Mass Spectrometry Imaging Datasets Research Articles

ELIMS: Ensemble Learning-Based Spatial Segmentation of Mass Spectrometry Imaging to Explore Metabolic Heterogeneity.

Spatial segmentation is an essential processing method for image analysis aiming to identify the characteristic suborgans or microregions from mass spectrometry imaging (MSI) data, which is critical for understanding the spatial heterogeneity of biological information and function and the underlying molecular signatures. Due to the intrinsic characteristics of MSI data including spectral nonlinearity, high-dimensionality, and large data size, the common segmentation methods lack the capability for capturing the accurate microregions associated with biological functions. Here we proposed an ensemble learning-based spatial segmentation strategy, named eLIMS, that combines a randomized unified manifold approximation and projection (r-UMAP) dimensionality reduction module for extracting significant features and an ensemble pixel clustering module for aggregating the clustering maps from r-UMAP. Three MSI datasets are used to evaluate the performance of eLIMS, including mouse fetus, human adenocarcinoma, and mouse brain. Experimental results demonstrate that the proposed method has potential in partitioning the heterogeneous tissues into several subregions associated with anatomical structure, i.e., the suborgans of the brain region in mouse fetus data are identified as dorsal pallium, midbrain, and brainstem. Furthermore, it effectively discovers critical microregions related to physiological and pathological variations offering new insight into metabolic heterogeneity.

Application of clustering strategy for automatic segmentation of tissue regions in mass spectrometry imaging.

Mass spectrometry imaging (MSI) has been widely used in biomedical research fields. Each pixel in MSI consists of a mass spectrum that reflects the molecule feature of the tissue spot. Because MSI contains high-dimensional datasets, it is highly desired to develop computational methods for data mining and constructing tissue segmentation maps. To visualize different tissue regions based on mass spectrum features and improve the efficiency in processing enormous data, we proposed a computational strategy that consists of four procedures including preprocessing, data reduction, clustering, and quantitative validation. In this study, we examined the combination of t-distributed stochastic neighbor embedding (t-SNE) and hierarchical clustering (HC) for MSI data analysis. Using publicly available MSI datasets, one dataset of mouse urinary bladder, and one dataset of human colorectal cancer, we demonstrated that the generated tissue segmentation maps from this combination were superior to other data reduction and clustering algorithms. Using the staining image as a reference, we assessed the performance of clustering algorithms with external and internal clustering validation measures, including purity, adjusted Rand index (ARI), Davies-Bouldin index (DBI), and spatial aggregation index (SAI). The result indicated that SAI delivered excellent performance for automatic segmentation of tissue regions in MSI. We used a clustering algorithm to construct tissue automatic segmentation in MSI datasets. The performance was evaluated by comparing it with the stained image and calculating clustering validation indexes. The results indicated that SAI is important for automatic tissue segmentation in MSI, different from traditional clustering validation measures. Compared to the reports that used internal clustering validation measures such as DBI, our method offers more effective evaluation of clustering results for MSI segmentation. We envision that the proposed automatic image segmentation strategy can facilitate deep learning in molecular feature extraction and biomarker discovery for the biomedical applications of MSI.

Advances in multimodal mass spectrometry for single-cell analysis and imaging enhancement.

Multimodal mass spectrometry (MMS) incorporates an imaging modality with probe-based mass spectrometry (MS) to enable precise, targeted data acquisition and provide additional biological and chemical data not available by MS alone. Two categories of MMS are covered; in the first, an imaging modality guides the MS probe to target individual cells and to reduce acquisition time by automatically defining regions of interest. In the second category, imaging and MS data are coupled in the data analysis pipeline to increase the effective spatial resolution using a higher resolution imaging method, correct for tissue deformation, and incorporate fine morphological features in an MS imaging dataset. Recent methodological and computational developments are covered along with their application to single-cell and imaging analyses.

Extended similarity methods for efficient data mining in imaging mass spectrometry.

