Mark Yandell - US grants

DESCRIPTION (provided by applicant): Medicine, agriculture, and other biology-related industries increasingly depend on the information contained in genomic DNA. Advancing our understanding of how genes are structured and regulated will eventually lead to novel therapeutics for combating cancer and other diseases, to cheaper and more nutritious food, and to less wasteful materials and energy sources. Sequencing genomes is now a relatively straightforward task. Annotating them, however, is proving to be a much more difficult one, especially for eukaryotic genomes. The gene catalogs of all but the simplest eukaryotic genomes are still incomplete, even for well-known organisms with a long history of genetics and molecular biology. Because annotations are the focal point for many kinds of research and technological applications, it is essential that they be correct. Incomplete and incorrect annotations poison every experiment that employs them. We believe the key to improving genome annotation lies in better software for the creation and quality control of genome annotations. This proposal describes our design for a system we call GenomeInvestigator. GenomeInvestigator consists of three components: MAKER, EVALUATOR and VERIFIER. MAKER creates genome annotations, EVALUATOR performs quality control analyses on extant annotations, and VERIFIER automatically designs experiments to problematic portions of annotations. By analogy to genome sequencing, MAKER and EVALUATOR produce draft annotation with quality values, and VERIFIER directs annotation finishing efforts. GenomeInvestigator is designed to be easily portable and will be freely available. Its outputs will be Sequence Ontology compliant and GMOD compatible. Medicine, agriculture, and other biology-related industries increasingly depend on the information contained in genomic DNA. Advancing our understanding of how genes are structured and regulated will eventually lead to novel therapeutics for combating cancer and other diseases, to cheaper and more nutritious food, and to less wasteful materials and energy sources.

DESCRIPTION (provided by applicant): Abstract Current estimates place the number of personal variants at approximately 4 million per genome. Given the rapid advances in genome sequencing technologies and the future democratization of human genome sequencing, small groups and even individual scientists will soon be performing their own human genome projects. We believe that the ability to automatically annotate the millions of variants that these projects will produce, to combine data from multiple projects, and to recover subsets of annotated variants for diverse downstream analyses will become a critical analysis bottleneck. Despite the need, there are no publically available tools that automate these procedures. In response to the NHGRI's RFA "Development and Application of Statistical and Computational Data Analysis Methods for DNA Sequence, Variation, GWAS, Genomic Function, Chemical Biology and Related Genomic Data Sets" we propose in this GO grant to develop a standalone software tool called VAAST-Variant Annotation, Analysis and Selection Tool. This system will fulfill NHGRI's need for a technology to assess data quality and call variants and will allow for analysis of data from all sequencing centers and will be useable for data from all sequencing platforms. We believe VAAST will fill a huge void in the software landscape by helping individual scientists to extract meaningful results from whole genome variant files. PUBLIC HEALTH RELEVANCE: It is now known that on average any two individual human genomes differ by approximately 4 million positions. These differences, called sequence variants, underlie the inherited physical differences between individuals, including their predisposition to develop certain diseases. This project proposes to develop a tool called VAAST- Variation Annotation, Analysis and Selection Tool. VAAST will help researchers sort through these millions of variants in their quest to identify which of them underlie different phenotypic traits of individuals and susceptibility to diseases.

