Canonical Correlation Research Articles

Rational selection of morphological phenotypic traits to extract essential similarities in chemical perturbation in the ergosterol pathway

Terbinafine, fluconazole, and amorolfine inhibit fungal ergosterol synthesis by acting on their target enzymes at different steps in the synthetic pathway, causing the accumulation of various intermediates. We found that the effects of these three in- hibitors on yeast morphology were different. The number of morphological parameters commonly altered by these drugs was only approximately 6% of the total. Using a rational strategy to find commonly changed parameters,we focused on hidden essential similarities in the phenotypes possibly due to decreased ergosterol levels. This resulted in higher apparent morphological similarity. Improvements in morphological similarity were observed even when canonical correlation analysis was used to select biologically meaningful morphological parameters related to gene function. In addition to changes in cell morphology, we also observed differences in the synergistic effects among the three inhibitors and in their fungicidal effects against pathogenic fungi possibly due to the accumulation of different intermediates. This study provided a comprehensive understanding of the properties of inhibitors acting in the same biosynthetic pathway.

Surface quality evaluation through new optical imaging system-based objective speckle for additive manufactured titanium samples

Laser cladding is an effective additive manufacturing technology used for materials’ surface modification to enhance their surface and mechanical properties. Surface roughness is a crucial feature that affects the materials’ quality and lifespan. Additionally, it is a real indicator of hardness and wear resistance. Although the existing methods may precisely measure surface roughness, they require delicate adjustment, can damage surfaces, and have limited working distances. This study presents a novel optical imaging system to quantitatively estimate the quality modifications of additive manufacturing samples by measuring their surface roughness based on objective speckle and advanced multivariate analysis methods. The raw speckle patterns are generated from deposited layers on Ti6Al4V titanium alloys obtained at different laser processing parameters. The proposed analysis approach produces collections of local statistical matrices from which histogram features are extracted. Canonical correlation analysis (CCA) is then proposed to distinguish the most significant features. The correlation between the features (with and without applying CCA) and surface roughness is established by using two machine learning regression algorithms, nonlinear support vector regression (SVR), random forest regression (RF), and k-nearest neighbor regression (kNN). The results confirmed that combining RF and CCA provided a feasible regression model to estimate the surface roughness, with 0.998 for R2 in training samples and maximum values of 0.258, 0.484, and 4.593 %, respectively, for MAE, RMSE, and MAPE in test samples. This demonstrates that the proposed objective speckle imaging system is effective in estimating the quality modification of additive manufacturing samples by measuring their surface roughness.

Augmentation-Free Graph Contrastive Learning of Invariant-Discriminative Representations.

Graph contrastive learning (GCL) is a promising direction toward alleviating the label dependence, poor generalization and weak robustness of graph neural networks, learning representations with invariance, and discriminability by solving pretasks. The pretasks are mainly built on mutual information estimation, which requires data augmentation to construct positive samples with similar semantics to learn invariant signals and negative samples with dissimilar semantics to empower representation discriminability. However, an appropriate data augmentation configuration depends heavily on lots of empirical trials such as choosing the compositions of data augmentation techniques and the corresponding hyperparameter settings. We propose an augmentation-free GCL method, invariant-discriminative GCL (iGCL), that does not intrinsically require negative samples. iGCL designs the invariant-discriminative loss (ID loss) to learn invariant and discriminative representations. On the one hand, ID loss learns invariant signals by directly minimizing the mean square error (MSE) between the target samples and positive samples in the representation space. On the other hand, ID loss ensures that the representations are discriminative by an orthonormal constraint forcing the different dimensions of representations to be independent of each other. This prevents representations from collapsing to a point or subspace. Our theoretical analysis explains the effectiveness of ID loss from the perspectives of the redundancy reduction criterion, canonical correlation analysis (CCA), and information bottleneck (IB) principle. The experimental results demonstrate that iGCL outperforms all baselines on five node classification benchmark datasets. iGCL also shows superior performance for different label ratios and is capable of resisting graph attacks, which indicates that iGCL has excellent generalization and robustness. The source code is available at https://github.com/lehaifeng/ T-GCN/tree/master/iGCL.