Imaging mass spectrometry is a label-free imaging modality that allows for the spatial mapping of many compounds directly in tissues. In an imaging mass spectrometry experiment, a raster of the tissue surface produces a mass spectrum at each sampled x, y position, resulting in thousands of individual mass spectra, each comprising a pixel in the resulting ion images. However, efficient analysis of imaging mass spectrometry datasets can be challenging due to the hyperspectral characteristics of the data. Each spectrum contains several thousand unique compounds at discrete m/z values that result in unique ion images, which demands robust and efficient algorithms for searching, statistical analysis, and visualization. Some traditional post-processing techniques are fundamentally ill-equipped to dissect these types of data. For example, while principal component analysis (PCA) has long served as a useful tool for mining imaging mass spectrometry datasets to identify correlated analytes and biological regions of interest, the interpretation of the PCA scores and loadings can be non-trivial. The loadings often contain negative peaks in the PCA-derived pseudo-spectra, which are difficult to ascribe to underlying tissue biology. Herein, we have utilized extended similarity indices to streamline the interpretation of imaging mass spectrometry data. This novel workflow uses PCA as a pixel-selection method to parse out the most and least correlated pixels, which are then compared using the extended similarity indices. The extended similarity indices complement PCA by removing all non-physical artifacts and streamlining the interpretation of large volumes of imaging mass spectrometry spectra simultaneously. The linear complexity, O(N), of these indices suggests that large imaging mass spectrometry datasets can be analyzed in a 1 : 1 scale of time and space with respect to the size of the input data. The extended similarity indices algorithmic workflow is exemplified here by identifying discrete biological regions of mouse brain tissue.

Multi-MSIProcessor: Data Visualizing and Analysis Software for Spatial Metabolomics Research.

Mass spectrometry imaging (MSI) has emerged as a revolutionary analytical strategy in biomedical research for molecular visualization. By linking the characterization of functional metabolites with tissue architecture, it is now possible to reveal unknown biological functions of tissues. However, due to the complexity and high dimensionality of MSI data, mining bioinformatics-related peaks from batch MSI data sets and achieving complete spatially resolved metabolomics analysis remain a great challenge. Here, we propose novel MSI data processing software, Multi-MSIProcessor (MMP), which integrates the data read-in, MSI visualization, processed data preservation, and biomarker discovery functions. The MMP focuses on the AFADESI-MSI data platform but also supports mzXML and imzmL data input formats for compatibility with data generated by other MSI platforms such as MALDI/SIMS-MSI. MMP enables deep mining of batch MSI data and has flexible adaptability with the source code opened that welcomes new functions and personalized analysis strategies. Using multiple clinical biosamples with complex heterogeneity, we demonstrated that MMP can rapidly establish complete MSI analysis workflows, assess batch sample data quality, screen and annotate differential MS peaks, and obtain abnormal metabolic pathways. MMP provides a novel platform for spatial metabolomics analysis of multiple samples that could meet the diverse analysis requirements of scholars.

High-Throughput Deconvolution of Native Protein Mass Spectrometry Imaging Data Sets for Mass Domain Analysis.

Protein mass spectrometry imaging (MSI) with electrospray-based ambient ionization techniques, such as nanospray desorption electrospray ionization (nano-DESI), generates data sets in which each pixel corresponds to a mass spectrum populated by peaks corresponding to multiply charged protein ions. Importantly, the signal associated with each protein is split among multiple charge states. These peaks can be transformed into the mass domain by spectral deconvolution. When proteins are imaged under native/non-denaturing conditions to retain non-covalent interactions, deconvolution is particularly valuable in helping interpret the data. To improve the acquisition speed, signal-to-noise ratio, and sensitivity, native MSI is usually performed using mass resolving powers that do not provide isotopic resolution, and conventional algorithms for deconvolution of lower-resolution data are not suitable for these large data sets. UniDec was originally developed to enable rapid deconvolution of complex protein mass spectra. Here, we developed an updated feature set harnessing the high-throughput module, MetaUniDec, to deconvolve each pixel of native MSI data sets and transform m/z-domain image files to the mass domain. New tools enable the reading, processing, and output of open format .imzML files for downstream analysis. Transformation of data into the mass domain also provides greater accessibility, with mass information readily interpretable by users of established protein biology tools such as sodium dodecyl sulfate polyacrylamide gel electrophoresis.

