Richard M Caprioli

Vanderbilt University, Нашвилл, Michigan, United States

Are you Richard M Caprioli?

Claim your profile

Publications (380)1806.23 Total impact

  • Human pathology 06/2015; DOI:10.1016/j.humpath.2015.06.009 · 2.81 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Motivation: Matrix Assisted Laser Desorption Ionization-Imaging Mass Spectrometry (MALDI-IMS) in “omics” data acquisition generates detailed information about the spatial distribution of molecules in a given biological sample. Various data processing methods have been developed for exploring the resultant high volume data. However, most of these methods process data in the spectral domain and do not make the most of the important spatial information available through this technology. Therefore, we propose a novel streamlined data analysis pipeline specifically developed for MALDI-IMS data utilizing significant spatial information for identifying hidden significant molecular distribution patterns in these complex datasets. Methods: The proposed unsupervised algorithm uses Sliding Window Normalization (SWN) and a new spatial distribution based peak picking method developed based on Gray level Co-Occurrence (GCO) matrices followed by clustering of biomolecules. We also use gist descriptors and an improved version of GCO matrices to extract features from molecular images and minimum medoid distance to automatically estimate the number of possible groups. Results: We evaluated our algorithm using a new MALDI-IMS metabolomics dataset of a plant (Eucalypt) leaf. The algorithm revealed hidden significant molecular distribution patterns in the dataset, which the current Component Analysis and Segmentation Map based approaches failed to extract. We further demonstrate the performance of our peak picking method over other traditional approaches by using a publicly available MALDI-IMS proteomics dataset of a rat brain. Although SWN did not show any significant improvement as compared with using no normalization, the visual assessment showed an improvement as compared to using the median normalization. Availability and implementation: The source code and sample data are freely available at http://exims.sourceforge.net/.
    Bioinformatics 06/2015; DOI:10.1093/bioinformatics/btv356 · 4.62 Impact Factor
  • Journal of Cutaneous Pathology 05/2015; DOI:10.1111/cup.12523 · 1.56 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: MALDI imaging mass spectrometry is a highly sensitive and selective tool used to visualize biomolecules in tissue. However, identification of detected proteins remains a difficult task. Indirect identification strategies have been limited by insufficient mass accuracy to confidently link ion images to proteomics data. Here, we demonstrate the capabilities of MALDI FTICR MS for imaging intact proteins. MALDI FTICR IMS provides an unprecedented combination of mass resolving power (~75,000 at m/z 5000) and accuracy (<5ppm) for proteins up to ~12kDa, enabling identification based on correlation with LC-MS/MS proteomics data. Analysis of rat brain tissue was performed as a proof-of-concept highlighting the capabilities of this approach by imaging and identifying a number of proteins including N-terminally acetylated thymosin β4 (m/z 4,963.502, 0.6ppm) and ATP synthase subunit ε (m/z 5,636.074, -2.3ppm). MALDI FTICR IMS was also used to differentiate a series of oxidation products of S100A8 (m/z 10,164.03, -2.1ppm), a subunit of the heterodimer calprotectin, in kidney tissue from mice infected with Staphylococcus aureus. S100A8 - M37O/C42O3 (m/z 10228.00, -2.6ppm) was found to co-localize with bacterial microcolonies at the center of infectious foci. The ability of MALDI FTICR IMS to distinguish S100A8 modifications is critical to understanding calprotectin's roll in nutritional immunity.