Water quality management of the Aras River in Iran using two modeling methods: canonical correlation analysis and principal component analysis

Water quality management of the Aras River in Iran using two modeling methods: canonical correlation analysis and principal component analysis

Personal Growth Initiative: Relation to Coping Styles, Strategies, and Self-Efficacy

Personal growth initiative (PGI) refers to active and intentional participation in the growth process. PGI includes behavioral and cognitive skills and attitudes that are captured by four factors: Readiness for Change, Planfulness, Using Resources, and Intentional Behavior. There is substantial evidence supporting the positive relations between PGI and various domains of well-being. However, a lack of nuance regarding how the four facets of PGI differentially relate to other aspects of optimal functioning, such as coping, persists. Additionally, PGI has been theoretically tied to coping, but there is limited empirical evidence substantiating this link. Thus, the current study examined the relations between PGI and coping styles, strategies, and self-efficacy in a sample of 789 college students through a series of three canonical correlations. The findings indicated different combinations of the four aspects of PGI related significantly to 13 coping styles, three coping strategies, and three forms of coping self-efficacy. These findings have implications for both the theory and operationalization of PGI, such as the viability of the four separate aspects of PGI, as well as for the application of PGI and coping in college settings, including the development of trainings to increase PGI and adaptive aspects of coping.

Automated Measurement of Cerebral Hemorrhagic Contusions and Outcomes After Traumatic Brain Injury in the TRACK-TBI Study

Because withdrawal of life-sustaining therapy based on perceived poor prognosis is the most common cause of death after moderate or severe traumatic brain injury (TBI), the accuracy of clinical prognoses is directly associated with mortality. Although the location of brain injury is known to be important for determining recovery potential after TBI, the best available prognostic models, such as the International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT) score, do not currently incorporate brain injury location. To test whether automated measurement of cerebral hemorrhagic contusion size and location is associated with improved prognostic performance of the IMPACT score. This prognostic cohort study was performed in 18 US level 1 trauma centers between February 26, 2014, and August 8, 2018. Adult participants aged 17 years or older from the US-based Transforming Research and Clinical Knowledge in TBI (TRACK-TBI) study with moderate or severe TBI (Glasgow Coma Scale score 3-12) and contusions detected on brain computed tomography (CT) scans were included. The data analysis was performed between January 2023 and February 2024. Labeled contusions detected on CT scans using Brain Lesion Analysis and Segmentation Tool for Computed Tomography (BLAST-CT), a validated artificial intelligence algorithm. The primary outcome was a Glasgow Outcome Scale-Extended (GOSE) score of 4 or less at 6 months after injury. Whether frontal or temporal lobe contusion volumes improved the performance of the IMPACT score was tested using logistic regression and area under the receiver operating characteristic curve comparisons. Sparse canonical correlation analysis was used to generate a disability heat map to visualize the strongest brainwide associations with outcomes. The cohort included 291 patients with moderate or severe TBI and contusions (mean [SD] age, 42 [18] years; 221 [76%] male; median [IQR] emergency department arrival Glasgow Coma Scale score, 5 [3-10]). Only temporal contusion volumes improved the discrimination of the IMPACT score (area under the receiver operating characteristic curve, 0.86 vs 0.84; P = .03). The data-derived disability heat map of contusion locations showed that the strongest association with unfavorable outcomes was within the bilateral temporal and medial frontal lobes. These findings suggest that CT-based automated contusion measurement may be an immediately translatable strategy for improving TBI prognostic models.

Correlates of multidimensional sleep in premenopausal women: The BioCycle study

PurposeTo identify sleep dimensions (characteristics) that co-occur in premenopausal women. The second aim was to examine associations between multiple dimensions of sleep and a set of demographic, lifestyle, and health correlates. The overarching goal was to uncover patterns of poor-sleep correlates that might inform interventions to improve sleep health of women in this age group. MethodsThe BioCycle Study included 259 healthy women aged 18–44y recruited between 2005 and 2007 from Western New York. Participants reported sleep data through daily diaries and questionnaires that were used to create five sleep health dimensions (duration, variability, timing, latency, and continuity). We used multivariate analysis – canonical correlation methods – to identify links among dimensions of sleep health and patterns of demographic, psychological, and occupational correlates. ResultsTwo distinct combinations of sleep dimensions were identified. The first - primarily determined by low variability in nightly sleep duration, low variability in bedtime (timing), greater nocturnal awakening, and less sleep onset latency – was distinguished from the second – primarily determined by sleep duration.The first combination of sleep dimensions was associated with older age and higher parity, fewer depressive symptoms, and higher stress level. The second combination of sleep dimensions was associated with perception of longer sleep duration as optimal, lower parity, not engaging in shift work, older age, lower stress level, higher prevalence of depressive symptoms, and White race. ConclusionAmong premenopausal women, we demonstrated distinct patterns of sleep dimensions that co-occur and vary by demographic, health, and lifestyle correlates. These findings shed light on the correlates of sleep health vulnerabilities among young women.

