Two-Phase and Graph-Based Clustering Methods for Accurate and Efficient Segmentation of Large Mass Spectrometry Images

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

RATIONALE

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.

METHODS

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.

RESULTS

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.

CONCLUSIONS

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.

Spatial segmentation of mass spectrometry imaging data featuring selected principal components

Spatial segmentation aims to find homogeneous/heterogeneous subgroups of spectra or ion images in mass spectrometry imaging (MSI) data. The maps it generated inform researchers of vital characteristics of the data and thus provide the basis for strategizing further biological analysis. Dimensional reduction and clustering are two basic steps of segmentation. Due to the variations in the quality, resolution, density of spectral information, and sizes, not all datasets could be segmented ideally with combinations of different dimensional reduction and clustering algorithms. Here, we proposed a segmentation pipeline that utilized pattern compression by principal component analysis (PCA) and represented by principal components. Instead of preprocessed or raw MSI data, normalized principal components were used for the segmentation process. Multiple datasets of rat brains and mouse kidneys were tested, and the proposed segmentation pipeline presented the obvious advantage of easy-to-use and can be readily intergraded with other existing innovative pipelines.

Delineating regions of interest for mass spectrometry imaging by multimodally corroborated spatial segmentation

Mass spectrometry imaging (MSI), which localizes molecules in a tag-free, spatially resolved manner, is a powerful tool for the understanding of underlying biochemical mechanisms of biological phenomena. When analyzing MSI data, it is essential to delineate regions of interest (ROIs) that correspond to tissue areas of different anatomical or pathological labels. Spatial segmentation, obtained by clustering MSI pixels according to their mass spectral similarities, is a popular approach to automate ROI definition. However, how to select the number of clusters (#Clusters), which determines the granularity of segmentation, remains to be resolved, and an inappropriate #Clusters may lead to ROIs not biologically real. Here we report a multimodal fusion strategy to enable an objective and trustworthy selection of #Clusters by utilizing additional information from corresponding histology images. A deep learning-based algorithm is proposed to extract "histomorphological feature spectra" across an entire hematoxylin and eosin image. Clustering is then similarly performed to produce histology segmentation. Since ROIs originating from instrumental noise or artifacts would not be reproduced cross-modally, the consistency between histology and MSI segmentation becomes an effective measure of the biological validity of the results. So, #Clusters that maximize the consistency is deemed as most probable. We validated our strategy on mouse kidney and renal tumor specimens by producing multimodally corroborated ROIs that agreed excellently with ground truths. Downstream analysis based on the said ROIs revealed lipid molecules highly specific to tissue anatomy or pathology. Our work will greatly facilitate MSI-mediated spatial lipidomics, metabolomics, and proteomics research by providing intelligent software to automatically and reliably generate ROIs.

DiviK: divisive intelligent K-means for hands-free unsupervised clustering in big biological data

BACKGROUND

Investigating molecular heterogeneity provides insights into tumour origin and metabolomics. The increasing amount of data gathered makes manual analyses infeasible-therefore, automated unsupervised learning approaches are utilised for discovering tissue heterogeneity. However, automated analyses require experience setting the algorithms' hyperparameters and expert knowledge about the analysed biological processes. Moreover, feature engineering is needed to obtain valuable results because of the numerous features measured.

RESULTS

We propose DiviK: a scalable stepwise algorithm with local data-driven feature space adaptation for segmenting high-dimensional datasets. The algorithm is compared to the optional solutions (regular k-means, spatial and spectral approaches) combined with different feature engineering techniques (None, PCA, EXIMS, UMAP, Neural Ions). Three quality indices: Dice Index, Rand Index and EXIMS score, focusing on the overall composition of the clustering, coverage of the tumour region and spatial cluster consistency, are used to assess the quality of unsupervised analyses. Algorithms were validated on mass spectrometry imaging (MSI) datasets-2D human cancer tissue samples and 3D mouse kidney images. DiviK algorithm performed the best among the four clustering algorithms compared (overall quality score 1.24, 0.58 and 162 for d(0, 0, 0), d(1, 1, 1) and the sum of ranks, respectively), with spectral clustering being mostly second. Feature engineering techniques impact the overall clustering results less than the algorithms themselves (partial [Formula: see text] effect size: 0.141 versus 0.345, Kendall's concordance index: 0.424 versus 0.138 for d(0, 0, 0)).