    Journal of the American Society for Mass Spectrometry 04/2015; 26(6). DOI:10.1007/s13361-015-1147-5 · 3.19 Impact Factor
  • Boone M. Prentice, Chad W. Chumbley, Richard M. Caprioli
    [Show abstract] [Hide abstract]
    ABSTRACT: A matrix-assisted laser desorption/ionization time of flight/time of flight tandem mass spectrometer (MALDI TOF/TOF) has been used for high-speed precursor/fragment ion transition image acquisition. High-throughput analysis is facilitated by an Nd:YLF solid state laser capable of pulse repetition rates up to 5 kHz, a high digitizer acquisition rate (up to 50 pixels/s), and continuous laser raster sampling. MS/MS experiments are enabled through the use of a precision timed ion selector, second source acceleration, and a dedicated collision cell. Continuous raster sampling is shown here to facilitate rapid MS/MS ion image acquisition from thin tissue sections for the drug rifampicin and for a common kidney lipid, SM4s(d18:1/24:1). The ability to confirm the structural identity of an analyte as part of the MS/MS imaging experiment is an essential part of the analysis. Additionally, the increase in sensitivity and specificity afforded by an MS/MS approach is highly advantageous, especially when interrogating complex chemical environments such as those in biological tissues. Herein, we report continuous laser raster sampling TOF/TOF imaging methodologies which demonstrate 8 to 14-fold increases in throughput compared with existing MS/MS instrumentation, an important advantage when imaging large areas on tissues. Copyright © 2015 John Wiley & Sons, Ltd.
    Journal of Mass Spectrometry 04/2015; 50(4). DOI:10.1002/jms.3579 · 2.71 Impact Factor
  • Richard M Caprioli
    [Show abstract] [Hide abstract]
    ABSTRACT: Imaging mass spectrometry (IMS) has become a valuable tool for the production of molecular maps in samples ranging from solid inorganic materials to biologicals such as cells and tissues. The unique features of IMS are its ability to map a wide variety of different types of molecules, its superb molecular specificity, and its potential for discovery since no target-specific reagents are needed. IMS has made significant contributions in biology and medicine and promises to be a next generation tool in anatomic pathology.
    Journal of the American Society for Mass Spectrometry 03/2015; 26(6). DOI:10.1007/s13361-015-1108-z · 3.19 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The 2013 Children's Oncology Group (COG) blueprint for renal tumor research challenges investigators to develop new, risk-specific biological therapies for unfavorable histology and higher-risk Wilms tumor (WT) in an effort to close a persistent survival gap and to reduce treatment toxicities. As an initial response to this call from the COG, we used imaging mass spectrometry to determine peptide profiles of WT associated with adverse outcomes. We created a WT tissue microarray containing 2-mm punches of formalin-fixed, paraffin-embedded specimens archived from 48 sequentially treated WT patients at our institutions. Imaging mass spectrometry was performed to compare peptide spectra between three patient groups as follows: unfavorable versus favorable histology, treatment success versus failure, and COG higher- versus lower-risk disease. Statistically significant peptide peaks differentiating groups were identified and incorporated into a predictive model using a genetic algorithm. One hundred thirty-one peptide peaks were differentially expressed in unfavorable versus favorable histology WT (P < 0.05). Two hundred three peaks differentiated treatment failure from success (P < 0.05). Seventy-one peaks differentiated COG higher-risk disease from the very-low, low, and standard-risk groups (P < 0.05). These peaks were used to develop predictive models that could differentiate among patient groups 98.49%, 94.46%, and 98.55% of the time, respectively. Spectral patterns were internally cross-validated using a leave-20% out model. Peptide spectra can discriminate adverse behavior of WT. After future external validation and refinement, these models could be used to predict WT behavior and to stratify intensity of chemotherapy regimens. Furthermore, peptides discovered in the model could be sequenced to identify potential risk-specific drug targets. Copyright © 2015 Elsevier Inc. All rights reserved.
    Journal of Surgical Research 03/2015; 196(2). DOI:10.1016/j.jss.2015.03.020 · 2.12 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Non-small cell lung cancer (NSCLC) is the leading cause of cancer death globally. In order to develop better diagnostics and more effective treatments, research in the past decades has focused on identification of molecular changes in the genome, transcriptome, proteome, and more recently also the metabolome. Phospholipids, which nevertheless play a central role in cell functioning, remain poorly explored. Here, using a mass spectrometry (MS)-based phospholipidomics approach, we profiled 179 phospholipid species in malignant and matched non-malignant lung tissue of 162 NSCLC patients (73 in a discovery cohort and 89 in a validation cohort). We identified 91 phospholipid species that were differentially expressed in cancer versus non-malignant tissues. Most prominent changes included a decrease in sphingomyelins (SMs) and an increase in specific phosphatidylinositols (PIs). Also a decrease in multiple phosphatidylserines (PSs) was observed, along with an increase in several phosphatidylethanolamine (PE) and phosphatidylcholine (PC) species, particularly those with 40 or 42 carbon atoms in both fatty acyl chains together. 2D-imaging MS of the most differentially expressed phospholipids confirmed their differential abundance in cancer cells. We identified lipid markers that can discriminate tumor versus normal tissue and different NSCLC subtypes with an AUC (area under the ROC curve) of 0.999 and 0.885, respectively. In conclusion, using both shotgun and 2D-imaging lipidomics analysis, we uncovered a hitherto unrecognized alteration in phospholipid profiles in NSCLC. These changes may have important biological implications and may have significant potential for biomarker development. This article is protected by copyright. All rights reserved. © 2015 UICC.
