2015 Midwest Bioinformatics Conference Agenda

Thursday, October 15, 2015


Keith A. Gary, PhD
Vice President
BioNexus KC

Lawrence A. Dreyfus, PhD
Vice Chancellor for Research and Economic Development
University of Missouri-Kansas City

Mark Hoffman, PhD
Director, Center for Health Insights, University of Missouri-Kansas City
UMKC Associate Professor in Biomedical and Health Informatics, and Pediatrics
Director, Translational Bioinformatics, Children’s Mercy

10:30 – 11:30 am: KEYNOTE SPEAKER

Jessica Tenenbaum, PhD
Associate Director, Bioinformatics
Duke Translational Medicine Institute

“Translational Bioinformatics to Enable Precision Medicine: Achievements, Obstacles and Opportunities”

Abstract: The past decade has seen the emergence and expansion of the new discipline of translational bioinformatics (TBI). The field of TBI centers around the development of novel methods to transform increasingly voluminous amounts of molecular and biomedical data into improved human health. Evidence of the rise of TBI can be seen through new journal issues, textbooks, and conferences devoted to the topic. The 2011 NRC “Toward Precision Medicine” Report and the Precision Medicine Initiative announced by President Obama in 2015 have both served to magnify the importance of TBI. The analytic, interpretive, and data storage methods enabled through TBI will be a critical component in realizing the vision for Precision Medicine, in which disease is classified not solely by macroscopic symptoms, many of which have been observed for centuries, but rather based on underlying molecular and environmental causes. This paradigm shift promises to do nothing short of rewrite the textbook of medicine moving forward. It will change the way we approach biomedical research and practice across the physical spectrum, from molecules to populations. As technology continues to advance, assay costs to decrease, and methods are further refined, the next decade is likely to feature increasingly pervasive examples of applied translational bioinformatics, both in healthcare and in daily life. This talk will highlight success stories and outstanding achievements in, or enabled by, translational bioinformatics. It will also address some important caveats and obstacles we face in this rapidly advancing field, as well as some ideas on how to address those hurdles. The talk will conclude with discussion of the tremendous opportunities ahead for TBI to facilitate the practice of precision medicine.

11:30 am – 1:00 pm: LUNCH AND POSTER SESSION

1:00 – 2:00 pm: DATA STRUCTURE

Alignment of data structure with anticipated applications is important at all levels of bioinformatics and has major influence on the success of analysis and other applications.

Andrew Keightley, PhD
Mass Spectrometry and Proteomics Core Facility
School of Biological Sciences
University of Missouri-Kansas City


Gerald J. Wyckoff, PhD
Associate Professor in Molecular Biology & Biochemistry, School of Biological Sciences
University of Missouri-Kansas City

“Structuring Data for Screening Studies in Drug Development”

Abstract: Most in silico screens for drug discovery today focus on one or a small number of targets, and examine potentially hundreds of thousands or millions of potential small molecule binders. Several problems arise when numerous targets are to be examined. First, how do we capture knowledge about target and small molecule similarity in a way that can be easily visualized? Second, how do we structure our experiments to avoid the potential pitfalls of promiscuous binders cropping up throughout our results? Lastly, what’re the data structures we can use to solve these and other potential problems? In this talk we discuss a pilot study looking at a matrix-based approach to in silico screening that enables a wider range of potential experimental scenarios, and discuss the pros and cons of frameworks we used to examine the resulting data.

Dmitriy Shin, PhD, MSc
Assistant Professor, Translational and Cancer Bioinformatics, Department of Pathology and Anatomical Sciences, School of Medicine
Pathobiology Doctoral Faculty, Pathobiology Area PhD Program, Veterinary School
University of Missouri

“Knowledge Representation and Computational Reasoning for Precision Medicine”

