Repeated measures ASCA+ for analysis of longitudinal intervention studies with multivariate outcome data

Madssen, Torfinn Støve; Giskeødegård, Guro F.; Smilde, Age K.; Westerhuis, Johan A.

doi:10.1371/journal.pcbi.1009585

Cited by 24 publications

(41 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Repeated measures ANOVA simultaneous component analysis+ (RM-ASCA+) was used for multivariate analysis ( 29 ). This method extends repeated measures linear mixed models to the multivariate case, by first decomposing the multivariate response matrix into effect matrices according to the specified LMM.…”

Section: Methodsmentioning

confidence: 99%

Longitudinal Changes in Circulating Metabolites and Lipoproteins After Breast Cancer Treatment

et al. 2022

Self Cite

View full text Add to dashboard Cite

The multimodal treatment of breast cancer may induce long term effects on the metabolic profile and increase the risk of future cardiovascular disease. In this study, we characterized longitudinal changes in serum lipoprotein subfractions and metabolites after breast cancer treatment, aiming to determine the long-term effect of different treatment modalities. Further, we investigated the prognostic value of treatment-induced changes in breast cancer-specific and overall 10-year survival. In this study, serum samples from breast cancer patients (n = 250) were collected repeatedly before and after radiotherapy, and serum metabolites and lipoprotein subfractions were quantified by NMR spectroscopy. Longitudinal changes were assessed by univariate and multivariate data analysis methods applicable for repeated measures. Distinct changes were detectable in levels of lipoprotein subfractions and circulating metabolites during the first year, with similar changes despite large differences in treatment regimens. We detect increased free cholesterol and decreased esterified cholesterol levels of HDL subfractions, a switch towards larger LDL particles and higher total LDL-cholesterol, in addition to a switch in the glutamine-glutamate ratio. Non-survivors had different lipid profiles from survivors already at baseline. To conclude, our results show development towards an atherogenic lipid profile in breast cancer patients with different treatment regimens.

show abstract

Section: Methodsmentioning

confidence: 99%

Longitudinal Changes in Circulating Metabolites and Lipoproteins After Breast Cancer Treatment

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another method that uses LMMs is repeated measures ASCA+ (RM-ASCA+). 73 This method is aimed at analysis of longitudinal metabolomic data, and involves the use of repeated measures LMMs to estimate temporal multivariate trajectories, as well as time-dependent treatment effects. In contrast to LiMM-PCA, RM-ASCA+ is applied directly to the response matrix Y.…”

Section: Extensions To Mixed Modelsmentioning

confidence: 99%

Multivariate analysis of NMR‐based metabolomic data

et al. 2021

Self Cite

View full text Add to dashboard Cite

Nuclear magnetic resonance (NMR) spectroscopy allows for simultaneous detection of a wide range of metabolites and lipids. As metabolites act together in complex metabolic networks, they are often highly correlated, and optimal biological insight is achieved when using methods that take the correlation into account. For this reason, latent-variable-based methods, such as principal component analysis and partial leastsquares discriminant analysis, are widely used in metabolomic studies. However, with increasing availability of larger population cohorts, and a shift from analysis of spectral data to using quantified metabolite levels, both more traditional statistical approaches and alternative machine learning methods have become more widely used. This review aims at providing an overview of the current state-of-the-art multivariate methods for the analysis of NMR-based metabolomic data as well as alternative methods, highlighting their strengths and limitations.

show abstract

“…One powerful approach for analysis of multivariate data is the ANOVA simultaneous component analysis (ASCA) framework that combines ANOVA with PCA ( Smilde et al, 2005 ; Smilde et al, 2012 ). More recently, extended ASCA methods such as ASCA + ( Thiel et al, 2017 ), LiMM-PCA, and repeated measures ASCA + (RM-ASCA + , Martin and Govaerts, 2020 ; Madssen et al, 2021 ) have emerged that combine general linear (mixed) models with PCA. In this way the flexibility of regression models are merged with the visualization of multivariate analysis, providing excellent interpretability by allowing to separate and display the complex multivariate patterns originating from different experimental factors.…”

Section: Introductionmentioning

confidence: 99%

“…In ordinary ASCA, M 0 usually represents the grand mean matrix, whereas in RM-ASCA + it typically either represents the baseline mean of all, or one of the groups, depending on how the effects are coded in the model. The effects h reflect the statistical and experimental design (for examples, see Madssen et al, 2021 ). In the context of a longitudinal study, an effect matrix M T would represent the effect of time, i.e., the change from baseline.…”

Section: Introductionmentioning

confidence: 99%

“…The multiblock package for R offers a comprehensive set of methods for multiblock analysis, including various ASCA-based methods such as LiMM-PCA, generalized ASCA, RM-ASCA + , and covariates ASCA ( Liland, 2022 ; Smilde et al, 2022 ). A Matlab implementation of RM-ASCA + has been published by Madssen et al (2021) , (scripts available at https://github.com/ntnu-mr-cancer/RM_ASCA ). An extension of RM-ASCA + has been proposed in the case of zero-inflated count data, namely the zero-inflated counts (ZIC)RM-ASCA + by applying zero-inflated negative binomial mixed models, with code available for R ( https://github.com/AukeHaver/ZICRM-ASCA_plus ).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

ALASCA: An R package for longitudinal and cross-sectional analysis of multivariate data by ASCA-based methods

Jarmund

Madssen²,

Giskeødegård³

2022

Front. Mol. Biosci.

View full text Add to dashboard Cite

The increasing availability of multivariate data within biomedical research calls for appropriate statistical methods that can describe and model complex relationships between variables. The extended ANOVA simultaneous component analysis (ASCA+) framework combines general linear models and principal component analysis (PCA) to decompose and visualize the separate effects of experimental factors. It has recently been demonstrated how linear mixed models can be included in the framework to analyze data from longitudinal experimental designs with repeated measurements (RM-ASCA+). The ALASCA package for R makes the ASCA+ framework accessible for general use and includes multiple methods for validation and visualization. The package is especially useful for longitudinal data and the ability to easily adjust for covariates is an important strength. This paper demonstrates how the ALASCA package can be applied to gain insights into multivariate data from interventional as well as observational designs. Publicly available data sets from four studies are used to demonstrate the methods available (proteomics, metabolomics, and transcriptomics).

show abstract

Repeated measures ASCA+ for analysis of longitudinal intervention studies with multivariate outcome data

Cited by 24 publications

References 38 publications

Longitudinal Changes in Circulating Metabolites and Lipoproteins After Breast Cancer Treatment

Longitudinal Changes in Circulating Metabolites and Lipoproteins After Breast Cancer Treatment

Multivariate analysis of NMR‐based metabolomic data

ALASCA: An R package for longitudinal and cross-sectional analysis of multivariate data by ASCA-based methods

Contact Info

Product

Resources

About