ClusterMatch aligns single-cell RNA-sequencing data at the multi-scale cluster level via stable matching.

Unsupervised clustering of single-cell RNA sequencing (scRNA-seq) data holds the promise of characterizing known and novel cell type in various biological and clinical contexts. However, intrinsic multi-scale clustering resolutions poses challenges to deal with multiple sources of variability in the high-dimensional and noisy data. We present ClusterMatch, a stable match optimization model to align scRNA-seq data at the cluster level. In one hand, ClusterMatch leverages the mutual correspondence by canonical correlation analysis and multi-scale Louvain clustering algorithms to identify cluster with optimized resolutions. In the other hand, it utilizes stable matching framework to align scRNA-seq data in the latent space while maintaining interpretability with overlapped marker gene set. Through extensive experiments, we demonstrate the efficacy of ClusterMatch in data integration, cell type annotation, and cross-species/timepoint alignment scenarios. Our results show ClusterMatch's ability to utilize both global and local information of scRNA-seq data, sets the appropriate resolution of multi-scale clustering, and offers interpretability by utilizing marker genes. The code of ClusterMatch software is freely available at https://github.com/AMSSwanglab/ClusterMatch.

Fecal Microbial Dysbiosis is Associated with Colorectal Cancer Risk in a Korean Population.

The association between the fecal microbiota and colorectal cancer (CRC) risk has been suggested in epidemiologic studies. However, data from large-scale population-based studies are lacking. In this case-control study, we recruited 283 CRC patients from the Center for Colorectal Cancer, National Cancer Center Hospital, Korea to perform 16S rRNA gene sequencing of fecal samples. A total of 283 age- and sex-matched healthy participants were selected from 890 cohort of healthy Koreans that are publicly available (PRJEB33905). The microbial dysbiosis index (MDI) was calculated based on the differentially abundant species. The association between MDI and CRC risk was observed using conditional logistic regression. Sparse Canonical Correlation Analysis was performed to integrate species data with microbial pathways obtained by PICRUSt2. There is a significant divergence of the microbial composition between CRC patients and controls (PERMANOVA p=0.001). Those who were in third tertile of the MDI showed a significantly increased risk of CRC in the total population (OR: 6.93, 95% CI: 3.98-12.06, p-trend<0.001) compared to those in the lowest tertile. Similar results were found for men (OR: 6.28, 95% CI: 3.04-12.98-, p-trend<0.001) and women (OR: 7.39, 95% CI: 3.10-17.63, p-trend<0.001). Bacteroides coprocola and Bacteroides plebeius species and 12 metabolic pathways were interrelated in healthy controls that explain 91% covariation across samples. Dysbiosis in the fecal microbiota may be associated with an increased risk of CRC. Due to the potentially modifiable nature of the gut microbiota, our findings may have implications for CRC prevention among Koreans.

Exploring the Relationship between Self-Compassion and Psychological Pain: A Canonical Correlation Analysis.

This study investigated the association between self-compassion and psychological pain across various demographic variables. Using canonical correlation analysis, we observed an inverse relationship between the combined factors of the Self-Compassion Scale (SCS) and the Orbach and Mikulincer Mental Pain Scale (OMMP-8). Subgroup analyses revealed differences in SCS subscales among demographic groups with females, individuals with mental health diagnoses, and non-athletes displaying higher scores on negative SCS subscales and PsyPn. Injury status did not significantly affect self-compassion levels, although injured individuals scored higher on the irreversibility subscale of PsyPn. Negative SCS factors exhibited larger group differences and stronger correlations with PsyPn, indicating the potency of negative thinking in influencing psychological pain. These findings underscore the importance of self-compassion in mental health and suggest potential implications for intervention strategies.