DESCRIPTION (provided by applicant): Whole genome screen for novel regulators of tissue homeostasis and regeneration Schmidtea mediterranea is a model system for addressing human health issues. Planarians are well known for their ability to regenerate complete animals from fragments of their bodies. Following recent demonstrations that S. mediterranea is amenable to modern cell, molecular, and RNAi techniques, it is becoming the model organism of choice for addressing research questions that cannot be easily studied in Drosophila melanogaster or Caenorhabditis elegans, including wound healing of adult tissues, regeneration, somatic stem cells, and tissue homeostasis. In 2007 we annotated the S. mediterranea genome and constructed a publicly available genome database, SmedGD containing the gene models. The objective of this grant is to use the S. mediterranea gene annotations in a high-throughput image-based screen for novel regulators of tissue regeneration and homeostasis. Toward this end we will (1) employ a battery of molecular and immunological techniques, including a genome-wide RNAi screen, and (2) leverage existing image processing, management and annotation tools to construct a cyberinfrastructure that can both support our experiments and distribute our results to the scientific community. PUBLIC HEALTH RELEVANCE: This project will use the sequenced genome of the planarian Schmidtea mediterranea to search for genes involved in tissue maintenance and regeneration. The project employs a number of exciting new technologies, including RNA interference, an automated confocal microscope, and image processing and annotation tools to achieve its aims. Because we will restrict our screen to planarian genes with bona fide human homologs, any stem cell function deficiency phenotype obtained becomes a potential model to study human stem-cell function, regeneration and wound healing, effectively advancing efforts in these frontiers of human health.

PI: Mark Yandell (University of Utah)

CoPIs: Kevin Childs, Ning Jiang, Shin-Han Shiu, and Yanni Sun (Michigan State University)

Today's DNA sequencing technologies are greatly increasing the availability of genome sequences. These genome sequences have great potential to increase agricultural productivity and provide a foundation for modern, genome-based research. Unfortunately a genome sequence in-and-of itself is of little use unless the locations of the genes within it are identified and their parts properly mapped. The process of locating and mapping the structures of genes within a genome sequence is called genome annotation. Researchers are finding that plant genomes are difficult to annotate for several technical reasons; for example, plant genomes tend to be repeat rich and are often polyploid. The purpose of this project is to adapt and extend MAKER, an established annotation tool, for use on plant genomes. Doing so will help to unlock the potential of sequenced plant genomes to empower genomics- based plant research.

The overarching goal of the project is to produce a plant genome optimized annotation pipeline that can be widely used by the plant research community and will be incorporated as part of the iPlant Cyberinfrastructure. The project will also provide outreach to the plant genomics community by means of tutorials given at various scientific meetings in genome annotation on how to use the MAKER-based plant genome optimized annotation tool. These tutorials will allow smaller plant genome projects to annotate their genomes and to distribute the resulting gene models; in turn, this will enable downstream comparative and functional genomics studies by third parties. Source code will be freely available and distributed in publicly announced releases at http://www.yandell-lab.org. Stable versions will be available by anonymous FTP and HTTP as software archives (*.tar.gz files). Finally, developer and candidate releases will be managed via subversion repositories. MAKER download pages can be accessed at http://www.yandell-lab.org/software/index.html.

DESCRIPTION (provided by applicant): Genomics-based venom studies using the cone snail C. bullatus. The marine snail, Conus bullatus is a fish predator and must quickly immobilize its prey. This is accomplished by injecting prey with a lethal cocktail of conopeptide venoms, small cysteine-rich peptides, each with a high affinity to a different ligand or voltage- gated ion channel. Over the last decade, cone venoms have proven indispensable reagents for the study of vertebrate neurotransmission, and the FDA has approved one for the treatment of chronic pain. There is good reason to believe that collectively the cone snails still harbor a large repertoire of uncharacterized venoms (<100,000) of pharmacological interest. Conopeptides also undergo many unusual posttranslational modifications, carried out by enzymes that are themselves of pharmacological interest. This process of venom maturation is poorly characterized. Unfortunately, cone venom research today is hindered by a lack of significant genomic resources. Hence, the purpose this application is (1) to obtain funds for Conus transcriptome and genome sequencing and analyses to identify new venoms and their posttranslational modifiers, and (2) for functional characterizations of these new venoms and the enzymes involved in their posttranslational modifications. .