A review on recent machine learning applications for imaging mass spectrometry studies

Imaging mass spectrometry (IMS) is a powerful analytical technique widely used in biology, chemistry, and materials science fields that continue to expand. IMS provides a qualitative compositional analysis and spatial mapping with high chemical specificity. The spatial mapping information can be 2D or 3D depending on the analysis technique employed. Due to the combination of complex mass spectra coupled with spatial information, large high-dimensional datasets (hyperspectral) are often produced. Therefore, the use of automated computational methods for an exploratory analysis is highly beneficial. The fast-paced development of artificial intelligence (AI) and machine learning (ML) tools has received significant attention in recent years. These tools, in principle, can enable the unification of data collection and analysis into a single pipeline to make sampling and analysis decisions on the go. There are various ML approaches that have been applied to IMS data over the last decade. In this review, we discuss recent examples of the common unsupervised (principal component analysis, non-negative matrix factorization, k-means clustering, uniform manifold approximation and projection), supervised (random forest, logistic regression, XGboost, support vector machine), and other methods applied to various IMS datasets in the past five years. The information from this review will be useful for specialists from both IMS and ML fields since it summarizes current and representative studies of computational ML-based exploratory methods for IMS.

ISegMSI: An Interactive Strategy to Improve Spatial Segmentation of Mass Spectrometry Imaging Data.

Spatial segmentation is a critical procedure in mass spectrometry imaging (MSI)-based biochemical analysis. However, the commonly used unsupervised MSI segmentation methods may lead to inappropriate segmentation results as the MSI data is characterized by high dimensionality and low signal-to-noise ratio. This process can be improved by the incorporation of precise prior knowledge, which is hard to obtain in most cases. In this study, we show that the incorporation of partial or coarse prior knowledge from different sources such as reference images or biological knowledge may also help to improve MSI segmentation results. Here, we propose a novel interactive segmentation strategy for MSI data called iSegMSI, which incorporates prior information in the form of scribble-regularization of the unsupervised model to fine-tune the segmentation results. By using two typical MSI data sets (including a whole-body mouse fetus and human thyroid cancer), the present results demonstrate the effectiveness of the iSegMSI strategy in improving the MSI segmentations. Specifically, the method can be used to subdivide a region into several subregions specified by the user-defined scribbles or to merge several subregions into a single region. Additionally, these fine-tuned results are highly tolerant to the imprecision of the scribbles. Our results suggest that the proposed iSegMSI method may be an effective preprocessing strategy to facilitate the analysis of MSI data.

Data-Driven Deciphering of Latent Lesions in Heterogeneous Tissue Using Function-Directed t-SNE of Mass Spectrometry Imaging Data.

Mass spectrometry imaging (MSI), which quantifies the underlying chemistry with molecular spatial information in tissue, represents an emerging tool for the functional exploration of pathological progression. Unsupervised machine learning of MSI datasets usually gives an overall interpretation of the metabolic features derived from the abundant ions. However, the features related to the latent lesions are always concealed by the abundant ion features, which hinders precise delineation of the lesions. Herein, we report a data-driven MSI data segmentation approach for recognizing the hidden lesions in the heterogeneous tissue without prior knowledge, which utilizes one-step prediction for feature selection to generate function-specific segmentation maps of the tissue. The performance and robustness of this approach are demonstrated on the MSI datasets of the ischemic rat brain tissues and the human glioma tissue, both possessing different structural complexity and metabolic heterogeneity. Application of the approach to the MSI datasets of the ischemic rat brain tissues reveals the location of the ischemic penumbra, a hidden zone between the ischemic core and the healthy tissue, and instantly discovers the metabolic signatures related to the penumbra. In view of the precise demarcation of latent lesions and the screening of lesion-specific metabolic signatures in tissues, this approach has great potential for in-depth exploration of the metabolic organization of complex tissue.

