Considerably Improving Clustering Algorithms Using UMAP Dimensionality Reduction Technique: A Comparative Study

Allaoui, Mebarka; Kherfi, Mohammed Lamine; Cheriet, Abdelhakim

doi:10.1007/978-3-030-51935-3_34

Cited by 132 publications

(83 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, deep learning approaches open new avenues from feature mining ( Dollar et al, 2007 ), over retinal ganglion counting ( Masin et al, 2021 ), to fundus and OCT analysis ( Badar et al, 2020 ). Furthermore, this is accompanied by an increased assessment of feature and data connectivity and relationships, such as principal component analysis and Uniform Manifold Approximation and Projection ( Allaoui et al, 2020 ). Additionally, carefully designed data analysis workflows will provide new insights into our data and the NVU, most likely in a way that is beyond current comprehension.…”

Section: Future Directions and Areas Of Scientific Interestmentioning

confidence: 99%

The “Neuro-Glial-Vascular” Unit: The Role of Glia in Neurovascular Unit Formation and Dysfunction

Kugler

Greenwood²,

Macdonald³

2021

Front. Cell Dev. Biol.

111

View full text Add to dashboard Cite

The neurovascular unit (NVU) is a complex multi-cellular structure consisting of endothelial cells (ECs), neurons, glia, smooth muscle cells (SMCs), and pericytes. Each component is closely linked to each other, establishing a structural and functional unit, regulating central nervous system (CNS) blood flow and energy metabolism as well as forming the blood-brain barrier (BBB) and inner blood-retina barrier (BRB). As the name suggests, the “neuro” and “vascular” components of the NVU are well recognized and neurovascular coupling is the key function of the NVU. However, the NVU consists of multiple cell types and its functionality goes beyond the resulting neurovascular coupling, with cross-component links of signaling, metabolism, and homeostasis. Within the NVU, glia cells have gained increased attention and it is increasingly clear that they fulfill various multi-level functions in the NVU. Glial dysfunctions were shown to precede neuronal and vascular pathologies suggesting central roles for glia in NVU functionality and pathogenesis of disease. In this review, we take a “glio-centric” view on NVU development and function in the retina and brain, how these change in disease, and how advancing experimental techniques will help us address unanswered questions.

show abstract

Section: Future Directions and Areas Of Scientific Interestmentioning

confidence: 99%

The “Neuro-Glial-Vascular” Unit: The Role of Glia in Neurovascular Unit Formation and Dysfunction

Kugler

Greenwood²,

Macdonald³

2021

Front. Cell Dev. Biol.

111

View full text Add to dashboard Cite

show abstract

“…While output cellular functionality is largely imposed by surface protein, the transcriptome is the vastly more complex precursor, now almost completely measurable by scRNA-seq without bias of preselect markers providing a holistic transcriptomic signature for an individual cell. Algorithms, outlined by Allaoui et al (2020), then "pull" similar RNA signatures together by k-means clustering, enabling visualized clusters in t-distributed stochastic neighborhood embedding (t-SNE) or uniform manifold approximation and projection (UMAP) plots to be grouped and independently characterized direct from sample suspension.…”

Section: Utility Of Scrna-seq: Unsupervised Holistic Dissection Of Signature Transcriptomesmentioning