A feasibility study of dosimetry for breast cancer radiotherapy based on body surface changes.

The requirement for precise and effective delivery of the actual dose to the patient grows along with the complexity of breast cancer radiotherapy. Dosimetry during treatment has become a crucial component of guaranteeing the efficacy and security. To propose a dosimetry method during breast cancer radiotherapy based on body surface changes. A total of 29 left breast cancer radiotherapy cases were retroactively retrieved from an earlier database for analysis. Non-rigid image registration and dose recalculation of the planning computed tomography (CT) referring to the Cone-beam computed tomography were performed to obtain dose changes. The study used 3D point cloud feature extraction to characterize body surface changes. Based on the correlation proof, a mapping model is developed between body surface changes and dose changes using neural network framework. The MSE metrics, the Euclidean distances of feature points and the 3D gamma pass rate metric were used to assess the prediction accuracy. A strong correlation exist between body surface changes and dose changes (first canonical correlation coefficient=0.950). For the dose deformation field and dose amplitude difference in the test set, the MSE of the predicted and actual values were 0.136 pixels and 0.229 cGy, respectively. After deforming the planning dose into a deformed one, the feature points' Euclidean distance between it and the recalculated dose changes from 9.267±1.879mm to 0.456±0.374mm. The 3D gamma pass rate of 90% or higher for the 2mm/2% criteria were achieved by 80.8% of all cases, with a minimum pass rate of 75.9% and a maximum pass rate of 99.6%. Pass rate for the 3mm/2% criteria ranged from 87.8% to 99.8%, with 92.3% of the cases having a pass rate of 90% or higher. This study provides a dosimetry method that is non-invasive, real-time, and requires no additional dose for breast cancer radiotherapy.

Quality of native goat's milk from a multivariate perspective

The aim of this study was to assess the relationship between physical components and the fatty acid (FA) profile of goat's milk, compare the physical and chemical variables of the milk using canonical correlation, and verify the contribution of the physical variables in predicting the FAs in the milk, contributing to the selection of the FA profile by the correlated response based on the physical profile of the milk. A total of 24 native goats, weighing 42 ± 4 kg and 30 ± 3 days into lactation, were used. The data were subjected to canonical analysis to identify the variables that best discriminate. The main objective of this procedure is to find the best set of variables to compose the discriminant function. Four components were needed to explain 70.68% of the total variation in the physical variables and FA profile in the milk of native goats. The percentage of variance accumulated in the first two factors is 41%. The most important variables in the first factor were the short-chain FAs, namely, caproic (6C), caprylic (8C), capric (10C), and lauric (C12), which are responsible for giving the milk its characteristic and even unpleasant odor, depending on their concentration. New studies with a larger sample of data should be carried out to better understand the relationships between physical and chemical variables in the milk of native goats as well as other breeds of goats for comparison. In addition, the weak simple correlations between the variables studied contributed to the results obtained.

Phytoplankton Diversity, Spatial Patterns, and Photosynthetic Characteristics Under Environmental Gradients and Anthropogenic Influence in the Pearl River Estuary.

The Pearl River Estuary (PRE) is one of the world's most urbanized subtropical coastal systems. It presents a typical environmental gradient suitable for studying estuarine phytoplankton communities' dynamics and photosynthetic physiology. In September 2018, the maximum photochemical quantum yield (Fv/Fm) of phytoplankton in different salinity habitats of PRE (oceanic, estuarine, and freshwater zones) was studied, revealing a complex correlation with the environment. Fv/Fm of phytoplankton ranged from 0.16 to 0.45, with taxa in the upper Lingdingyang found to be more stressed. Community composition and structure were analyzed using 18S rRNA, accompanied by a pigment analysis utilized as a supplementary method. Nonmetric multidimensional scaling analysis indicated differences in the phytoplankton spatial distribution along the estuarine gradients. Specificity-occupancy plots identified different specialist taxa for each salinity habitat. Dinophyta and Haptophyta were the predominant taxa in oceanic areas, while Chlorophyta and Cryptophyta dominated freshwater. Bacillariophyta prevailed across all salinity gradients. Canonical correlation analysis and Mantel tests revealed that temperature, salinity, and elevated nutrient levels (i.e., NO3--N, PO43--P, and SiO32--Si) associated with anthropogenic activities significantly influenced the heterogeneity of community structure. The spatial distribution of phytoplankton, along with in situ photosynthetic characteristics, serves as a foundational basis to access estuarine primary productivity, as well as community function and ecosystem health.