    International Journal of Cancer 03/2015; DOI:10.1002/ijc.29517 · 5.01 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We performed high-throughput mass spectrometry at high spatial resolution from individual regions (anterior cingulate and primary motor, somatosensory, and visual cortices) and layers of the neocortex (layers III, IV, and V) and cerebellum (granule cell layer), as well as the caudate nucleus in humans and chimpanzees. A total of 39 mass spectrometry peaks were matched with probable protein identifications in both species, allowing for direct comparison in expression. We explored how the pattern of protein expression varies across regions and cortical layers to provide insights into the differences in molecular phenotype of these neural structures between species. The expression of proteins differed principally in a region- and layer-specific pattern, with more subtle differences between species. Specifically, human and chimpanzee brains were similar in their distribution of proteins related to the regulation of transcription and enzyme activity but differed in their expression of proteins supporting aerobic metabolism. While most work assessing molecular expression differences in the brains of primates has been performed on gene transcripts, this dataset extends current understanding of differential molecular expression that may underlie human cognitive specializations.This article is protected by copyright. All rights reserved. © 2015 Wiley Periodicals, Inc.
    The Journal of Comparative Neurology 03/2015; DOI:10.1002/cne.23777 · 3.51 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Bacterial biofilms account for a significant number of hospital-acquired infections and complicate treatment options, because bacteria within biofilms are generally more tolerant to antibiotic treatment. This resilience is attributed to transient bacterial subpopulations that arise in response to variations in the microenvironment surrounding the biofilm. Here, we probed the spatial proteome of surface-associated single-species biofilms formed by uropathogenic Escherichia coli (UPEC), the major causative agent of community-acquired and catheter-associated urinary tract infections. We used matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) imaging mass spectrometry (IMS) to analyze the spatial proteome of intact biofilms in situ. MALDI-TOF IMS revealed protein species exhibiting distinct localizations within surface-associated UPEC biofilms, including two adhesive fibers critical for UPEC biofilm formation and virulence: type 1 pili (Fim) localized exclusively to the air-exposed region, while curli amyloid fibers localized to the air-liquid interface. Comparison of cells grown aerobically, fermentatively, or utilizing an alternative terminal electron acceptor showed that the phase-variable fim promoter switched to the "OFF" orientation under oxygen-deplete conditions, leading to marked reduction of type 1 pili on the bacterial cell surface. Conversely, S pili whose expression is inversely related to fim expression were up-regulated under anoxic conditions. Tethering the fim promoter in the "ON" orientation in anaerobically grown cells only restored type 1 pili production in the presence of an alternative terminal electron acceptor beyond oxygen. Together these data support the presence of at least two regulatory mechanisms controlling fim expression in response to oxygen availability and may contribute to the stratification of extracellular matrix components within the biofilm. MALDI IMS facilitated the discovery of these mechanisms, and we have demonstrated that this technology can be used to interrogate subpopulations within bacterial biofilms.