DESCRIPTION (provided by applicant): The overarching goal of this proposal is to produce a single deliverable: VAAST+, which will provide innovative and improved solutions for three major bottlenecks in analyses of personal genomes data: variant prioritization, risk assessment and disease-gene finding. Better variant prioritization and risk assessment will aid diagnostic laboratories and clinicians seeking to interpret the impact of rare variants discovered in the course of routine genetic testing; whereas a better tool for disease-gene finding will empower researchers seeking to employ whole-genome and exome sequences to identify novel genes and disease-causing alleles responsible for rare and common diseases. VAAST+ will leverage the VAAST platform, which was developed with support from an NHGRI Grand Opportunity Grant entitled Tool for annotation and analyses of human whole-genome sequence variation data. Doing so will allow us to rapidly implement VAAST+ and distribute it to the research community.

PROJECT SUMMARY/ABSTRACT The NHLBI has invested extensively in the Pediatric Cardiac Genomics Consortium (PCGC), recognizing that translating genomic discoveries into optimized management and therapeutic strategies for congenital heart disease (CHD) can only be achieved in the context of multi-center, collaborative research. Currently, the PCGC is lacking two fundamental capabilities that hinder its ability to define the genomic basis for CHD outcomes: (1) a robust mechanism for extracting pertinent, machine-readable clinical data from Electronic Health Records (EHRs) across multiple institutions; and (2) a robust Artificial Intelligence (AI) platform that is capable of teasing apart the complex interplay between maternal factors, phenotypes, genotypes, gene functions and clinical outcomes. Here, we propose innovative solutions to these challenges, by assembling teams of content experts to leverage existing infrastructure to extract relevant outcomes directly from the EHR of participating PCGC Centers and by designing best-practice AI tools for outcomes research. Our principal goal is provide the vision, infrastructure and expertise to collaboratively empower CHD outcomes research, foster knowledge exchange, and train the next generation of genomic scientists. We propose to leverage existing data infrastructure to obtain Electronic Health Records (EHR) and other clinical variables at scale by partnering with other research networks to create a PCGC Data Resource. Using this resource, we will create and deploy a platform of Artificial Intelligence (AI)-based predictors for CHD outcomes research, with the goal of translating genomic discoveries into improved management and therapeutic strategies for CHD.

Advances in next-generation sequencing have opened up additional avenues of investigation due to the reduction in costs and the bioinformatic tools and functional resources available. That said, there is still a critical need for advancing sequence visualization tools to aid in sequence analysis. This EAGER project will develop a new type of genome representation - a Variant Graph which provides an innovative way to visualize and identify variants in genome sequences across related populations within a species. If successfully implemented, a Variant Graph will provide a novel means to leverage next generation sequencing data to create more accurate gene models. In addition, a fully Variant Graph based MAKER-P genome annotation pipeline will provide a leap forwards in genome-annotation and project management by providing a means for simultaneously annotating and managing genome sequences generated from multiple plant cultivars through Gramene (www.gramene.org) and CyVerse (formerly the iPlant Collaborative; www.cyverse.org).

The goals of this project are to take advantage of the wealth of sequence assemblies available for plants to (1) create and deploy a Variant Graph for crop plants using genome sequences available for maize and rice through the Gramene database; (2) leverage MAKER-P to create the first genome-annotation pipeline capable of exploiting the full potential of the Variant Graph to improve the annotations of the maize and rice reference assemblies; and, (3) deploy a cutting-edge Variant Graph based variant-calling pipeline within Gramene. The project will provide interdisciplinary research training in software development and applications for plant biology and genomics for a postdoctoral associate. All software developed as part of this project will be open source and available for academic use.