Canonical correlation analysis based on robust covariance matrix by using deterministic of minimum covariance determinant

Canonical correlation analysis (CCA) study the linear combinations between a two multivariate set of variable that have the maximum association among these two sets of variables. The main computation of the CCA is depend on the sample mean and covariance matrices, where they are very sensitive and highly effected by presence of outliers. In this study a new procedure is used on CCA to obtain the robust canonical correlation analysis (RCCA) to control and conquer the deformities of the sample mean and covariance matrices in contaminated data set. The deterministic of minimum covariance determinant (DetMCD) of Mia et al. (1936) is applied on CCA and the superiority performance of RCCA is demonstrated CCA method. A simulation is running for both methods on real and generation data.

Financial security of Ukraine under martial law: Impact of macroeconomic determinants

Russia’s open aggression against Ukraine has resulted in significant changes across all sectors of the Ukrainian economy and its financial sphere, including financial security. The paper aims to identify the impact of the primary macroeconomic determinants, i.e., military defense spending, non-performing bank loans, exchange rate, foreign debt, and state (total) reserves, on the financial security of Ukraine under martial law. The canonical correlation analysis is employed to assess the strength of the relationship between the above macroeconomic indicators and the level of the state’s financial security. It was found that the reduction of the state’s financial security level in 2022 was 63.9%, explained exactly by the changes in the above macroeconomic determinants after the start of a full-scale invasion. The study determined the degree of influence of each indicator on Ukraine’s financial security level. An increase in the level of military defense spending, non-performing bank loans, hryvnia’s devaluation, and external debt growth had a direct negative impact on Ukraine’s financial security. At the same time, an upsurge in total reserves had an indirect negative impact (through the external debt growth). The research findings confirm the necessity for effective monitoring and management of the macroeconomic indicators to maintain both Ukraine’s financial security and macro-financial stability in order to ensure its’ sustainable economic development during the postwar recovery period. AcknowledgmentThis research is financially supported by the NATO SPS Program “Security of territorial communities: evidence from the Eastern European countries”.

Linked patterns of symptoms and cognitive covariation with functional brain controllability in major depressive disorder

Controllability analysis is an approach developed for evaluating the ability of a brain region to modulate function in other regions, which has been found to be altered in major depressive disorder (MDD). Both depressive symptoms and cognitive impairments are prominent features of MDD, but the case-control differences of controllability between MDD and controls can not fully interpret the contribution of both clinical symptoms and cognition to brain controllability and linked patterns among them in MDD. Sparse canonical correlation analysis was used to investigate the associations between resting-state functional brain controllability at the network level and clinical symptoms and cognition in 99 first-episode medication-naïve patients with MDD. Average controllability was significantly correlated with clinical features. The average controllability of the dorsal attention network (DAN) and visual network had the highest correlations with clinical variables. Among clinical variables, depressed mood, suicidal ideation and behaviour, impaired work and activities, and gastrointestinal symptoms were significantly negatively associated with average controllability, and reduced cognitive flexibility was associated with reduced average controllability. These findings highlight the importance of brain regions in modulating activity across brain networks in MDD, given their associations with symptoms and cognitive impairments observed in our study. Disrupted control of brain reconfiguration of DAN and visual network during their state transitions may represent a core brain mechanism for the behavioural impairments observed in MDD. National Natural Science Foundation of China (82001795 and 82027808), National Key R&D Program (2022YFC2009900), and Sichuan Science and Technology Program (2024NSFSC0653).

Emphasizing symbolic capital: its superior influence on the association between family socioeconomic status and adolescent subjective well-being uncovered by a large-scale multivariate network analysis.