    PLoS Pathogens 03/2015; 11(3):e1004697. DOI:10.1371/journal.ppat.1004697 · 8.06 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We describe a predictive imaging modality created by 'fusing' two distinct technologies: imaging mass spectrometry (IMS) and microscopy. IMS-generated molecular maps, rich in chemical information but having coarse spatial resolution, are combined with optical microscopy maps, which have relatively low chemical specificity but high spatial information. The resulting images combine the advantages of both technologies, enabling prediction of a molecular distribution both at high spatial resolution and with high chemical specificity. Multivariate regression is used to model variables in one technology, using variables from the other technology. We demonstrate the potential of image fusion through several applications: (i) 'sharpening' of IMS images, which uses microscopy measurements to predict ion distributions at a spatial resolution that exceeds that of measured ion images by ten times or more; (ii) prediction of ion distributions in tissue areas that were not measured by IMS; and (iii) enrichment of bi
    Nature Methods 02/2015; advance online publication(4). DOI:10.1038/nmeth.3296 · 25.95 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We have achieved protein imaging mass spectrometry capabilities at sub-cellular spatial resolution and at high acquisition speed by integrating a transmission geometry ion source with time of flight mass spectrometry. The transmission geometry principle allowed us to achieve a 1-μm laser spot diameter on target. A minimal raster step size of the instrument was 2.5 μm. Use of 2,5-dihydroxyacetophenone robotically sprayed on top of a tissue sample as a matrix together with additional sample preparation steps resulted in single pixel mass spectra from mouse cerebellum tissue sections having more than 20 peaks in a range 3-22 kDa. Mass spectrometry images were acquired in a standard step raster microprobe mode at 5 pixels/s and in a continuous raster mode at 40 pixels/s.
    Analytical and Bioanalytical Chemistry 02/2015; 407(8). DOI:10.1007/s00216-015-8532-6 · 3.58 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The present study was designed to demonstrate the potential of an optimized histology directed protein identification combined with imaging mass spectrometry technology to reveal and identify molecules associated to ectopic calcification in human tissue. As a proof of concept, mineralized and non-mineralized areas were compared within the same dermal tissue obtained from a patient affected by Pseudoxanthoma elasticum, a genetic disorder characterized by calcification only at specific sites of soft connective tissues. Data have been technically validated on a contralateral dermal tissue from the same subject and compared with those from control healthy skin. Results demonstrate that this approach 1) significantly reduces the effects generated by techniques that, disrupting tissue organization, blend data from affected and unaffected areas; 2) demonstrates that, abolishing differences due to inter-individual variability, mineralized and non-mineralized areas within the same sample have a specific protein profile and have a different distribution of molecules; 3) avoiding the bias of focusing on already known molecules, reveals a number of proteins that have been never related to the disease nor to the calcification process, thus paving the way for the selection of new molecules to be validated as pathogenic or as potential pharmacological targets.
    Bone 01/2015; 24. DOI:10.1016/j.bone.2015.01.004 · 4.46 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Imaging mass spectrometry (IMS) was employed for the analysis of frozen skin biopsies to investigate the differences between stage IV pressure ulcers that remain stalled, stagnant and unhealed versus those exhibiting clinical and histological signs of improvement. Our data reveal a rich diversity of proteins that are dynamically modulated and we selectively highlight a family of calcium binding proteins (S-100 molecules) including calcyclin (S100-A6), calgranulins A (S100-A8) and B (S100-A9), and calgizzarin (S100-A11). IMS allowed us to target 3 discrete regions of interest: the wound bed, adjacent dermis, and hypertrophic epidermis. Plots derived using unsupervised principal component analysis of the global protein signatures within these 3 spatial niches indicate that these data from wound signatures have potential as a prognostic tool since they appear to delineate wounds that are favorably responding to therapeutic interventions versus those that remain stagnant or intractable in their healing status. Our discovery based approach with IMS augments current knowledge of the molecular signatures within pressure ulcers while providing a rationale for a focused examination of the role of calcium modulators within the context of impaired wound healing.
