorously assesses current genomics strategies, pinpointing their drawbacks and suggesting areas for future investigation, while delving into the potential of machine learning to overcome these hurdles and offer valuable insights in genomics. Keywords- Gene Expression , Dimensionality Reduction, Biomarkers, Cancer Prediction, Epigenetics, Machine Learning, Feature extraction.

Phase division and recognition of crystal HRTEM images based on machine learning and deep learning

The High Resolution Transmission Electron Microscope (HRTEM) images provide valuable insights into the atomic microstructure, dislocation patterns, defects, and phase characteristics of materials. However, the current analysis and research of HRTEM images of crystal materials heavily rely on manual expertise, which is labor-intensive and susceptible to subjective errors. This study proposes a combined machine learning and deep learning approach to automatically partition the same phase regions in crystal HRTEM images. The entire image is traversed by a sliding window to compute the amplitude spectrum of the Fast Fourier Transform (FFT) in each window. The generated data is transformed into a 4-dimensional (4D) format. Principal component analysis (PCA) on this 4D data estimates the number of feature regions. Non-negative matrix factorization (NMF) then decomposes the data into a coefficient matrix representing feature region distribution, and a feature matrix corresponding to the FFT magnitude spectra. Phase recognition based on deep learning enables identifying the phase of each feature region, thereby achieving automatic segmentation and recognition of phase regions in HRTEM images of crystals. Experiments on zirconium and oxide nanoparticle HRTEM images demonstrate the proposed method achieve the consistency of manual analysis. Code and supplementary material are available at https://github.com/rememberBr/HRTEM2.

Quali-quantitative water behaviour in an intensive swine production catchment in the Atlantic Forest biome, Southern Brazil.

Agriculture is an essential economic activity in Brazil. However, it is also the main source of water quality degradation. Monitoring catchments with agricultural land use is a way to generate information on a scale to identify causes and sources of water quality degradation. This work used monitoring data derived from hydrology and the quality of surface and underground water in an intensive agricultural catchment in the Atlantic Forest biome. The Fortaleza River catchment is located in the western part of Santa Catarina state in southern Brazil and has 62 km2 of drainage area. Hydrological and water quality monitoring was conducted for 7years at two fluviometric stations, three lysimeters, one meteorological station, and one piezometer. Data on precipitation, temperature, water flow, surface runoff, drainage, and water quality were used. Statistical analyses were also developed. Precipitation between 2013 and 2019 presented a homogeneous distribution in monthly and annual data, with January and July the months with the highest and lowest values, respectively. Statistical difference in the average and Q95 flows was found in upstream and downstream fluviometric sections. In terms of quality, statistical differences were identified for ammonium, nitrate, and potassium concentrations, which had higher concentrations in lysimeter runoff, indicating direct influence of agricultural activity on water quality. Principal component analysis (PCA) indicated that (i) surface water presented a positive relationship in Component 1 for the magnesium-calcium, sulphate-chloride, and acetate-bromide groups and a negative relationship for phosphate-nitrate; (ii) in lysimeters, the positive relationship occurred for Component 2 for the phosphate-chloride and sulphate-nitrate groups and was negative for ammonium-lithium and calcium-potassium-magnesium; and (iii) in piezometer, positive relationships were found for chloride-sodium and phosphate-nitrite pairs, while negative relationships were found for calcium-magnesium.

One year of active moss biomonitoring in the identification of PAHs in an urbanized area—prospects and implications

Classical monitoring of air pollution provides information on environmental quality but involves high costs. An alternative to this method is the use of bioindicators. The purpose of our work was to evaluate atmospheric aerosol pollution by selected polycyclic aromatic hydrocarbons conducted as part of annual active biomonitoring (“moss-bag” technique) with the use of three moss species: Pleurozium schreberi, Sphagnum fallax, and Dicranum polysetum. The gas chromatography-mass spectrometry (GC–MS) was utilized to determine certain 13 polycyclic aromatic hydrocarbons (PAHs). Three seasonal variations in PAH concentrations have been observed as a result of the study. A fire on the toilet paper plant caused an increase of five new compounds: benzo(k)fluoranthene (BkF), benzo(a)pyrene (BaP), indeno(1.2.3)-cd_pyrene (IP), dibenzo(a.h)anthracene (Dah), and benzo(g.h.i)perylene (Bghi) in proximity after 8 months of exposure compared to previous months. The effect of meteorological conditions on the deposition of PAHs (mainly wind direction) in mosses was confirmed by principal component analysis (PCA). Dicranum polysetum moss accumulated on average 26.5% more PAHs than the other species, which allows considering its broader use in active biomonitoring. The “moss-bag” technique demonstrates its feasibility in assessing the source of PAH air pollution in a long-term study. It is recommended to use this biological method as a valuable tool in air quality monitoring.