Abstract: Effective utilization of “omics” data is crucial to realizing precision medicine ideas. However, majority of existing computational tools for omics data analysis are mostly designed for use by scientists and not effectively tailored for use by medical practitioners. Integrative software tools with greater explanatory power that can lead to actionable outcomes are needed to enable physicians to practice precision medicine. In this talk we will discuss knowledge representation and computational reasoning for high explanatory decision support systems (DSS) that can effectively manage complexity of omics data to bring about actionable outcomes in personalized diagnostics and therapeutics, also known as theranostics. To this end, we have been investigating advanced inference methods to map clinical biomarkers data to biological pathways to recreate interplay of signaling proteomic networks for individualized patient cases. Our new computational formalism called Resource Description Framework (RDF)-induced Influgrams (RIIG) has been shown in a recent proof-of-concept study to exhibit qualities sufficient to provide case-specific reasoning for theranostics. RIIG takes advantage of vast amounts of publicly available curated biological knowledge represented as RDF format and introduces the notions of RDF relevance and RDF collider, which mimic conditional independence and “explaining away” mechanisms of probabilistic systems, respectively. Using these constructs RIIG can uncover complex relationships among biological entities and allow effective management of knowledge granularity, causality and complexity reduction for precision medicine.

Palak Sheth
Product Manager, Bioinformatics
BioNano Genomics

“Whole Genome De Novo Assembly and Structural Variation Discovery with Single Molecule Next Generation Genome Mapping”

Abstract: Structural variation analysis (SVA) of human genomes is usually a reference-based process and therefore biased and incomplete. In order to have a comprehensive analysis of structural variation, a de novo approach is needed. As a result of the remaining limitations of DNA sequencing and analysis technologies, it is not feasible to create high quality de novo assemblies of individuals for detecting and interpreting the many types of structural variation that are refractory to high-throughput or short-read technologies. Using a single-molecule genome analysis system, the Irys® System, we produced high resolution genome maps that were assembled de novo. These maps preserve long-range structural information necessary for structural variation detection.


Integrating disparate data sources, whether for clinical or biological research, requires effective data standardization. Likewise, preparing data for translation from basic research settings into clinical applications requires understanding of the terminologies and standards used in each venue.

Daniel A. Andresen, PhD
Associate Professor, Computing & Information Sciences
Kansas State University


Ian Z. Chuang, MD, MS
Senior Vice President, Healthcare Informatics & Chief Medical Officer
Netsmart Technologies

“Medical Terminology and Codified Data”

Abstract: Clinical data is a full of descriptive data that provides clinical color and details for clinicians to understand what is going on with the care of an individual. Physical medicine has the benefit of discrete quantitative data in terms of measurements from laboratory and measuring instruments to supplement descriptive data. Behavioral health care is heavily dependent on narrative and assessment data. Useful for clinicians reading the notes, yet challenging for digital uses such as decision support, analytics and comparative benchmarking. What does HCIT in behavioral health need to consider to tap the wealth of clinical data for knowledge-driven care?

Devin Koestler, PhD
Assistant Professor, Department of Biostatistics
University of Kansas Medical Center

“An Algorithm for Estimating the Cellular Composition of Blood using DNA Methylation Biomarkers: The Rich Get Richer and the Poor Get Poorer”

Abstract: In this talk, I will discuss ongoing work that involves the development of a statistical methodology for estimating the cell composition of blood; a heterogeneous tissue-type that is frequently used for assessing DNAm. Our approach is similar to regression calibration and involves projecting the DNAm signature blood sample onto a reference data set, which consists of the DNAm signatures across a spectrum isolated white blood cell types. Under certain constraints, this approach can be used to estimate the underlying distribution of cell proportions of the original blood sample, thus providing researchers with tool for controlling for the potential confounding effects of cell heterogeneity in downstream analyses. A colorful and lively overview of epigenetics, DNAm arrary data, and the results of our approach applied several DNAm array data sets will be provided. Analogies and audio/visual examples will be integrated throughout the presentation to help orient the audience to the field of epigenetics and our contributions.

Nitin Tyagi
Vice President Enterprise Solutions
Cambridge Technology Enterprises

“Big Data Analytics for Your Lab”

Abstract:We have seen a quantum leap in the technology for sequencing DNA, which drastically reduced the time and cost of identifying genetic code. The cloud provides various tools and services to help scientific computing customers who are experiencing high levels of data throughput and volume. Molecular modeling, QM MM, protein structure prediction, machine learning approaches, GPU enabled simulations and more can benefit from the elastic and high performance cloud services.