Hidden molecular clues in marine sediments revealed by untargeted mass spectrometry imaging

Mass spectrometry imaging (MSI) in sedimentary archives can produce records of molecular proxies at μm-scale resolution. For example, in annually varved sediments of the Santa Barbara Basin, such a fine resolution allows deciphering sub-annual distributions of archaeal tetraether lipids, haptophyte-derived alkenones, and sterols. Herein, we reported the establishment of an untargeted data processing workflow aimed at dissecting the MSI datasets and extracting information beyond that obtained by targeted analysis of known molecular proxies. The combination of MSI and the untargeted workflow not only increases the spatial resolution for molecular stratigraphy but also dramatically broadens the number and diversity of molecular signals evaluated, enabling us to discover unique molecular signatures imprinted by various biogeochemical processes. We applied the proposed workflow to two MSI datasets that were both measured on the uppermost ∼10 cm of the Santa Barbara Basin sediments while covering different mass ranges. Two matrices of 18625×293 and 18963×323 (number of spectra × number of peaks) were, respectively, extracted after peak alignment using bin-wise kernel density estimation and subsequent peak picking by peak prominence filtering combined with geochemical context-based filtering. Feature extraction by non-negative matrix factorization revealed in total 15 stable molecular clusters with distinct spatial distributions in the sediments. Each cluster typically comprised several to dozens of compounds, with the majority of compounds in each cluster likely belonging to similar chemical taxonomies. Some of these clusters can be linked to specific biogeochemical processes. For example, chlorin-like compounds are possibly related to diatom production, alkenones are related to coccolithophorid production, and steranes and long-chain fatty acids likely represent terrigenous input. Supervised learning from these data mining results further extracted molecular signatures with proxy potential that appear to be linked to specific environmental conditions inferred from historical oceanographic data. However, generalizability to other sedimentary settings will require further investigation.

Sequential paired covariance for improved visualization of mass spectrometry imaging datasets.

Untargeted analyses in mass spectrometry imaging produce hundreds of ion images representing spatial distributions of biomolecules in biological tissues. Due to the large diversity of ions detected in untargeted analyses, normalization standards are often difficult to implement to account for pixel‐to‐pixel variability in imaging studies. Many normalization strategies exist to account for this variability, but they largely do not improve image quality. In this study, we present a new approach for improving image quality and visualization of tissue features by application of sequential paired covariance (SPC). This approach was demonstrated using previously published tissue datasets such as rat brain and human prostate with different biomolecules like metabolites and N‐linked glycans. Data transformation by SPC improved ion images resulting in increased smoothing of biological features compared with commonly used normalization approaches.

ImShot: An Open-Source Software for Probabilistic Identification of Proteins In Situ and Visualization of Proteomics Data.

Imaging mass spectrometry (IMS) has developed into a powerful tool allowing label-free detection of numerous biomolecules in situ. In contrast to shotgun proteomics, proteins/peptides can be detected directly from biological tissues and correlated to its morphology leading to a gain of crucial clinical information. However, direct identification of the detected molecules is currently challenging for MALDI–IMS, thereby compelling researchers to use complementary techniques and resource intensive experimental setups. Despite these strategies, sufficient information could not be extracted because of lack of an optimum data combination strategy/software. Here, we introduce a new open-source software ImShot that aims at identifying peptides obtained in MALDI–IMS. This is achieved by combining information from IMS and shotgun proteomics (LC–MS) measurements of serial sections of the same tissue. The software takes advantage of a two-group comparison to determine the search space of IMS masses after deisotoping the corresponding spectra. Ambiguity in annotations of IMS peptides is eliminated by introduction of a novel scoring system that identifies the most likely parent protein of a detected peptide in the corresponding IMS dataset. Thanks to its modular structure, the software can also handle LC–MS data separately and display interactive enrichment plots and enriched Gene Ontology terms or cellular pathways. The software has been built as a desktop application with a conveniently designed graphic user interface to provide users with a seamless experience in data analysis. ImShot can run on all the three major desktop operating systems and is freely available under Massachusetts Institute of Technology license.

An Organ-Specific Metabolite Annotation Approach for Ambient Mass Spectrometry Imaging Reveals Spatial Metabolic Alterations of a Whole Mouse Body.