CONCLUSIONS

DiviK could be the default choice in the exploration of MSI data. Thanks to its unique, GMM-based local optimisation of the feature space and deglomerative schema, DiviK results do not strongly depend on the feature engineering technique applied and can reveal the hidden structure in a tissue sample. Additionally, DiviK shows high scalability, and it can process at once the big omics data with more than 1.5 mln instances and a few thousand features. Finally, due to its simplicity, DiviK is easily generalisable to an even more flexible framework. Therefore, it is helpful for other -omics data (as single cell spatial transcriptomic) or tabular data in general (including medical images after appropriate embedding). A generic implementation is freely available under Apache 2.0 license at https://github.com/gmrukwa/divik .

Next Generation Digital Pathology: Emerging Trends and Measurement Challenges for Molecular Pathology

Digital pathology is revolutionising the analysis of histological features and is becoming more and more widespread in both the clinic and research. Molecular pathology extends the tissue morphology information provided by conventional histopathology by providing spatially resolved molecular information to complement the structural information provided by histopathology. The multidimensional nature of the molecular data poses significant challenge for data processing, mining, and analysis. One of the key challenges faced by new and existing pathology practitioners is how to choose the most suitable molecular pathology technique for a given diagnosis. By providing a comparison of different methods, this narrative review aims to introduce the field of molecular pathology, providing a high-level overview of many different methods. Since each pixel of an image contains a wealth of molecular information, data processing in molecular pathology is more complex. The key data processing steps and variables, and their effect on the data, are also discussed.

Automatic Registration of the Mass Spectrometry Imaging Data of Sagittal Brain Slices to the Reference Atlas

The registration of the mass spectrometry imaging (MSI) data with mouse brain tissue slices from the atlases could perform automatic anatomical interpretation, and the comparison of MSI data in particular brain regions from different mice could be accelerated. However, the current registration of MSI data with mouse brain tissue slices is mainly focused on the coronal. Although the sagittal plane is able to provide more information about brain regions on a single histological slice than the coronal, it is difficult to directly register the complete sagittal brain slices of a mouse as a result of the more significant individualized differences and more positional shifts of brain regions. Herein, by adding the auxiliary line on the two brain regions of central canal (CC) and cerebral peduncle (CP), the registration accuracy of the MSI data with sagittal brain slices has been improved (∼2-5-folds for different brain regions). Moreover, the histological sections with different degrees deformation and different dyeing effects have been used to verify that this pipeline has a certain universality. Our method facilitates the rapid comparison of sagittal plane MSI data from different animals and accelerates the application in the discovery of disease markers.

Data Filtering and Its Prioritization in Pipelines for Spatial Segmentation of Mass Spectrometry Imaging

Mass spectrometry imaging (MSI) could provide vast amounts of data at the temporal-spatial scale in heterogeneous biological specimens, which challenges us to segment accurately suborgans/microregions from complex MSI data. Several pipelines had been proposed for MSI spatial segmentation in the past decade. More importantly, data filtering was found to be an efficient procedure to improve the outcomes of MSI segmentation pipelines. It is not clear, however, how the filtering procedure affects the MSI segmentation. An improved pipeline was established by elaborating the filtering prioritization and filtering algorithm. Lipidomic-characteristic-based MSI data of a whole-body mouse fetus was used to evaluate the established pipeline on localization of the physiological position of suborgans by comparing with three commonly used pipelines and commercial SCiLS Lab software. Two structural measurements were used to quantify the performances of the pipelines including the percentage of abnormal edge pixel (PAEP) and CHAOS. Our results demonstrated that the established pipeline outperformed the other pipelines in visual inspection, spatial consistence, time-cost, and robustness analysis. For example, the dorsal pallium (isocortex) and hippocampal formation (Hpf) regions, midbrain, cerebellum, and brainstem on the mouse brain were annotated and located by the established pipeline. As a generic pipeline, the established pipeline could help with the accurate assessment and screening of drug/chemical-induced targeted organs and exploration of the progression and molecular mechanisms of diseases. The filter-based strategy is expected to become a critical component in the standard operating procedure of MSI data sets.