? DESCRIPTION (provided by applicant): Sudden death in the young (SDY) is a tragic event, with devastating consequences for the family who must endure both the unexpected loss of their child and the possibility of harboring a familial disorder that threatens the health and survival of the remaining members. Due to lack of evidence, there is little consensus within the scientific community around the best method for preventing SDY. Thus, the National Heart, Lung, and Blood Institute (NHLBI), National Institute of Neurologic Disorders and Stroke (NINDS) and Centers for Disease Control and Prevention (CDC) have invested extensively in the SDY Case Registry by acquiring detailed clinical information and archiving DNA samples from SDY cases across the United States, with the goal of defining the genomic and mechanistic basis for SDY. We propose to establish the Utah SDY Center as an integral partner of the SDY Registry, by providing unparalleled bioinformatics, genomics and clinical expertise, resources, and an infrastructure for collaboration to accomplish the Registry's goals. Our proposal addresses several critical barriers in the field of SDY research: 1) standard genetic testing fails to identify a molecular cause in the majority of autopsy-negative SDY; 2) current strategies do not validate the functionality of identified variants; and 3) a comprehensive family cardiac evaluation is inconsistently performed despite Class I expert consensus recommendations. We propose to overcome these barriers by achieving the following Aims: Specific Aim 1: Provide the SDY Case Registry with the bioinformatics expertise that will enable the discovery of the genomic basis for autopsy-negative SDY, using whole-exome/genome sequencing. Specific Aim 2: Characterize novel SDY disease genes, alleles and disease mechanisms, using cell- and model organism-based assays. Specific Aim 3: Integrate our clinical expertise in SDY to phenotype, genotype, risk-stratify and counsel surviving family members. The Utah SDY Center's multi-disciplinary research plan will allow us to discover, validate and characterize new sudden death genes, alleles and mechanisms at a scale and resolution not previously possible. By partnering with other SDY Case Registry Centers, the Utah Center will promote a mechanistic understanding of SDY and thus establish the foundation for future screening strategies and preventative measures.

SUMMARY Venomous marine snails in the superfamily Conoidea capture their prey by injecting a complex mixture of ribosomally-synthesized peptides that undergo extensive post-translational modification. These conopeptides target receptors and ion channels in the prey's nervous, endocrine and sensory system with remarkable potency and specificity. Owing to their diversity and target selectivity, conopeptides have become invaluable tools for ion channel research and as therapeutics. The rationale of using cone snail venoms as a source for drug discovery is that homologs of many molecular targets expressed in the prey of cone snails are also found in humans where they are implicated in diverse physiological disorders, including inflammation, epilepsy, neuropathic pain and diabetes. Several recent discoveries made in my group now demonstrate that each of the ~700 cone snail species produces a distinct set of conopeptides that are finely tuned for a specific set of receptors in its prey. Thus, the central hypothesis of this grant is that drug discovery can be maximized by sequencing and characterizing the venom composition of many species from diverse lineages of cone snails, including those that induce diverse physiological endpoints in their prey. This is a highly innovative approach because it takes full advantage of the unique strategies that evolved in these animals for prey capture: species that induce rapid paralysis in their prey are likely to express toxins that target the neuromuscular junction and pain circuits whereas those that induce hypoactivity and sedation are more likely to have evolved toxins that target the sensory and endocrine system. Our preliminary research has already identified several unique drug leads for the treatment of diabetes, a disease that has been recognized as a global epidemic, and pain, a leading cause for the current opioid epidemic. This proposal will enable us to efficiently scale these promising initial efforts. The specific aims of this project are (Aim 1) to undertake a large-scale, evolution-guided collection and next-generation sequencing effort of venoms from all ~50 major lineages of cone snails, (Aim 2) to develop an innovative computational pipeline, the Taxonomer Venoms Module, to analyze these large sequencing datasets, and (Aim 3) to use a tiered, data-driven selection process to pharmacologically characterize the most promising novel toxins from these large datasets. We will also seek to identify and characterize conopeptide biosynthetic pathways. Doing so will improve synthetic and recombinant means for production of conopeptides for functional studies. The expected outcomes are significant. We will provide a computational pipeline for drug discovery that will lead to the identification of many novel classes of conopeptides and their biosynthetic enzymes that will fuel scientific discovery and drug development activities for decades to come.