Rapid and accurate metabolite annotation in mass spectrometry imaging (MSI) can improve the efficiency of spatially resolved metabolomics studies and accelerate the discovery of reliable in situ disease biomarkers. To date, metabolite annotation tools in MSI generally utilize isotopic patterns, but high-throughput fragmentation-based identification and biological and technical factors that influence structure elucidation are active challenges. Here, we proposed an organ-specific, metabolite-database-driven approach to facilitate efficient and accurate MSI metabolite annotation. Using data-dependent acquisition (DDA) in liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) to generate high-coverage product ions, we identified 1620 unique metabolites from eight mouse organs (brain, liver, kidney, heart, spleen, lung, muscle, and pancreas) and serum. Following the evaluation of the adduct form difference of metabolite ions between LC-MS and airflow-assisted desorption electrospray ionization (AFADESI)-MSI and deciphering organ-specific metabolites, we constructed a metabolite database for MSI consisting of 27,407 adduct ions. An automated annotation tool, MSIannotator, was then created to conduct metabolite annotation in the MSI dataset with high efficiency and confidence. We applied this approach to profile the spatially resolved landscape of the whole mouse body and discovered that metabolites were distributed across the body in an organ-specific manner, which even spanned different mouse strains. Furthermore, the spatial metabolic alteration in diabetic mice was delineated across different organs, exhibiting that differentially expressed metabolites were mainly located in the liver, brain, and kidney, and the alanine, aspartate, and glutamate metabolism pathway was simultaneously altered in these three organs. This approach not only enables robust metabolite annotation and visualization on a body-wide level but also provides a valuable database resource for underlying organ-specific metabolic mechanisms.

Compound Dereplication and De Novo Characterization of Small Molecules by Mass Spectrometry

We describe the molecular dereplication principles and de novo characterization of small molecules obtained from liquid-chromatography mass spectrometry and imaging mass spectrometry data sets. Our methodology aims at supporting chemists and computer programmers to understand the hidden computing algorithms used for metabolomics mass spectrometry data processing. The approaches have been made available in the open-source tool CycloBranch. The presented tutorial extends the interpretation of mass spectra portfolia described in a series of papers published in Chemicke Listy, issues 2/2020 and 3/2020.

Ratiometric Mass Spectrometry Imaging for Stain-Free Delineation of Ischemic Tissue and Spatial Profiling of Ischemia-Related Molecular Signatures.

Mass spectrometry imaging (MSI) serves as an emerging tool for spatial profiling of metabolic dysfunction in ischemic tissue. Prior to MSI data analysis, commonly used staining methods, e.g., triphenyltetrazole chloride (TTC) staining, need to be implemented on the adjacent tissue for delineating lesion area and evaluating infarction, resulting in extra consumption of the tissue sample as well as morphological mismatch. Here, we propose an in situ ratiometric MSI method for simultaneous demarcation of lesion border and spatial annotation of metabolic and enzymatic signatures in ischemic tissue on identical tissue sections. In this method, the ion abundance ratio of a reactant pair in the TCA cycle, e.g., fumarate to malate, is extracted pixel-by-pixel from an ambient MSI dataset of ischemic tissue and used as a surrogate indicator for metabolic activity of mitochondria to delineate lesion area as if the tissue has been chemically stained. This method is shown to be precise and robust in identifying lesions in brain tissues and tissue samples from different ischemic models including heart, liver, and kidney. Furthermore, the proposed method allows screening and predicting metabolic and enzymatic alterations which are related to mitochondrial dysfunction. Being capable of concurrent lesion identification, in situ metabolomics analysis, and screening of enzymatic alterations, the ratiometric MSI method bears great potential to explore ischemic damages at both metabolic and enzymatic levels in biological research.

Venoms for all occasions: The functional toxin profiles of different anatomical regions in sea anemones are related to their ecological function.

The phylum Cnidaria is the oldest extant venomous group and is defined by the presence of nematocysts, specialized organelles responsible for venom production and delivery. Although toxin peptides and the cells housing nematocysts are distributed across the entire animal, nematocyte and venom profiles have been shown to differ across morphological structures in actiniarians. In this study, we explore the relationship between patterns of toxin expression and the ecological roles of discrete anatomical structures in Telmatactis stephensoni. Specifically, using a combination of proteomic and transcriptomic approaches, we examined whether there is a direct correlation between the functional similarity of regions and the similarity of their associated toxin expression profiles. We report that the regionalization of toxin production is consistent with the partitioning of the ecological roles of venom across envenomating structures, and that three major functional regions are present in T.stephensoni: tentacles, epidermis and gastrodermis. Additionally, we find that most structures that serve similar functions not only have comparable putative toxin profiles but also similar nematocyst types. There was no overlap in the putative toxins identified using proteomics and transcriptomics, but the expression patterns of specific milked venom peptides were conserved across RNA-sequencing and mass spectrometry imaging data sets. Furthermore, based on our data, it appears that acontia of T.stephensoni may be transcriptionally inactive and only mature nematocysts are present in the distal portions of the threads. Overall, we find that the venom profile of different anatomical regions in sea anemones varies according to its ecological functions.