confidence: 99%

Visual Genomics Analysis Studio as a Tool to Analyze Multiomic Data

et al. 2021

View full text Add to dashboard Cite

Type B adverse drug reactions (ADRs) are iatrogenic immune-mediated syndromes with mechanistic etiologies that remain incompletely understood. Some of the most severe ADRs, including delayed drug hypersensitivity reactions, are T-cell mediated, restricted by specific human leukocyte antigen risk alleles and sometimes by public or oligoclonal T-cell receptors (TCRs), central to the immunopathogenesis of tissue-damaging response. However, the specific cellular signatures of effector, regulatory, and accessory immune populations that mediate disease, define reaction phenotype, and determine severity have not been defined. Recent development of single-cell platforms bringing together advances in genomics and immunology provides the tools to simultaneously examine the full transcriptome, TCRs, and surface protein markers of highly heterogeneous immune cell populations at the site of the pathological response at a single-cell level. However, the requirement for advanced bioinformatics expertise and computational hardware and software has often limited the ability of investigators with the understanding of diseases and biological models to exploit these new approaches. Here we describe the features and use of a state-of-the-art, fully integrated application for analysis and visualization of multiomic single-cell data called Visual Genomics Analysis Studio (VGAS). This unique user-friendly, Windows-based graphical user interface is specifically designed to enable investigators to interrogate their own data. While VGAS also includes tools for sequence alignment and identification of associations with host or organism genetic polymorphisms, in this review we focus on its application for analysis of single-cell TCR–RNA–Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE)-seq, enabling holistic cellular characterization by unbiased transcriptome and select surface proteome. Critically, VGAS does not require user-directed coding or access to high-performance computers, instead incorporating performance-optimized hidden code to provide application-based fast and intuitive tools for data analyses and production of high-resolution publication-ready graphics on standard specification laptops. Specifically, it allows analyses of comprehensive single-cell TCR sequencing (scTCR-seq) data, detailing (i) functional pairings of α–β heterodimer TCRs, (ii) one-click histograms to display entropy and gene rearrangements, and (iii) Circos and Sankey plots to visualize clonality and dominance. For unbiased single-cell RNA sequencing (scRNA-seq) analyses, users extract cell transcriptome signatures according to global structure via principal component analysis, t-distributed stochastic neighborhood embedding, or uniform manifold approximation and projection plots, with overlay of scTCR-seq enabling identification and selection of the immunodominant TCR-expressing populations. Further integration with similar sequence-based detection of surface protein markers using oligo-labeled antibodies (CITE-seq) provides comparative understanding of surface protein expression, with differential gene or protein analyses visualized using volcano plot or heatmap functions. These data can be compared to reference cell atlases or suitable controls to reveal discrete disease-specific subsets, from epithelial to tissue-resident memory T-cells, and activation status, from senescence through exhaustion, with more finite transcript expression displayed as violin and box plots. Importantly, guided tutorial videos are available, as are regular application updates based on the latest advances in bioinformatics and user feedback.

show abstract

“…Unlike t-SNE, which is a locally focused method, the UMAP preserves both local and global structure. It also boasts shorter run times and applicability to big datasets (Allaoui et al, 2020). Briefly, UMAP uses the k-nearest neighbor concept and builds a high-dimensional graph of the input data.…”

Section: Dimensionality Reduction Tools For Visualizing Organellar Mapmentioning

confidence: 99%

Bioinformatic Analysis of Temporal and Spatial Proteome Alternations During Infections

2021

View full text Add to dashboard Cite

Microbial pathogens have evolved numerous mechanisms to hijack host’s systems, thus causing disease. This is mediated by alterations in the combined host-pathogen proteome in time and space. Mass spectrometry-based proteomics approaches have been developed and tailored to map disease progression. The result is complex multidimensional data that pose numerous analytic challenges for downstream interpretation. However, a systematic review of approaches for the downstream analysis of such data has been lacking in the field. In this review, we detail the steps of a typical temporal and spatial analysis, including data pre-processing steps (i.e., quality control, data normalization, the imputation of missing values, and dimensionality reduction), different statistical and machine learning approaches, validation, interpretation, and the extraction of biological information from mass spectrometry data. We also discuss current best practices for these steps based on a collection of independent studies to guide users in selecting the most suitable strategies for their dataset and analysis objectives. Moreover, we also compiled the list of commonly used R software packages for each step of the analysis. These could be easily integrated into one’s analysis pipeline. Furthermore, we guide readers through various analysis steps by applying these workflows to mock and host-pathogen interaction data from public datasets. The workflows presented in this review will serve as an introduction for data analysis novices, while also helping established users update their data analysis pipelines. We conclude the review by discussing future directions and developments in temporal and spatial proteomics and data analysis approaches. Data analysis codes, prepared for this review are available from https://github.com/BabuLab-UofR/TempSpac, where guidelines and sample datasets are also offered for testing purposes.

show abstract

Considerably Improving Clustering Algorithms Using UMAP Dimensionality Reduction Technique: A Comparative Study

Cited by 132 publications

References 12 publications

The “Neuro-Glial-Vascular” Unit: The Role of Glia in Neurovascular Unit Formation and Dysfunction

The “Neuro-Glial-Vascular” Unit: The Role of Glia in Neurovascular Unit Formation and Dysfunction

Visual Genomics Analysis Studio as a Tool to Analyze Multiomic Data

Bioinformatic Analysis of Temporal and Spatial Proteome Alternations During Infections

Contact Info

Product

Resources

About