    Journal of Proteome Research 12/2014; 14(2). DOI:10.1021/pr5010218 · 5.00 Impact Factor
  • Domenico Taverna, Jeremy L Norris, Richard M Caprioli
    [Show abstract] [Hide abstract]
    ABSTRACT: This study presents an on-tissue proteolytic digestion and peptide extraction method using microwave irradiation for in situ analysis of proteins from spatially defined regions of a tissue section. The methodology utilizes hydrogel discs (1 mm diameter) embedded with trypsin solution. The hydrogel discs are applied to a tissue section, directing enzymatic digestion to a spatially confined area of the tissue. By applying microwave radiation, protein digestion is performed in 2 minutes on-tissue, and the extracted peptides are then analyzed by MALDI MS and LC-MS/MS. The reliability and reproducibility of the microwave-assisted hydrogel mediated on-tissue digestion is demonstrated by the comparison with other on-tissue digestion strategies, including comparisons with conventional heating and in-solution digestion. LC-MS/MS data were evaluated considering the number of identified proteins as well as the number of protein groups and distinct peptides. The results of this study demonstrate that a rapid and reliable protein digestion can be performed on a single thin tissue section while preserving the tissue architecture, and the resulting peptides can be extracted in sufficient abundance to permit analysis using LC-MS/MS. This approach will be most useful for samples that have limited availability but are needed for multiple analyses, especially for the correlation of proteomics data with histology and immunohistochemistry.
    Analytical Chemistry 11/2014; 87(1). DOI:10.1021/ac503479a · 5.83 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Although physicochemical fractionation techniques play a crucial role in the analysis of complex mixtures, they are not necessarily the best solution to separate specific molecular classes, such as lipids and peptides. Any physical fractionation step such as, for example, those based on liquid chromatography, will introduce its own variation and noise. In this paper we investigate to what extent the high sensitivity and resolution of contemporary mass spectrometers offers viable opportunities for computational separation of signals in full scan spectra. We introduce an automatic method that can discriminate peptide from lipid peaks in full scan mass spectra, based on their isotopic properties. We systematically evaluate which features maximally contribute to a peptide versus lipid classification. The selected features are subsequently used to build a random forest classifier that enables almost perfect separation between lipid and peptide signals without requiring ion fragmentation and classical tandem MS-based identification approaches. The classifier is trained on in silico data, but is also capable of discriminating signals in real world experiments. We evaluate the influence of typical data inaccuracies of common classes of mass spectrometry instruments on the optimal set of discriminant features. Finally, the method is successfully extended towards the classification of individual lipid classes from full scan mass spectral features, based on input data defined by the Lipid Maps Consortium.
    09/2014; 4. DOI:10.1016/j.euprot.2014.05.002
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Imaging mass spectrometry (IMS) has become a prime tool for studying the distribution of biomolecules in tissue. Although IMS data sets can become very large, computational methods have made it practically feasible to search these experiments for relevant findings. However, these methods lack access to an important source of information that many human interpretations rely upon: anatomical insight. In this work, we address this need by (1) integrating a curated anatomical data source with an empirically acquired IMS data source, establishing an algorithm-accessible link between them and (2) demonstrating the potential of such an IMS-anatomical atlas link by applying it toward automated anatomical interpretation of ion distributions in tissue. The concept is demonstrated in mouse brain tissue, using the Allen Mouse Brain Atlas as the curated anatomical data source that is linked to MALDI-based IMS experiments. We first develop a method to spatially map the anatomical atlas to the IMS data sets using nonrigid registration techniques. Once a mapping is established, a second computational method, called correlation-based querying, gives an elementary demonstration of the link by delivering basic insight into relationships between ion images and anatomical structures. Finally, a third algorithm moves further beyond both registration and correlation by providing automated anatomical interpretation of ion images. This task is approached as an optimization problem that deconstructs ion distributions as combinations of known anatomical structures. We demonstrate that establishing a link between an IMS experiment and an anatomical atlas enables automated anatomical annotation, which can serve as an important accelerator both for human and machine-guided exploration of IMS experiments.
    Analytical Chemistry 08/2014; 86(18). DOI:10.1021/ac502838t · 5.83 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Imaging mass spectrometry (IMS) studies increasingly focus on endogenous small molecular weight metabolites and consequently bring special analytical challenges. Since analytical tissue blanks do not exist for endogenous metabolites, careful consideration must be given to confirm molecular identity. Here, we present approaches for the improvement in detection of endogenous amine metabolites such as amino acids and neurotransmitters in tissues through chemical derivatization and matrix-assisted laser desorption/ionization (MALDI) IMS. Chemical derivatization with 4-hydroxy-3-methoxycinnamaldehyde (CA) was used to improve sensitivity and specificity. CA was applied to the tissue via MALDI sample targets precoated with a mixture of derivatization reagent and ferulic acid as a MALDI matrix. Spatial distributions of chemically derivatized endogenous metabolites in tissue were determined by high-mass resolution and MSn IMS. We highlight an analytical strategy for metabolite validation whereby tissue extracts are analyzed by high-performance liquid chromatography (HPLC)-MS/MS to unambiguously identify metabolites and distinguish them from isobaric compounds. Copyright © 2014 John Wiley & Sons, Ltd.