Ability to compare full genomes from individuals, rather than entire species, leading to a much more detailed genetic map of where we, as individuals, have genetic similarities and differences. This will ultimately give us better insight into human health and disease.

This session will provide introduction to: Leveraging cloud for analyzing big data genomic pipelines, high performance computing for Bioinformatics, and AWS Cloud compute platform.

Andrew Keightley, PhD
Mass Spectrometry and Proteomics Core Facility
School of Biological Sciences
University of Missouri-Kansas City

“Correction Factors for Precise Protein Quantitation Using Tandem Mass Tags”

Abstract: New Mass Spectrometer technology with high scan rates (>10/second), provide ultrasensitivity (femtomolar), with single ppm level mass accuracy (+/- 0.001 Da) are available now and facilitating whole cell/tissue simultaneous protein quantitation/identification. Meanwhile, breakthroughs in better sample preparation and much better methods for high resolution peptidome fractionation (new multidimensional liquid chromatography methods) have also been critical to allow routine differential expression analysis of over 8000 proteins from up to 10 separate patients (quantitative 10Plex) in a single experiment. One successful set of experiments can provide enormous amounts of data for biomarker discovery, and pathways analysis to provide the necessary information to recognize clinical pathologies early in disease progression. But to make this data precise, corrections must be integrated in the bioinformatics analysis process to adjust quantitative measurements due to measureable isotopic contaminants in the reagents. We can do that, too.

3:00 – 3:30 pm: BREAK

3:30 – 4:30 pm: DATA VISUALIZATION

Data visualization is increasingly important for hypothesis generation, exploratory research and qualitative analysis. Emerging technologies that generate highly visual representations of data can foster recognition of novel associations.

Susan J. Brown, PhD
University Distinguished Professor of Biology
Kansas State University


Gregory W. King, PhD
Associate Professor, Department of Civil and Mechanical Engineering
University of Missouri-Kansas City

“Kinematic Analysis of Provider Performance in Simulated Clinical Environments”

Abstract: Simulated medical procedures are becoming more viable for clinician training as technology advances. Simulation is a safe and effective way for clinicians to gain proficiency, and can produce favorable effects on patient outcomes. However, many of the performance metrics associated with simulation-based training are qualitative, and few data-driven metrics exist to quantify provider performance. Biomechanical motion analysis technology has the potential to address this limitation, at least from a kinematics perspective. In a recent pilot project, our research team used a marker-based motion analysis system to quantify the kinematic differences between novice and experienced clinicians performing simulated procedures, which included endotracheal intubation, central line placement, and a laparoscopic surgery training exercise. The resulting motion data was processed to extract kinematic performance variables including joint angles and instrument path lengths. Comparison of these metrics between groups provided evidence that experienced clinicians, compared to novice, generally use more efficient movements, characterized by smaller instrument path lengths and joint angle excursions. While the primary focus of this work was on quantitative performance metrics, the motion data may also be used to generate software-based animations and/or 3D renderings for visual analysis. Thus, the combination of visual and quantitative analyses may represent a more effective means to assess and improve provider performance in simulated clinical environments.

Edward Gogol, PhD
Associate Professor, Cell Biology and Biophysics
University of Missouri-Kansas City

“Display of Results of 3D EM Structural Analysis”

Abstract: Visualization of macromolecular structures obtained from electron microscopy of single particles relies on computational displays used for atomic-resolution structures, as obtained mainly by X-ray crystallography or structural NMR. However, in all but a few cases, single-particle EM studies lack resolution to specify atomic coordinates, limiting the information needed to unambiguously identify surfaces and other features of molecules. Furthermore, highly flexible regions of the specimen will not appear at full, if any, density, so limits of interpretation must be considered. Several examples and an outline of approaches will be presented.

