Biology subjects

Patterson, N. H.

Publications and source records attributed to Patterson, N. H..

5 recordsLinked to original sources

bioRxiv2021

Autofluorescence microscopy as a label-free tool for renal histology and glomerular segmentation

Automated spatial segmentation models can enrich spatio-molecular omics analyses by providing a link to relevant biological structures. We developed segmentation models that use label-free autofluorescence (AF) microscopy to recognize multicellular functional tissue units (FTUs) (glomerulus, proximal tubule, descending thin limb, ascending thick limb, distal tubule, and collecting duct) and gross morphological structures (cortex, outer medulla, and inner medulla) in the human kidney. Annotations were curated using highly specific multiplex immunofluorescence and transferred to co-registered AF for model training. All FTUs (except the descending thin limb) and gross kidney morphology were segmented with high accuracy: >0.85 F1-score, and Dice-Sorensen coefficients >0.80, respectively. This workflow allowed lipids, profiled by imaging mass spectrometry, to be quantitatively associated with segmented FTUs. The segmentation masks were also used to acquire spatial transcriptomics data from collecting ducts. Consistent with previous literature, we demonstrated differing transcript expression of collecting ducts in the inner and outer medulla.

bioinformatics↗

bioRxiv2021

Tissue fixation effects on human retinal lipid analysis by MALDI imaging and LC-MS/MS technologies

Imaging mass spectrometry (IMS) allows the location and abundance of lipids to be mapped across tissue sections of human retina. For reproducible and accurate information, sample preparation methods need to be optimized. Paraformaldehyde fixation of a delicate multilayer structure like human retina facilitates the preservation of tissue morphology by forming methylene bridge cross-links between formaldehyde and amine/ thiols in biomolecules; however, retina sections analyzed by IMS are typically fresh-frozen. To determine if clinically significant inferences could be reliably based on fixed tissue, we evaluated the effect of fixation on analyte detection, spatial localization, and introduction of artefactual signals. Hence, we assessed the molecular identity of lipids generated by matrix-assisted laser desorption ionization (MALDI-IMS) and liquid chromatography coupled tandem mass spectrometry (LC-MS/MS) for fixed and fresh-frozen retina tissues in positive and negative ion modes. Based on MALDI-IMS analysis, more lipid signals were observed in fixed compared to fresh-frozen retina. More potassium adducts were observed in fresh-frozen tissues than fixed as the fixation process caused displacement of potassium adducts to protonated and sodiated species in ion positive ion mode. LC-MS/MS analysis revealed an overall decrease in lipid signals due to fixation that reduced glycerophospholipids and glycerolipids and conserved most sphingolipids and cholesteryl esters. The high quality and reproducible information from untargeted lipidomics analysis of fixed retina informs on all major lipid classes, similar to fresh-frozen retina, and serves as a steppingstone towards understanding of lipid alterations in retinal diseases.

biochemistry↗

bioRxiv2020

Automated Biomarker Candidate Discovery in Imaging Mass Spectrometry Data Through Spatially Localized Shapley Additive Explanations