    Journal of Mass Spectrometry 08/2014; 49(8). DOI:10.1002/jms.3411 · 2.71 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Imaging Mass Spectrometry is increasingly being applied to the study of small endogenous compounds, including metabolites, neurotransmitters, lipids and other compounds as well. However, due to the high degree of structural homology and the lack of true "blank" samples, generation of images of unequivocal molecular identity is challenging. In this special feature perspective article, Richard Caprioli and colleagues at Vanderbilt University Medical Center discuss these challenges and describe an analytical strategy that combines a number of advanced instrumental methods to identify and confirm the accurate spatial localization of select amino acids and amine-containing metabolites. By combining derivatization, MS(n) , and accurate mass, followed by confirmation via HPLC-MS, they are able to demonstrate the localization of several endogenous metabolites in biological tissue specimens.
    Journal of Mass Spectrometry 08/2014; 49(8). DOI:10.1002/jms.3273 · 2.71 Impact Factor
  • Jessica L Moore, Richard M Caprioli, Eric P Skaar
    [Show abstract] [Hide abstract]
    ABSTRACT: Matrix-assisted laser desorption/ionization mass spectrometry (MALDI MS) has been successfully applied to the field of microbial pathogenesis with promising results, principally in diagnostic microbiology to rapidly identify bacteria based on the molecular profiles of small cell populations. Direct profiling of molecules from serum and tissue samples by MALDI MS provides a means to study the pathogen-host interaction and to discover potential markers of infection. Systematic molecular profiling across tissue sections represents a new imaging modality, enabling regiospecific molecular measurements to be made in situ, in both two-dimensional and three-dimensional analyses. Herein, we briefly summarize work that employs MALDI MS to study the pathogenesis of microbial infection.
    Current Opinion in Microbiology 07/2014; 19C:45-51. DOI:10.1016/j.mib.2014.05.023 · 7.22 Impact Factor

Publication Stats

15k Citations
1,806.23 Total Impact Points

Institutions

  • 1998–2015
    • Vanderbilt University
      • • Department of Medicine
      • • Department of Biochemistry
      • • Department of Neurological Surgery
      • • Department of Chemistry
      • • Mass Spectrometry Research Center
      Нашвилл, Michigan, United States
  • 2010
    • David H. Murdock Research Institute
      North Carolina, United States
  • 2009
    • American Society for Mass Spectrometry
      Nashville, Tennessee, United States
  • 2007
    • Johns Hopkins University
      Baltimore, Maryland, United States
    • University of Pittsburgh
      Pittsburgh, Pennsylvania, United States
  • 2004–2007
    • Uppsala University
      • Department of Pharmaceutical Biosciences
      Uppsala, Uppsala, Sweden
  • 2006
    • University of Alabama at Birmingham
      Birmingham, Alabama, United States
  • 2005
    • University of Greifswald
      • Institute of Pharmacy
      Griefswald, Mecklenburg-Vorpommern, Germany
  • 2003
    • Gateway-Vanderbilt Cancer Treatment Center
      Clarksville, Tennessee, United States
  • 2001
    • Indian Institute of Technology Ropar
      Rūpar, Punjab, India
  • 1978–1998
    • University of Texas Medical School
      • • Department of Biochemistry and Molecular Biology
      • • Department of Neurobiology and Anatomy
      • • Department of Anesthesiology
      • • Department of Internal Medicine
      Houston, Texas, United States
  • 1987
    • The University of Manchester
      Manchester, England, United Kingdom
  • 1982
    • University of Texas Health Science Center at Houston
      Houston, Texas, United States
  • 1978–1979
    • Purdue University
      ウェストラファイエット, Indiana, United States