Doina Caragea, PhD
Associate Professor
Machine Learning and Bioinformatics Laboratory
Department of Computing and Information Sciences
Kansas State University

“Visual Methods for Examining Machine Learning Classifiers”

Abstract: The availability of large amounts of data in many application domains (e.g., bioinformatics or medical informatics) offers unprecedented opportunities for knowledge discovery in such domains. The classification community in machine learning has focused primarily on building accurate predictive models from the available data. Highly accurate algorithms that can be used for complex classification problems have been designed. Although the predictive accuracy is an important measure of the quality of a classification model, for many data mining tasks, understanding the model is as important as the accuracy of the model itself. Finding the role different variables play in classification provides an analyst with a deeper understanding of the domain. For example, in medical informatics applications, such an understanding can lead to more effective screening, preventive measures and therapies. Visualization of the data in the training stage of building a classifier can help by providing guidance in choosing variables and input parameters for the machine learning algorithms. We illustrate the usefulness of visualization in the context of support vector machines that are used to learn classifiers for discriminating between normal and cancerous cells based on gene expression levels.

Hari S. Hariharan
Vice President, Analytics Consulting
Karvy Analytics Limited

“Facilitating Early Detection of AD Using Machine Learning Techniques”

Abstract: Our focus is to use the power of machine learning techniques to facilitate the early detection of Alzheimer’ diseases (AD). We propose a Computer Aided Diagnosis system to enable a comprehensive analytical method for extracting the significant features of Alzheimer’s disease from MRI images. We used MRI images from the Open Access Series of Imaging Studies (OASIS) database for testing the proposed model and the results are very encouraging.

4:30 – 5:30 pm: SYSTEMS STRATEGIES

Systems analysis requires managing multiple scales and disparate data sources. Strategies for sharing data across non-affiliated systems and for analyzing biological data using biological pathways will be shared.

Anthony Persechini, PhD
Professor, Division of Molecular Biology and Biochemistry
University of Missouri-Kansas City


Russ Waitman, PhD
Director of Medical Informatics
Assistant Vice Chancellor for Enterprise Analytics
Associate Professor of Internal Medicine
University of Kansas Medical Center

“Introducing PCORnet: The National Patient-Centered Clinical Research Network from a Plains Perspective”

Abstract: Will present ongoing efforts by the Patient-Centered Outcomes Research Institute (PCORI) to address means to enhance large scale, real-world–setting, pragmatic trials. The National Patient-Centered Clinical Research Network (PCORNet), a network for conducting clinical outcomes research that will establish a resource of clinical data gathered in real-world settings will be described. PCORNet requires patients, clinicians, and healthcare systems to be actively involved in the governance and use of the data generated by the network. These programs will provide direction, resources, and tools to support a new approach to clinical research.

Douglas C. Bittel, PhD
Associate Professor
The Ward Family Heart Center
Children’s Mercy and University of Missouri-Kansas City

“The Genetic Source of Congenital Heart Defects May Be Hiding in Plain Sight”

Abstract: Congenital heart defects (CHDs) are the most common birth defect and represent a substantial health care burden even in countries with advanced health care systems. In the last 10 years, genetic studies of heart development have made tremendous strides in identifying genetic mechanisms that control vertebrate heart development. Never the less, the genetic contribution to Congenital Heart Defects (CHDs) remains unknown for the majority of CHD Cases. Alternative splicing (AS) plays an important role in regulating mammalian heart development, but a link between misregulated splicing and congenital heart defects (CHDs) has not been shown. We recently examined the transcriptome in myocardial tissue from children with CHDs and observed changes in mRNA splice isoforms of genes that are critical for regulating heart development. In parallel, we found a modest but significant reduction in the levels of 12 small cajal body-specific RNAs (scaRNAs direct the biochemical processing of spliceosomal RNA). To explore the potential relevance of these findings, we used primary cells derived from the RV of infants with TOF to show a direct link between scaRNA levels and splice isoforms of several genes that regulate heart development. In addition, we used antisense morpholinos to knock down the expression of two scaRNAs in zebrafish and saw a corresponding disruption of heart development with an accompanying alteration in splice isoforms of cardiac regulatory genes. Our observations are consistent with disruption of splicing patterns during early embryonic development leading to insufficient regulatory control in the developing heart resulting in the congenital defect. Our observations open the door on a new paradigm in developmental regulation and potentially may explain a substantial portion of the “missing” genetic heritability of CHDs.

