Parameter clustering in Bayesian functional principal component analysis of neuroscientific data

Margaritella, Nicoló; Inácio, Vanda; King, Ruth

doi:10.1002/sim.8768

Cited by 10 publications

(4 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Clustering is a common goal for applied statistical analysis across many fields, and has grown in popularity alongside other unsupervised machine learning methods in recent years [ 62 ]. In the context of health and medical research, clustering methods comprise a versatile set of statistical tools with a wide variety of potential applications including design of clinical trials [ 63 ], building data-driven profiles of individuals using functional biosignals data [ 64 ], and identifying clinical or epidemiological subtypes based on multivariate longitudinal observations [ 65 ]. Bayesian model averaging offers an intuitive and elegant framework to access more robust insights by combining inference across multiple clustering solutions.…”

Section: Discussionmentioning

confidence: 99%

clusterBMA: Bayesian model averaging for clustering

Forbes,

Santos-Fernandez,

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

Various methods have been developed to combine inference across multiple sets of results for unsupervised clustering, within the ensemble clustering literature. The approach of reporting results from one ‘best’ model out of several candidate clustering models generally ignores the uncertainty that arises from model selection, and results in inferences that are sensitive to the particular model and parameters chosen. Bayesian model averaging (BMA) is a popular approach for combining results across multiple models that offers some attractive benefits in this setting, including probabilistic interpretation of the combined cluster structure and quantification of model-based uncertainty. In this work we introduce clusterBMA, a method that enables weighted model averaging across results from multiple unsupervised clustering algorithms. We use clustering internal validation criteria to develop an approximation of the posterior model probability, used for weighting the results from each model. From a combined posterior similarity matrix representing a weighted average of the clustering solutions across models, we apply symmetric simplex matrix factorisation to calculate final probabilistic cluster allocations. In addition to outperforming other ensemble clustering methods on simulated data, clusterBMA offers unique features including probabilistic allocation to averaged clusters, combining allocation probabilities from ‘hard’ and ‘soft’ clustering algorithms, and measuring model-based uncertainty in averaged cluster allocation. This method is implemented in an accompanying R package of the same name. We use simulated datasets to explore the ability of the proposed technique to identify robust integrated clusters with varying levels of separation between subgroups, and with varying numbers of clusters between models. Benchmarking accuracy against four other ensemble methods previously demonstrated to be highly effective in the literature, clusterBMA matches or exceeds the performance of competing approaches under various conditions of dimensionality and cluster separation. clusterBMA substantially outperformed other ensemble methods for high dimensional simulated data with low cluster separation, with 1.16 to 7.12 times better performance as measured by the Adjusted Rand Index. We also explore the performance of this approach through a case study that aims to identify probabilistic clusters of individuals based on electroencephalography (EEG) data. In applied settings for clustering individuals based on health data, the features of probabilistic allocation and measurement of model-based uncertainty in averaged clusters are useful for clinical relevance and statistical communication.

show abstract

Section: Discussionmentioning

confidence: 99%

clusterBMA: Bayesian model averaging for clustering

Forbes,

Santos-Fernandez,

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…An unsupervised principal component analysis (PCA) and heatmaps with dendrograms were performed on all analyzed samples based on the fraction of ten infiltrating immune cells as determined by QuanTIseq. The analysis was based on the methods reported by Vichi et al and Granato et al [19,20]. Principal Component Analysis (PCA) was visualized through the built-in R functions prcomp().and the visualized first two (PC1 and PC2) projection scatterplots were built using the ggplot2 package of R function.…”

Section: Identification and Visualization Of Immuno-subtypes Of The Tumor Samplesmentioning

confidence: 99%

Comprehensive analysis of the immunogenomic landscape and clinical features in cervical cancer

Fan

2021

Preprint

View full text Add to dashboard Cite

Immunotherapy has changed the standard of treatment for many cancers. However, the same treatments showed disappointing outcomes in cervical cancer so far. Thus, understanding the mechanisms that support the immune tolerance of cervical cancer will provide a way to design new strategies to enhance immunotherapies. Here, we characterized cellular compositions of the immune infiltrates in cervical cancer and investigated if the tumor immune landscape is a predictor for patient prognosis. The fraction of ten immune infiltrates of cervical and other cancers were analyzed by using QuanTIseq software base on the bulk mRNA sequencing data from The Cancer Genome Atlas Program (TCGA). Cervical cancer is one of the cancers that had the lowest percentage of total immune infiltrates, but it had the highest ratio for CD8 T cells to all immune infiltrates among all solid cancers. Both the principal components (PCA) analysis and heatmap with dendrogram analysis showed that cervical cancer had a similar immune infiltrated microenvironment with other squamous cell carcinomas, such as head and neck cancer and lung squamous cell cancer. The PCA and heatmap with dendrogram analysis showed that cervical cancer and HPV positive head and neck cancers were clustered more closer and partially separated with HPV negative head and neck cancer. Further analysis showed that HPV-positive cervical and head and neck cancers had a significantly higher level of CD8 T cells and M1-liked macrophages, but a lower level of M2 macrophages. The survival analysis showed that a higher level of CD8 T cells was associated with a better patient prognosis. However, immuno-suppressive immune infiltrates including M2 macrophages and Treg cells that are known to suppress anti-tumor immunity also demonstrated positive patient overall survival. Our study provided a conceptual framework to understand the tumor immune microenvironment of cervical cancer. Our results also demonstrated that the immune infiltrates can be a prognosis marker for cervical cancer.

show abstract

“…Various alternative approaches to dimensionality reduction of fMRI data have been proposed in the literature, including adaptations of Principal Component Analysis (PCA) to the context of large-scale datasets (see, e.g., [7]). Moreover, Blood-Oxygenation Level Dependent (BOLD) signals have also been considered as functional data [8][9][10][11], and analysed through functional PCA methods (see, e.g., [12,13]). However, high dimensionality is not the only challenge when applying PCA to the analysis of fMRI data.…”

Section: Introductionmentioning

confidence: 99%

Anatomically compliant modes of variations: New tools for brain connectivity

Clementi,

Arnone,

Santambrogio

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

Anatomical complexity and data dimensionality present major issues when analysing brain connectivity data. The functional and anatomical aspects of the connections taking place in the brain are in fact equally relevant and strongly intertwined. However, due to theoretical challenges and computational issues, their relationship is often overlooked in neuroscience and clinical research. In this work, we propose to tackle this problem through Smooth Functional Principal Component Analysis, which enables to perform dimensional reduction and exploration of the variability in functional connectivity maps, complying with the formidably complicated anatomy of the grey matter volume. In particular, we analyse a population that includes controls and subjects affected by schizophrenia, starting from fMRI data acquired at rest and during a task-switching paradigm. For both sessions, we first identify the common modes of variation in the entire population. We hence explore whether the subjects’ expressions along these common modes of variation differ between controls and pathological subjects. In each session, we find principal components that are significantly differently expressed in the healthy vs pathological subjects (with p-values < 0.001), highlighting clearly interpretable differences in the connectivity in the two subpopulations. For instance, the second and third principal components for the rest session capture the imbalance between the Default Mode and Executive Networks characterizing schizophrenia patients.

show abstract

Parameter clustering in Bayesian functional principal component analysis of neuroscientific data

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Cited by 10 publications

References 66 publications

clusterBMA: Bayesian model averaging for clustering

clusterBMA: Bayesian model averaging for clustering

Comprehensive analysis of the immunogenomic landscape and clinical features in cervical cancer

Anatomically compliant modes of variations: New tools for brain connectivity

Contact Info

Product

Resources

About