Evaluation of Distance Metrics and Spatial Autocorrelation in Uniform Manifold Approximation and Projection Applied to Mass Spectrometry Imaging Data

Smets, Tina; Verbeeck, Nico; Claesen, Marc; Asperger, Arndt; Griffioen, Gerard; Tousseyn, Thomas; Waelput, Wim; Waelkens, Etienne; Moor, Bart De

doi:10.1021/acs.analchem.8b05827

Cited by 88 publications

(118 citation statements)

References 18 publications

Supporting

Mentioning

114

Contrasting

Order By: Relevance

“…The recent appearance of single-cell expression profiling data paired with CRISPR-induced mutations will be an especially useful source of data of this type, as these experiments include increasingly larger numbers of mutations. 19 While many of the most useful applications of dimensionality reduction tend to arise from single-cell genomics, for which typical datasets necessitate approaches like UMAP to define relationships between cells, these approaches may also prove useful in visualizing the spatial relationships of biomolecules in tissues, 20 genetic interactions, or relationships between human populations. 21…”

mentioning

confidence: 99%

Dimensionality reduction by UMAP to visualize physical and genetic interactions

Dorrity

Saunders

Queitsch

et al. 2019

Preprint

View full text Add to dashboard Cite

Dimensionality reduction is often used to visualize complex expression profilingdata. Here, we use the Uniform Manifold Approximation and Projection (UMAP) method on published transcript profiles of 1484 single gene deletions of Saccharomyces cerevisiae. Proximity in low-dimensional UMAP space identifies clusters of genes that correspond to protein complexes and pathways, and finds novel protein interactions even within well-characterized complexes. This approach is more sensitive than previous methods and should be broadly useful as additional transcriptome datasets become available for other organisms.A central goal of biological studies is the identification and characterization of proteins that act in a common cellular pathway. Efforts towards this goal have been greatly aided by large-scale perturbation analyses coupled with whole-transcriptome profiling, in which each gene's transcriptional response to a perturbation is measured. If a sufficient database of expression profiles exists, then a pathway affected by an uncharacterized

show abstract

mentioning

confidence: 99%

Dimensionality reduction by UMAP to visualize physical and genetic interactions

Dorrity

Saunders

Queitsch

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Briefly, UMAP aims to form a topological representation of the original high-dimensional data by finding local manifold approximates using their local fuzzy simplicial set representation. Prior work has shown promising results of applying UMAP on spectra in MSI data [36] (as opposed to ion images as we are doing here).…”

Section: Clustering Stepsmentioning

confidence: 87%

Spatially-Aware Clustering of Ion Images in Mass Spectrometry Imaging Data Using Deep Learning

Zhang

Claesen

Moerman

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Computational analysis is crucial to capitalize on the wealth of spatio-molecular information generated by mass spectrometry imaging (MSI) experiments. Currently, the spatial information available in MSI data is often under-utilized, due to the challenges of in-depth spatial pattern extraction.The advent of deep learning has greatly facilitated such complex spatial analysis. In this work, we use a pre-trained neural network to extract high-level features from ion images in MSI data, and test whether this improves downstream data analysis. The resulting neural network interpretation of ion images, coined neural ion images, are used to cluster ion images based on spatial expressions.We evaluate the impact of neural ion images on two ion image clustering pipelines, namely DBSCAN clustering, combined with UMAP-based dimensionality reduction, and k-means clustering. In both pipelines, we compare regular and neural ion images from two different MSI datasets. All tested pipelines could extract underlying spatial patterns, but the neural network-based pipelines provided better assignment of ion images, with more fine-grained clusters, and greater consistency in the spatial structures assigned to individual clusters.Additionally, we introduce the Relative Isotope Ratio metric to quantitatively evaluate clustering quality. The resulting scores show that isotopical m/z values are more often clustered together in the neural network-based pipeline, indicating improved clustering outcomes.The usefulness of neural ion images extends beyond clustering towards a generic framework to incorporate spatial information into any MSI-focused machine learning pipeline, both supervised and unsupervised.

show abstract

“…We plotted the expression counts of the samples class-wise (i.e. "Mtb Infected", "Ag85 Stimulated", "ESAT Stimulated" and "Unstimulated") by Uniform Manifold Approximation and Projection (UMAP) [26]. Interestingly, we found the cluster among three classes (i.e.…”

Section: Exploratory Data Analysismentioning

confidence: 95%

lncRNA Mediated Hijacking of T-cell Hypoxia Response Pathway byMycobacterium tuberculosisPredicts Latent to Active Progression in Humans

Mehra

Kumar

Srivastava

et al. 2020

Preprint

View full text Add to dashboard Cite

Cytosolic functions of Long non-coding RNAs including mRNA translation masking and sponging are major regulators of biological pathways. Formation of T cell-bounded hypoxic granuloma is a host immune defense for containing infected Mtb-macrophages. Our study exploits the mechanistic pathway of Mtb-induced HIF1A silencing by the antisense lncRNA-HIF1A-AS2 in T cells. Computational analysis of in-vitro T-cell stimulation assays in progressors(n=119) versus latent(n=221) tuberculosis patients revealed the role of lncRNA mediated disruption of hypoxia adaptation pathways in progressors.We found 291 upregulated and 227 downregulated lncRNAs that were correlated at mRNA level with HIF1A and HILPDA which are major players in hypoxia response. We also report novel lncRNA-AC010655 (AC010655.4 and AC010655.2) in cis with HILPDA, both of which contain binding sites for the BARX2 transcription factor, thus indicating a mechanistic role. Detailed comparison of infection with antigenic stimulation showed a non-random enrichment of lncRNAs in the cytoplasmic fraction of the cell in progressors. The lack of this pattern in non-progressors indicates the hijacking of the lncRNA dynamics by Mtb. The in-vitro manifestation of this response in the absence of granuloma indicates pre-programmed host-pathogen interaction between T-cells and Mtb regulated through lncRNAs, thus tipping this balance towards progression or containment of Mtb. Finally, we trained multiple machine learning classifiers for reliable prediction of latent to the active progression of patients, yielding a model to guide aggressive treatment.

show abstract

Evaluation of Distance Metrics and Spatial Autocorrelation in Uniform Manifold Approximation and Projection Applied to Mass Spectrometry Imaging Data

Cited by 88 publications

References 18 publications

Dimensionality reduction by UMAP to visualize physical and genetic interactions

Dimensionality reduction by UMAP to visualize physical and genetic interactions

Spatially-Aware Clustering of Ion Images in Mass Spectrometry Imaging Data Using Deep Learning

lncRNA Mediated Hijacking of T-cell Hypoxia Response Pathway byMycobacterium tuberculosisPredicts Latent to Active Progression in Humans

Contact Info

Product

Resources

About