Eric J. Deeds, PhD
Assistant Professor, Department of Molecular Biosciences
University of Kansas

“The Evolution of Crosstalk in Signaling Networks”

Abstract: The degree of crosstalk observed in signaling networks varies widely across evolution. In eukaryotes, crosstalk is widespread, with some kinases and phosphatases acting on hundreds of downstream targets. In bacteria, however, signaling pathways are often completely isolated from one another. It is currently unclear exactly what pressures have driven the evolution of these vastly different topologies. The basic “building block” of eukaryotic signaling networks is a pair of enzymes, one that modifies a substrate (e.g. a kinase) and one that undoes this modification (e.g. a phosphatase). We recently used mathematical models to show that adding crosstalk to this type of system can increase ultrasensitivity and couple signal responses, behaviors that could yield phenotypic benefits for eukaryotic cells. In contrast, bacterial networks utilize Two-Component Signaling (TCS), in which a single enzyme (the sensor Histidine Kinase, or HK) acts as both kinase and phosphatase for its downstream Response Regulator (RR). We found that crosstalk always reduces signal response in TCS, which likely underlies the dramatic decrease in fitness that has been observed experimentally when crosstalk is engineered into bacterial cells. Our work thus indicates that the different topologies of eukaryotic and bacterial signaling networks likely arise from fundamental differences in the behavior of the basic motifs from which the networks themselves are constructed. These differences have important consequences for both the function and the evolutionary dynamics of information processing systems within cells.

Christopher P. Baines, PhD
Assistant Professor, Biomedical Sciences, University of Missouri
Investigator, Dalton Cardiovascular Research Center, University of Missouri

“Using Proteomics to Understand Cardiac Myocyte Necrosis”

Abstract: Cardiac injury induces myocyte apoptosis and necrosis, processes that result in the secretion and/or release of intracellular proteins. Currently, myocardial injury can be detected by analysis of a limited number of biomarkers in blood or coronary artery perfusate. However, the complete proteomic signature of protein release from necrotic cardiac myocytes is unknown. Therefore, we undertook an unbiased, proteomic-based study of proteins released from necrotic and apoptotic rat cardiac myocytes to identify novel specific markers of cardiac myocyte cell death. We were able to confirm the presence of classical necrotic markers in the extracellular milieu of necrotic myocytes. We also were able to identify novel markers of necrotic cell death with relatively early release profiles compared to classical protein markers of necrosis. These results have implications for the discovery of novel biomarkers of necrotic myocyte injury, especially in the context of myocardial infarction. In addition, they provide additional insight into the mechanisms by which myocyte necrosis leads to activation of both the inflammatory and fibrotic responses.


Friday, October 16

7:15 am: BREAKFAST

7:45 – 8:00 am: WELCOME AND COMMENTS

Keith A. Gary, PhD
Vice President
BioNexus KC

Mark Hoffman, PhD
Director, Center for Health Insights, University of Missouri-Kansas City
UMKC Associate Professor in Biomedical and Health Informatics, and Pediatrics
Director, Translational Bioinformatics, Children’s Mercy

8:00 – 9:00 am: KEYNOTE SPEAKER

Christopher A. Longhurst, MD, MS
Chief Medical Information Officer, Stanford Children’s Health
Clinical Associate Professor of Pediatrics and Biomedical Informatics,
Stanford University School of Medicine
Clinical Informatics Fellowship Director, Stanford Medicine

“Bringing Big Data to the Bedside”

Abstract: “Big Data” is an overhyped term in biomedicine. In this talk, Dr. Longhurst will share some real-world examples of the value realized by bringing big data to the bedside in both clinical research at Stanford and commercial startups in the Silicon Valley.

9:00 – 10:00 am: DATA ANALYSIS

The computational and statistical toolkit for quantitative analysis of large data sets is developing rapidly, partly through the focus on “Big Data”.