Family socioeconomic status (FSES) serves as a significant determinant for subjective well-being. However, extant research has provided conflicting evidence on the correlation between FSES and adolescent students' subjective well-being (SSWB). Data were collected from 12,058 adolescent students (16 years of age) by the Programme for International Student Assessment (PISA) 2018. Multivariate canonical correlation and Mantel test were utilized to investigate the specific connection between FSES and SSWB. Furthermore, a Gaussian EBICglasso graph-theoretical model was used to capture the topological properties of the FSES-SSWB network and reveal the interplay among multifarious components of FSES and SSWB. FSES was positively correlated with SSWB. In the FSES-SSWB network, parental educational attainment and occupation status demonstrated the highest centrality values, thereby contributing significantly to the relationship between FSES and SSWB. However, family wealth, along with educational and cultural resources, displayed lower centrality values, signifying their weaker roles in this relationship. Our findings suggest that symbolic capital, rather than family affluence, exerts a dominant influence on adolescent SSWB. In other words, SSWB may not be detrimentally influenced by a deficiency in monetary resources. However, it is more susceptible to being unfavorably impacted by inferior parental educational attainment and occupational standing.

Team interaction behaviors correlates with team creativity among nursing students: Canonical correlation and moderation analyses

BackgroundPrior studies have indicated team members’ interaction behaviors may predict creativity among nursing students.MethodsThis study investigated the correlation between interaction behaviors and creativity, both individual- and team-level, among nursing students. In this cross-sectional quantitative study, data were obtained from self-reported questionnaires. Individual creativity was assessed using the Torrance Tests of Creative Thinking scale; the perceived team interaction behavior and team creativity were assessed using validated instruments. Canonical correlation analysis was conducted to determine the overall correlation between interaction behaviors, and creativity, and the moderating effect of female proportion dominance was also examined.ResultsA total of 164 nursing students (84.1% female) arranged into 14 teams were included in this study. Canonical correlation analysis showed a positive correlation between interaction behaviors and creativity (correlation = 0.88). All dimensions of interactive behaviors were positively related to creativity dimensions. A stronger correlation to team creativity (correlation = 1) was found compared to individual creativity (correlation = 0.07). This study demonstrated that individual interactive behaviors including spontaneous communication and helping behavior predicted high team creativity.ConclusionsThis insight may be valuable for nursing education programs seeking to foster creativity and effective teamwork. The potential moderating effect of female proportions on team interaction behaviors and creativity should be investigated further.

Robust statistical industrial fault monitoring: A machine learning-based distributed CCA and low frequency control charts

Over the past two decades, there has been a notable increase in the complexity and dynamism of industrial and manufacturing systems. Traditional fault detection strategies exhibit limited robustness in terms of detection accuracy, inefficiency, and poor detection outcomes. This paper presents an innovative, robust machine learning (ML) based distributed canonical correlation analysis (DCCA) and low-frequency generated univariate control charts CUSUM and EWMA using wavelet transforms (WT) to monitor industrial and manufacturing systems, specifically focusing on abnormalities and fault detection. The framework’s performance is tested using previous algorithms such as multiscale PCA and PLS, and the (CSTR) is utilized as the benchmark case application. The results illustrate that the ML-based multiscale low-frequency (DCCA) monitoring charts for fault cases exhibited an (MDR) of 0 %, (FAR) of 0.2 %, (FDR) of 100 %, precision of 99.8 %, and an F1-score of 99.9 %. This method enables robust detection and ensures complete monitoring of aberrant circumstances.

A novel NMF-DiCCA deep learning method and its application in wind turbine blade icing failure identification

Wind turbine blade icing data has the characteristics of multi-source and multi variable. It is difficult to characterize and identify the icing failure with multi-scale features. In this paper, a novel Non-negative Matrix Factorization-Dynamic-inner Canonical Correlation Analysis (NMF-DiCCA )based on GRU and SSAE algorithm is proposed to solve this problem. Firstly, using NMF instead of SVD decomposition method in DiCCA algorithm, the NMF-DiCCA is applied to obtain the dynamic latent variable of time serie. Secondly, the latent structure features S of dynamic latent variable is captured by SSAE. Thirdly, the temporal correlation hidden feature H of dynamic latent variable is extracted by Gated Recurrent Unit (GRU). Finally, the attention weight distribution between latent structure S and temporal correlation hidden feature H is integrated using the attention mechanism, and the fusion feature is reconstructed using the improved SSAE(ISSAE) based on GRU and SSAE.