Fast visual exploration of mass spectrometry images with interactive dynamic spectral similarity pseudocoloring

Mass Spectrometry Imaging (MSI) is an established and still evolving technique for the spatial analysis of molecular co-location in biological samples. Nowadays, MSI is expanding into new domains such as clinical pathology. In order to increase the value of MSI data, software for visual analysis is required that is intuitive and technique independent. Here, we present QUIMBI (QUIck exploration tool for Multivariate BioImages) a new tool for the visual analysis of MSI data. QUIMBI is an interactive visual exploration tool that provides the user with a convenient and straightforward visual exploration of morphological and spectral features of MSI data. To improve the overall quality of MSI data by reducing non-tissue specific signals and to ensure optimal compatibility with QUIMBI, the tool is combined with the new pre-processing tool ProViM (Processing for Visualization and multivariate analysis of MSI Data), presented in this work. The features of the proposed visual analysis approach for MSI data analysis are demonstrated with two use cases. The results show that the use of ProViM and QUIMBI not only provides a new fast and intuitive visual analysis, but also allows the detection of new co-location patterns in MSI data that are difficult to find with other methods.

Deep Learning-Based Annotation Transfer between Molecular Imaging Modalities: An Automated Workflow for Multimodal Data Integration

An ever-increasing array of imaging technologies are being used in the study of complex biological samples, each of which provides complementary, occasionally overlapping information at different length scales and spatial resolutions. It is important to understand the information provided by one technique in the context of the other to achieve a more holistic overview of such complex samples. One way to achieve this is to use annotations from one modality to investigate additional modalities. For microscopy-based techniques, these annotations could be manually generated using digital pathology software or automatically generated by machine learning (including deep learning) methods. Here, we present a generic method for using annotations from one microscopy modality to extract information from complementary modalities. We also present a fast, general, multimodal registration workflow [evaluated on multiple mass spectrometry imaging (MSI) modalities, matrix-assisted laser desorption/ionization, desorption electrospray ionization, and rapid evaporative ionization mass spectrometry] for automatic alignment of complex data sets, demonstrating an order of magnitude speed-up compared to previously published work. To demonstrate the power of the annotation transfer and multimodal registration workflows, we combine MSI, histological staining (such as hematoxylin and eosin), and deep learning (automatic annotation of histology images) to investigate a pancreatic cancer mouse model. Neoplastic pancreatic tissue regions, which were histologically indistinguishable from one another, were observed to be metabolically different. We demonstrate the use of the proposed methods to better understand tumor heterogeneity and the tumor microenvironment by transferring machine learning results freely between the two modalities.

Heterogeneous antiretroviral drug distribution and HIV/SHIV detection in the gut of three species

HIV replication within tissues may increase in response to a reduced exposure to antiretroviral drugs. Traditional approaches to measuring drug concentrations in tissues are unable to characterize a heterogeneous drug distribution. Here, we used mass spectrometry imaging (MSI) to visualize the distribution of six HIV antiretroviral drugs in gut tissue sections from three species (two strains of humanized mice, macaques, and humans). We measured drug concentrations in proximity to CD3+ T cells that are targeted by HIV, as well as expression of HIV or SHIV RNA and expression of the MDR1 drug efflux transporter in gut tissue from HIV-infected humanized mice, SHIV-infected macaques, and HIV-infected humans treated with combination antiretroviral drug therapy. Serial 10-μm sections of snap-frozen ileal and rectal tissue were analyzed by MSI for CD3+ T cells and MDR1 efflux transporter expression by immunofluorescence and immunohistochemistry, respectively. The tissue slices were analyzed for HIV/SHIV RNA expression by in situ hybridization and for antiretroviral drug concentrations by liquid chromatography-mass spectrometry. The gastrointestinal tissue distribution of the six drugs was heterogeneous. Fifty percent to 60% of CD3+ T cells did not colocalize with detectable drug concentrations in the gut tissue. In all three species, up to 90% of HIV/SHIV RNA was found to be expressed in gut tissue with no exposure to drug. These data suggest that there may be gut regions with little to no exposure to antiretroviral drugs, which may result in low-level HIV replication contributing to HIV persistence.