Brooke L. Fridley, PhD
Associate Professor, Department of Biostatistics
Director, Biostatistics and Informatics Shared Resource,
The University of Kansas Cancer Center
Site Director, K-INBRE Bioinformatics Core


Harlen D. Hays, MPH
Senior Manager, Quantitative Research and Biostatistics
Cerner Corporation

“Statistical Analysis in Observational Research: Harnessing EMR Data for Research”

Abstract: Utilization of clinical data for secondary purposes can provide multiple benefits and pitfalls for statistical and informatics research. This talk will focus on the type of available information from Electronic Medical Records and some of the known techniques for ensuring accurate conclusions from the data available. Key topics will include: Connecting to Open Data, mapping implications, exploring available continuous data, and free text.

Trupti Joshi, PhD
Director of Translational Bioinformatics, School of Medicine – Medical Research Office
Assistant Research Professor, Department of Molecular Microbiology & Immunology
Core Faculty MU Informatics Institute and Department of Computer Science
University of Missouri

“Soybean Knowledge Base (SoyKB) and NGS Resequencing Data Analysis for Trait Improvement”

Abstract: Soybean Knowledge Base (http://soykb.org) is a comprehensive web resource developed for bridging soybean translational genomics and molecular breeding research. It provides information for six entities including genes/proteins, microRNAs/sRNAs, metabolites, single nucleotide polymorphisms (SNPs), plant introduction lines and traits. It also incorporates many multi-omics datasets including transcriptomics, proteomics, metabolomics and molecular breeding data, such as quantitative trait loci, traits and germplasm information.

Soybean Knowledge Base has new suite of tools such as In Silico Breeding Program for soybean breeding, which includes a graphical chromosome visualizer for ease of navigation. It integrates quantitative trait loci, traits and germplasm information along with genomic variation data, such as SNPs, insertions, deletions and genome-wide association studies (GWAS) data, from multiple soybean cultivars and Glycine soja. SoyKB also has many new data analysis and visualization tools for RNA-seq and proteomics expression datasets including heatmaps, scatter plots and hierarchical clustering. It also provides new suite of tools for differential expression analysis of omics datasets. Various new types of data including DNA methylation, fast neutron mutations, phosphorylation, genotype by sequencing (GBS) data for molecular breeding and phenotypic inferences have also been incorporated.

SoyKB is now powered by the iPlant Cyber-Infrastructure and connected seamlessly to iPlant’s advanced computing infrastructure, XSEDE and TACC HPC resources, to leverage the data analysis capabilities. The infrastructure is been utilized for bioinformatics analysis of 800+ soybean germplasm lines next generation (NGS) resequencing data using Pegasus workflows. Data is available via SoyKB’s new NGS resequencing data browser tool and being utilized for trait discovery and GWAS analysis.

Mei Liu, PhD
Assistant Professor of Internal Medicine
Division of Medical Informatics
University of Kansas Medical Center

“Predictive Analytics in Drug Safety: Predicting Adverse Drug Reactions”

Abstract: We are spending billions of dollars on prescription drugs every year; however striking variations exist in drug therapy as characterized by adverse drug reactions (ADRs) or drug side-effects. ADR is one of the major causes of failure in the drug development process. Severe ADRs that go undetected until the post-marketing phase of a drug often lead to patient morbidity, causing numerous drug withdrawals. Therefore, accurate prediction of potential ADRs is required in the entire life cycle of a drug, including early stages of drug development, different phases of clinical trials, and post-marketing surveillance. This talk will focus on the methodology development for ADR prediction at various stages.

Hongying (Daisy) Dai, PhD
Senior Biostatistician / Associate Professor, Research Development and Clinical Investigation
Children’s Mercy
Department of Biomedical and Health Informatics
University of Missouri-Kansas City

“Compound Hierarchical Correlated Beta Mixture with an Application to Cluster Mouse Transcription Factor DNA Binding Data”