Comparisons of Different Representative Species Selection Schemes for Reduced-Order Modeling and Chemistry Acceleration of Complex Hydrocarbon Fuels

The simulation of engine combustion processes, such as autoignition, an important process in the co-optimization of fuel-engine design, can be computationally expensive due to the large number of thermo-chemical scalars needed to describe the full chemical system. Yet, the inherent correlations between the different chemical species during oxidation can significantly reduce the complexity of representing this system. One strategy is to select a subset of representative species that accurately captures the combustion process at a fraction of the computational cost of the full system. In this study, we compare the performance of four different techniques to select these species. They include the two-step principal component analysis (PCA) approach, directed relation graphs (DRGs), the global pathway selection (GPS) approach, and the manifold-informed species selection method. A parametric study of the representative species selection is carried out on data from the simulation of homogeneous and perfectly stirred reactors by investigating seven cumulative variances and 47 different cut-off percentages for the two-step PCA, and 65 and 51 thresholds for the DRGs and GPS, respectively. Results show that these selection methods capture key important species that can accurately describe the chemical system and track each stage of oxidation. The two-step PCA is sensitive to the cumulative variance, and DRGs and GPS are sensitive to the choice of target variables. By selecting key representative species and reducing the number of thermo-chemical scalars, these three methods can be used to develop computationally efficient hybrid chemistry schemes.

Exploring Drivers of Wildfires in Spain

Wildfires play a dual role in ecosystems by providing ecological benefits while posing catastrophic events; they also inflict non-catastrophic damage and yield long-term effects on biodiversity, soil quality, and air quality, among other factors, including public health. This study analysed the key determinants of wildland fires in Spain using openly available spatial data from 2008 to 2021, including fire perimeters, bioclimatic variables, topography, and socioeconomic datasets, at a resolution of 1 km2. Our methodology combined principal component analysis (PCA), linear regression analysis, and one-way analysis of variance (ANOVA). Our findings show that scrub/herbaceous vegetation (average 63 ± 1.45% SE) and forests (average 19 ± 0.76% SE) have been highly susceptible to wildfires. The population density exhibited a robust positive correlation with wildfire frequency (R2 = 0.88, p &lt; 0.0001). Although the study provides insights into some fire-related climatic drivers over Spain, it includes only temperature and precipitation-based variables and does not explicitly consider fuel dynamics. Therefore, a more advanced methodology should be applied in the future to understand the local specifics of regional wildfire dynamics. Our study identified that scrub/herbaceous areas and forests near densely populated regions should be prioritised for wildfire management in Spain, particularly under changing climate conditions.

PCA-ICA-LSTM: A Hybrid Deep Learning Model Based on Dimension Reduction Methods to Predict S&amp;P 500 Index Price

AbstractIn this paper, we propose a new hybrid model based on a deep learning network to predict the prices of financial assets. The study addresses two key limitations in existing research: (1) the lack of standardized datasets, time scales, and evaluation metrics, and (2) the focus on prediction return. The proposed model employs a two-stage preprocessing approach utilizing Principal Component Analysis (PCA) for dimensionality reduction and de-noising, followed by Independent Component Analysis (ICA) for feature extraction. A Long Short-Term Memory (LSTM) network with five layers is fed with this preprocessed data to predict the price of the next day using a 5 day time horizon. To ensure comparability with existing literature, experiments employ an 18 year dataset of the Standard &amp; Poor's 500 (S&amp;P500) index and include over 40 technical indicators. Performance evaluation encompasses six metrics, highlighting the model's superiority in accuracy and return rates. Comparative analyses demonstrate the superiority of the proposed PCA-ICA-LSTM model over single-stage statistical methods and other deep learning architectures, achieving notable improvements in evaluation metrics. Evaluation against previous studies using similar datasets corroborates the model's superior performance. Moreover, extensions to the study include adjustments to dataset parameters to account for the COVID-19 pandemic, resulting in improved return rates surpassing traditional trading strategies. PCA-ICA-LSTM achieves a 220% higher return compared to the “hold and wait” strategy in the extended S&amp;P500 dataset, along with a 260% higher return than its closest competitor in the comparison. Furthermore, it outperformed other models in additional case studies. Graphical Abstract

Comprehensive assessment of salt-alkali tolerance of different soybean varieties at the maturation stage

Soil salinization is a significant global challenge impacting food security and arable land safety. Biological amelioration for the development and utilization of salt-alkali land has gained attention in recent years but the breeding and selection of salt-alkali-tolerant crop varieties remain a challenge. Cultivated soybeans are moderately salt-tolerant crops; however, a precise threshold for salt-alkali conditions has not been determined and varies by genotype. Therefore, this study evaluated the salt-alkali tolerance of 15 different soybean varieties using correlation, principal component, membership function, and cluster analyses. The average soybean indices decreased with increasing salt-alkali stress concentrations. Under severe salt-alkali stress, the coefficients of variation for grain yield and biomass were significantly higher than those under mild salt-alkali stress. Plant height and node number were significantly positively correlated, whereas bottom pod height was significantly negatively correlated with branch number, effective pod number, and grain number per plant. Node and grain numbers per plant were significantly positively correlated. Branch number, effective pod number, grain number per plant, and hundred-grain weight were significantly and positively correlated. Effective pod number was significantly positively correlated with grain number per plant, biomass, and yield. Grain number per plant exhibited a significant negative correlation with hundred-grain weight. Biomass and yield were significantly positively correlated. The salt-alkali tolerance of 15 soybean varieties ranged from strong weak as follows: Cangdou 1438, 0734, 1426, 1418, 1453, 1327, 1846, 1412, 1817, 1301, 1821, 1819, 1815, 1814, and 1850. Salt-alkali-sensitive and salt-alkali-tolerant varieties were successfully identified.