The search for molecular species that are differentially expressed between biological states is an important step towards discovering promising biomarker candidates. In imaging mass spectrometry (IMS), performing this search manually is often impractical due to the large size and high-dimensionality of IMS datasets. Instead, we propose an interpretable machine learning workflow that automatically identifies biomarker candidates by their mass-to-charge ratios, and that quantitatively estimates their relevance to recognizing a given biological class using Shapley additive explanations (SHAP). The task of biomarker candidate discovery is translated into a feature ranking problem: given a classification model that assigns pixels to different biological classes on the basis of their mass spectra, the molecular species that the model uses as features are ranked in descending order of relative predictive importance such that the top-ranking features have a higher likelihood of being useful biomarkers. Besides providing the user with an experiment-wide measure of a molecular species biomarker potential, our workflow delivers spatially localized explanations of the classification models decision-making process in the form of a novel representation called SHAP maps. SHAP maps deliver insight into the spatial specificity of biomarker candidates by highlighting in which regions of the tissue sample each feature provides discriminative information and in which regions it does not. SHAP maps also enable one to determine whether the relationship between a biomarker candidate and a biological state of interest is correlative or anticorrelative. Our automated approach to estimating a molecular species potential for characterizing a user-provided biological class, combined with the untargeted and multiplexed nature of IMS, allows for the rapid screening of thousands of molecular species and the obtention of a broader biomarker candidate shortlist than would be possible through targeted manual assessment. Our biomarker candidate discovery workflow is demonstrated on mouse-pup and rat kidney case studies. HighlightsO_LIOur workflow automates the discovery of biomarker candidates in imaging mass spectrometry data by using state-of-the-art machine learning methodology to produce a shortlist of molecular species that are differentially expressed with regards to a user-provided biological class. C_LIO_LIA model interpretability method called Shapley additive explanations (SHAP), with observational Shapley values, enables us to quantify the local and global predictive importance of molecular species with respect to recognizing a user-provided biological class. C_LIO_LIBy providing spatially localized explanations for a classification models decision-making process, SHAP maps deliver insight into the spatial specificity of biomarker candidates and enable one to determine whether (and where) the relationship between a biomarker candidate and the class of interest is correlative or anticorrelative. C_LI

bioinformatics↗

bioRxiv2020

Highly Multiplexed Immunofluorescence of the Human Kidney using Co-Detection by Indexing (CODEX)

The human kidney is composed of many cell types that vary in their abundance and distribution from organ to organ. As these cell types perform unique and essential functions, it is important to confidently label each within a single tissue to more accurately assess tissue architecture. Towards this goal, we demonstrate the use of co-detection by indexing (CODEX) multiplexed immunofluorescence for visualizing 23 antigens within the human kidney. Using CODEX, many of the major cell types and substructures, such as collecting ducts, glomeruli, and thick ascending limb, were visualized within a single tissue section. Of these antibodies, 19 were conjugated in-house, demonstrating the flexibility and utility of this approach for studying the human kidney using traditional antibody markers. We performed a pilot study showing that the studied tissues had on average 84 {+/-} 11 cells per mm2 with the most variance seen within the cells containing vimentin and aquaporin 1, while cells containing -smooth muscle actin and CD31 possessed a high degree of uniformity between the samples. These precursory data show the power of CODEX multiplexed IF for surveying the cellular diversity of the human kidney and have potential applications within pathology, histology, and building anatomical atlases.

cell biology↗

bioRxiv2020

Lipid Landscape of the Human Retina and Supporting Tissues Revealed by High Resolution Imaging Mass Spectrometry

The human retina evolved to facilitate complex visual tasks. It supports vision at light levels ranging from starlight to sunlight, and its supporting tissues and vasculature regulate plasma-delivered lipophilic essentials for vision, including retinoids (vitamin A derivatives). The human retina is of particular interest because of its unique anatomic specializations for high-acuity and color vision that are also vulnerable to prevalent blinding diseases. The retinas exquisite cellular architecture is composed of numerous cell types that are aligned horizontally, giving rise to structurally distinct cell, synaptic, and vascular layers that are visible in histology and in diagnostic clinical imaging. Suitable for retinal investigations, MALDI imaging mass spectrometry (IMS) technologies are now capable of providing images at low micrometer spatial resolution with high levels of chemical specificity. In this study, a multimodal imaging approach combined with a recently developed method of high accuracy multi-image registration was used to define the localization of lipids in human retina tissue at laminar, cellular, and sub-cellular levels. Data acquired by IMS combined with autofluorescence and bright-field microscopy of human retina sections in macular and peripheral regions indicate differences in distributions and abundances of lipid species across and within single cell types. Of note is localization of signals within specific layers of macula, localization within different compartments of photoreceptors and RPE, complementarity of signals between macular retina and non-macular RPE, and evidence that lipids differing by a single double bond can have markedly different distributions.

biochemistry↗