Abstract: Modeling correlation structures is a challenge in bioinformatics, especially when dealing with high throughput genomic data. A compound hierarchical Correlated Beta Mixture (CBM) with an exchangeable correlation structure is proposed to cluster genetic vectors into mixture components. The correlation coefficient is homogenous within a mixture component and heterogeneous between mixture components. A random CBM brings more flexibility in explaining correlation variations among genetic variables. Expectation-Maximization (EM) algorithm and Stochastic Expectation-Maximization (SEM) algorithm are used to estimate parameters of CBM. The number of mixture components can be determined using model selection criteria such as AIC, BIC and ICL–BIC. Extensive simulation studies were conducted to compare EM, SEM and model selection criteria. Simulation results suggest that CBM outperforms the traditional beta mixture model with lower estimation bias and higher classification accuracy. The proposed method is applied to cluster transcription factor – DNA binding probability in mouse genome data. The results reveal distinct clusters of transcription factors when binding to promoter regions of genes in JAK-STAT, MAPK and other two pathways.

10:00 – 10:30: BREAK

10:30 – 11:30 am: IMAGE ANALYSIS

Images are used to store vast amounts of biological and clinical data. Increasingly powerful informatics methods to detect subtle variations in image data are utilized in a wide variety of disciplines and models.

Chi-Ren Shyu, PhD
Director, University of Missouri Informatics Institute
Chairman, Electrical and Computer Engineering Department
Paul K. and Dianne Shumaker Endowed Professor for Biomedical Informatics
University of Missouri


Tommi A. White, PhD
Assistant Research Professor of Biochemistry
Associate Director, Electron Microscopy Core Research Facility
University of Missouri

“Advances in 3-Dimensional Electron Microscopy”

Abstract: Recent developments in the field of electron microscopy have allowed 3D-visualization versus a traditional 2D image. Methods to capture this third dimension include serial block face imaging, FIBSEM tomography, and cryo-electron microscopy and cryo-electron tomography which utilize direct detection technologies incorporating time as the fourth dimension with the acquisition of “movies”. The image informatics challenges of not only the 2D, but now 3D & 4D electron microscopy are looming. These challenges include storage of large files >25 GB/dataset, quantitating and segmenting of features of interest in a semi- or automated method in noisy images, and rendering the data in 3D.

Liang Tang, PhD
Associate Professor of Biochemistry and Microbiology, Department of Molecular Biosciences
University of Kansas

“In and Out: Viral DNA Packaging and Injection Nanomachines”

Abstract: Many double-stranded DNA (dsDNA) viruses employ common mechanisms for packaging viral DNA into capsid and delivering DNA into host cells. We have integrated electron cryo-microscopy, X-ray crystallography, small-angel X-ray scattering and other biochemical and biophysical methods to investigate the molecular mechanisms underlying DNA packaging into virus particles and DNA injection into host cells in tailed dsDNA bacteriophages and herpesviruses. We have determined three-dimensional structures of the viral DNA-packaging machinery called terminase at atomic resolution for its isolated components and at low resolution for an in vitro assembled complex, shedding lights on intricate mechanisms for maturation of viral DNA prior to packaging and chemo-mechanical coupling during the ATP-driven packaging process. More recently, we have obtained an electron microscopic structure of a ~500-kDa, decameric assembly of the DNA-injection protein gp12 of bacteriophage Sf6, which shows a ~150-Å, mushroom-like architecture consisting of a crown domain and a tube-like domain and embracing a 25-Å-wide channel that precisely accommodates dsDNA. The structure suggests that gp12 mediates injection of phage DNA into host cells by providing a molecular conduit for DNA translocation. The 10-fold symmetry of the gp12 assembly suggests a new symmetry mismatch with respect to the 6-fold symmetric phage tail. The gp12 monomer is highly flexible in solution as shown by small-angle X-ray scattering, suggesting a mechanism for translocation of the protein through the conduit of the phage tail toward the host cell envelope, where it assembles into a DNA-injection device.

Gabriel A. Frank, PhD
Post-doctoral Fellow, Lab of Cell Biology
Center for Cancer Research at the National Institutes of Health

“3D Structure Determination of Conformationally Variable Protein Complexes Using Cryo-EM”

Abstract: Cryo-electron microscopy is unique in its ability to determine the structure of conformationlly variable protein complexes in their native environment as well as in purified isolates. In this talk I will discuss groundbreaking computational strategies for dealing with conformational variability of proteins and their application to our recent work on HIV-Env and human P-glycoprotein.