Rapid Automated Annotation and Analysis of N-Glycan Mass Spectrometry Imaging Data Sets Using NGlycDB in METASPACE.

Imaging N-glycan spatial distribution in tissues using mass spectrometry imaging (MSI) is emerging as a promising tool in biological and clinical applications. However, there is currently no high-throughput tool for visualization and molecular annotation of N-glycans in MSI data, which significantly slows down data processing and hampers the applicability of this approach. Here, we present how METASPACE, an open-source cloud engine for molecular annotation of MSI data, can be used to automatically annotate, visualize, analyze, and interpret high-resolution mass spectrometry-based spatial N-glycomics data. METASPACE is an emerging tool in spatial metabolomics, but the lack of compatible glycan databases has limited its application for comprehensive N-glycan annotations from MSI data sets. We created NGlycDB, a public database of N-glycans, by adapting available glycan databases. We demonstrate the applicability of NGlycDB in METASPACE by analyzing MALDI-MSI data from formalin-fixed paraffin-embedded (FFPE) human kidney and mouse lung tissue sections. We added NGlycDB to METASPACE for public use, thus, facilitating applications of MSI in glycobiology.

Peak learning of mass spectrometry imaging data using artificial neural networks

Mass spectrometry imaging (MSI) is an emerging technology that holds potential for improving, biomarker discovery, metabolomics research, pharmaceutical applications and clinical diagnosis. Despite many solutions being developed, the large data size and high dimensional nature of MSI, especially 3D datasets, still pose computational and memory complexities that hinder accurate identification of biologically relevant molecular patterns. Moreover, the subjectivity in the selection of parameters for conventional pre-processing approaches can lead to bias. Therefore, we assess if a probabilistic generative model based on a fully connected variational autoencoder can be used for unsupervised analysis and peak learning of MSI data to uncover hidden structures. The resulting msiPL method learns and visualizes the underlying non-linear spectral manifold, revealing biologically relevant clusters of tissue anatomy in a mouse kidney and tumor heterogeneity in human prostatectomy tissue, colorectal carcinoma, and glioblastoma mouse model, with identification of underlying m/z peaks. The method is applied for the analysis of MSI datasets ranging from 3.3 to 78.9 GB, without prior pre-processing and peak picking, and acquired using different mass spectrometers at different centers.

Mass spectrometry imaging-based multi-modal technique: Next-generation of biochemical analysis strategy

Mass spectrometry imaging-based multi-modal technique: Next-generation of biochemical analysis strategy

Contrast optimization of mass spectrometry imaging (MSI) data visualization by threshold intensity quantization (TrIQ).

Mass spectrometry imaging (MSI) enables the unbiased characterization of surfaces with respect to their chemical composition. In biological MSI, zones with differential mass profiles hint towards localized physiological processes, such as the tissue-specific accumulation of secondary metabolites, or diseases, such as cancer. Thus, the efficient discovery of ‘regions of interest’ (ROI) is of utmost importance in MSI. However, often the discovery of ROIs is hampered by high background noise and artifact signals. Especially in ambient ionization MSI, unmasking biologically relevant information from crude data sets is challenging. Therefore, we implemented a Threshold Intensity Quantization (TrIQ) algorithm for augmenting the contrast in MSI data visualizations. The simple algorithm reduces the impact of extreme values (‘outliers’) and rescales the dynamic range of mass signals. We provide an R script for post-processing MSI data in the imzML community format (https://bitbucket.org/lababi/msi.r) and implemented the TrIQ in our open-source imaging software RmsiGUI (https://bitbucket.org/lababi/rmsigui/). Applying these programs to different biological MSI data sets demonstrated the universal applicability of TrIQ for improving the contrast in the MSI data visualization. We show that TrIQ improves a subsequent detection of ROIs by sectioning. In addition, the adjustment of the dynamic signal intensity range makes MSI data sets comparable.