Mass Spectrometry Imaging and Integration with Other Imaging Modalities for Greater Molecular Understanding of Biological Tissues

Over the last two decades, mass spectrometry imaging (MSI) has been increasingly employed to investigate the spatial distribution of a wide variety of molecules in complex biological samples. MSI has demonstrated its potential in numerous applications from drug discovery, disease state evaluation through proteomic and/or metabolomic studies. Significant technological and methodological advancements have addressed natural limitations of the techniques, i.e., increased spatial resolution, increased detection sensitivity especially for large molecules, higher throughput analysis and data management. One of the next major evolutions of MSI is linked to the introduction of imaging mass cytometry (IMC). IMC is a multiplexed method for tissue phenotyping, imaging signalling pathway or cell marker assessment, at sub-cellular resolution (1 μm). It uses MSI to simultaneously detect and quantify up to 30 different antibodies within a tissue section. The combination of MSI with other molecular imaging techniques can also provide highly relevant complementary information to explore new scientific fields. Traditionally, classical histology (especially haematoxylin and eosin–stained sections) is overlaid with molecular profiles obtained by MSI. Thus, MSI-based molecular histology provides a snapshot of a tissue microenvironment and enables the correlation of drugs, metabolites, lipids, peptides or proteins with histological/pathological features or tissue substructures. Recently, many examples combining MSI with other imaging modalities such as fluorescence, confocal Raman spectroscopy and MRI have emerged. For instance, brain pathophysiology has been studied using both MRI and MSI, establishing correlations between in and ex vivo molecular imaging techniques. Endogenous metabolite and small peptide modulation were evaluated depending on disease state. Here, we review advanced ‘hot topics’ in MSI development and explore the combination of MSI with established molecular imaging techniques to improve our understanding of biological and pathophysiological processes.

Exploring Ion Suppression in Mass Spectrometry Imaging of a Heterogeneous Tissue

In this study we have explored several aspects of regional analyte suppression in mass spectrometry imaging (MSI) of a heterogeneous sample, transverse cryosections of mouse brain. Olanzapine was homogeneously coated across the section prior to desorption electrospray ionization (DESI) and matrix-assisted laser desorption ionization (MALDI) mass spectrometry imaging. We employed the concept of a tissue extinction coefficient (TEC) to assess suppression of an analyte on tissue relative to its intensity in an off tissue region. We expanded the use of TEC, by first segmenting anatomical regions using graph-cuts clustering and calculating a TEC for each cluster. The single ion image of the olanzapine [M + H]⁺ ion was seen to vary considerably across the image, with anatomical features such as the white matter and hippocampus visible. While trends in regional ion suppression were conserved across MSI modalities, significant changes in the magnitude of relative regional suppression effects between techniques were seen. Notably the intensity of olanzapine was less suppressed in DESI than for MALDI. In MALDI MSI, significant differences in the concentration dependence of regional TECs were seen, with the TEC of white matter clusters exhibiting a notably stronger correlation with concentration than for clusters associated with gray matter regions. We further employed cluster-specific TECs as regional normalization factors. In comparison to published pixel-by-pixel normalization methods, regional TEC normalization exhibited superior reduction ion suppression artifacts. We also considered the usefulness of a segmentation-based approach to compare spectral information obtained from complementary modalities.

Interactive Visual Exploration of 3D Mass Spectrometry Imaging Data Using Hierarchical Stochastic Neighbor Embedding Reveals Spatiomolecular Structures at Full Data Resolution

Technological advances in mass spectrometry imaging (MSI) have contributed to growing interest in 3D MSI. However, the large size of 3D MSI data sets has made their efficient analysis and visualization and the identification of informative molecular patterns computationally challenging. Hierarchical stochastic neighbor embedding (HSNE), a nonlinear dimensionality reduction technique that aims at finding hierarchical and multiscale representations of large data sets, is a recent development that enables the analysis of millions of data points, with manageable time and memory complexities. We demonstrate that HSNE can be used to analyze large 3D MSI data sets at full mass spectral and spatial resolution. To benchmark the technique as well as demonstrate its broad applicability, we have analyzed a number of publicly available 3D MSI data sets, recorded from various biological systems and spanning different mass-spectrometry ionization techniques. We demonstrate that HSNE is able to rapidly identify regions of interest within these large high-dimensionality data sets as well as aid the identification of molecular ions that characterize these regions of interest; furthermore, through clearly separating measurement artifacts, the HSNE analysis exhibits a degree of robustness to measurement batch effects, spatially correlated noise, and mass spectral misalignment.