11:30 am – 12:30 pm: DATA APPLICATION

The successful application of informatics requires understanding of data, technology, and the unique needs of the beneficiary; whether the consumer of food products or a critically ill patient.

Laura Fitzmaurice, MD, FAAP, FACEP
Pediatric Emergency Medicine Physician
Chief Medical Information Officer
Associate Executive Medical Director
Children’s Mercy


Neil Miller
Director of Informatics, Center for Pediatric Genomic Medicine
Children’s Mercy

“The CMH Variant Warehouse – A Catalog of Genetic Variation in Patients of a Children’s Hospital”

Abstract: Advances in high-throughput DNA sequencing have enabled the comprehensive identification of individual genetic variation on an unprecedented scale, powering the diagnosis of disease and personalized treatment. As our ability to detect genetic variation has grown, clinicians and researchers struggle to interpret the functional significance of the millions of variants found in each individual genome. The Variant Warehouse at the Center for Pediatric Genomic Medicine at Children’s Mercy, Kansas City, is a resource containing a record of over 120 million genomic variants detected in more than 3800 patients sequenced by the Center since 2011. Each variant has been characterized by the CPGM’s Rapid Understanding of Nucleotide Effect Software (RUNES) pipeline, which records database cross references and predicted functional consequences as generated by multiple in silico tools. Additionally, a local minor allele frequency is calculated for each variant every 6 hours enabling clinicans and researchers to rapidly identify rare disease causing mutations in patients. This talk will give a brief overview of genomic medicine at Children’s Mercy, Kansas City and the use of high performance computing and Big Data to improve patient treatment.

John A. Spertus, MD, MPH, FACC
Adjunct Professor of Medicine, Washington University School of Medicine, St. Louis, MO
Director, Health Outcomes Research, Saint Luke’s Mid America Heart Institute
Professor, Daniel J. Lauer Missouri Endowed Chair in Metabolism and Vascular Disease Research, University of Missouri-Kansas City

“Applying Precision Medicine Today”

Abstract: While the conference provides an amazing overview of the foundation for leveraging bioinformatics to improve healthcare through the delivery of precision medicine, the proof of this benefit remains ephemeral. In this presentation, Dr. Spertus will introduce a novel IT infrastructure, ePRISM, to enable the implementation of precision medicine within clinical workflow. He will provide empirical evidence of how precision medicine can improve patients’ experiences with care, the safety and outcomes of care and reduce costs. Using angioplasty as a prototypical example of the potential of precision medicine, he will share his experiences in improving the value of healthcare and highlight challenges in translating precision medicine to the bedside, along with an overview of strategies to improve its adoption.

Caroline Emery, PhD
Associate Director of Translational Research
BioMed Valley Discoveries

“Integration of Emerging Clinical Trial Experience to Inform Patient-Tailored Medical Decisions – A Case Study of BVD-523, a Novel Agent in Development for the Treatment of Advanced Malignancies”

Abstract: Precision medicine seeks to identify effective treatments for patients based on genetic, environmental and lifestyle factors. The understanding that cancer is a disease of the genome has led to a revolution in oncology therapeutics. For example, mutations that result in aberrant regulation of the mitogen-activated protein kinase (MAPK) pathway often result in cancer, and thus targeting inhibition of the pathway is an attractive clinical approach. BioMed Valley Discoveries is currently conducting clinical trials to evaluate a novel anti-cancer therapeutic, BVD-523 (ulixertinib), which targets the MAPK pathway. These studies enroll patients with advanced solid tumors or hematological malignancies having genetic alterations known to activate the MAPK pathway. In addition to tumor genetics, response to treatment is influenced by numerous patient-specific considerations (e.g., renal and biliary status, drug absorption, drug metabolism, concomitant medications, etc.). Here we highlight the successes and challenges of integrating large and diverse datasets to inform patient-tailored medical decisions.

12:30 – 1:30 pm: LUNCH

1:30 – 3:00 pm: MOCK INTERVIEWS