Maja Bucan - US grants

The long term objective of this proposal is to establish the physical map and characterize the molecular structure of the region surrounding the Ph, Rw and W loci on mouse chromosome 5. These three genes are located within 3 cM and are defined by dominant mutations that have similar effects on pigmentation and development. Alleles of two mutations (W, Ph) are associated with mutations/deletions of genes encoding growth factor receptors. Based on the way the third mutation, Rw, was generated it is likely that this mutation also represents a deletion. Chromosome walking using yeast artificial chromosome (YAC) and jumping libraries will be used to isolate overlapping sets of genomic clones (contigs) around available molecular probes. In order to determine the order, physical distance and overlap between contigs, their end clones will be mapped genetically and by pulsed field gel electrophoresis (PFGE). The PFGE analysis of end clones in W19H, Ph-and Rw DNA will facilitate physical mapping, and will also allow us to identify the breakpoints and determine the size of the chromosomal aberrations associated with theW19H, Ph and Rw mutations. A physical map of the region around Ph, Rw and W will contribute to our understanding of the chromosomal organization around these phenotypically related, developmentally interesting genes. In addition this study win provide a model of a combined genetic and molecular analysis of a delimited region of the mouse genome and will provide useful gene-probes for the homologous region on human chromosome 4.

The long term objective of these studies is to decode the genetic information contained within a relatively small chromosomal region in the central portion of mouse chromosome 5. Classical genetic studies have resulted in the identification of several developmental mutations (W, Ph, Rw, rs, l, pi) within this region. Recent studies have shown that two of these mutations (Ph and W) are caused by alterations in genes encoding receptor tyrosine kinases. The extensive information about the genetic structure of this region, and its involvement in key developmental processes, provide a rationale for the establishment of a more comprehensive functional map. The experiments proposed here will provide both a structural map consisting of cloned and characterized mutation breakpoints associated with chromosomal rearrangement and a transcription map consisting of an array of partially characterized physically mapped exons/genes. To define the functional role of the genes in this region, a hemizygous screen for novel visible / lethal / behavioral mutations among the progeny of chemically mutagenized mice will also be performed. Complementation analysis and phenotypic characterization of new and several preexisting mutations in this region will result in significant progress toward the long term goal of these studies - to determine the relationship between genes defined by mutations, molecularly defined transcription units and gene sequences identified by large scale genomic sequencing. Molecular genetic information about the function of genes within the central portion of mouse chromosome 5 will be directly applicable to the genetic dissection of the homologous region on human chromosome 4. Moreover, the results of this project will facilitate a global study of the mammalian genome and provide insight into genome alterations that lead to abnormal function / disease.

DESCRIPTION (Adapted from the applicant's description) The long term goal of our studies is to identify genetic factors that underlie molecular events involved in the regulation of sleep. A wealth of information on the neuroanatomy, neurochemistry and neurophysiology of sleep provides a firm foundation for the classical genetic approach to the studies of sleep proposed in this application. Due to the paucity of candidate genes for key regulators of sleep-wakefulness, we propose using forward genetics to screen for single gene mutations that affect sleep patterns in the mouse. This approach involves mutagenesis of the whole genome with a potent chemical mutagen, N-ethyl-N-nitrosourea (ENU), and the generation of a large number of progeny which will then be monitored for aberrant behavioral phenotypes (indicated by differences in levels of locomotor activity, the ratio between the length of the activity and rest phase, recovery rest following short term sleep loss, startle amplitude and pre-pulse inhibition of the startle response) that may point to sleep abnormalities. The relevance of the observed behavioral anomalies for identifying abnormal sleep will be verified with a standard electro-physiological assessment of sleep (by electroencephalogram and electromyogram) performed on the progeny of potential mutants. Further behavioral testing and neuropathological examination will be used to rule out possible pleiotropic mutations and/or identify neural correlates that may underlie the observed deviations in sleep parameters. The next step will involve genetic characterization and mapping of sleep mutants, which will facilitate the interpretation of behavioral findings, as well as provide the basis for the positional cloning of genes responsible for sleep anomalies. Human orthologs of loci defined by these single gene mutations may represent additive or interactive contributions to the polygenic component of inherited sleep disorders. Random mutagenesis and screening for aberrant behavioral traits in mice have not been extensively utilized and have great potential for discovering genes underlying the regulation of sleep and wakefulness in mammals. (end of description)

The long term objective of our studies is to identify novel behavioral mutations in mice which may allow the dissection of genetic pathways underlying neurobiological processes. The limited availability of appropriate animal models has hindered research in complex neuropsychiatric illnesses such as schizophrenia, major affective disorders, and sleep disorders. Pharmacologic and surgical manipulations have been used to induce behaviors that simulate aspects of human disorders in laboratory animal models; however, these models represent phenocopies and do not reflect genetic causes of aberrant behavior in the animal. The approach outlined in this proposal involves random mutagenesis of the mouse genome, using a potent mutagen N-ethyl-N nitrosourea (ENU), and screens for abnormal behavioral phenotypes; in particular, disrupted rest: activity behavior and altered acoustic startle response. Novel mutations will be phenotypically characterized based on their chromosomal location and complementation analysis. This project is part of a larger effort-NIH Investigator-Initiated Interactive Research Project (IRPG)- to perform a genome-wide search for dominant behavioral mutations and to saturate a portion of the mouse genome with recessive mutations. A combination of random mutagenesis (IRPG one proposal) with an effort to create deletions (IRPG two proposal) across the 30 cM region on mouse chromosome 5, well characterized at the genetic and molecular level, will allow the identification of recessive mutations in a two generation saturation screen for 2% of the mouse genome. The IRPG effort will allow identification, phenotypic analysis and mapping of development and behavioral mutants as well as identification of mutants whose dominant behavioral anomalies may be due to underlying development, and/or neuroanatomical and neuropathological defects.

DESCRIPTION (provided by applicant): The elucidation of the functional content of the genome of a model organism, such as the mouse, involves identification of genes and evolutionarily conserved non-transcribed sequences through genome sequencing, analysis of gene expression and analysis of normal and mutant phenotypes. This grant application is being prepared at an exciting time for mammalian biology, when each one of these efforts is being undertaken on a large scale - whole genome level. In this project we propose to initiate the integration of these functional genomics data, using a 100 Mb chromosomal region of the mouse genome as a model. Our aims are to (1) identify, characterize and clone a set of novel embryo-lethal mutations within the 30cM/60 Mb region covered by the Rw inversion (2) perform functional analysis of the cluster of GABAA receptor subunit genes by generation of Embryonic Stem (ES) cells with chromosomal deletions and phenotypic analysis of mice carrying these deletions and (3) assemble a gene index and perform comparative sequence analysis of genes shown to be involved in developmental processes (based on mutant analysis and/or available expression data). The minimal set of genes necessary for embryonic survival is a key question in developmental biology. In this proposal we use a defined chromosomal region, representing 2 percent of the mouse genome, and available genetic, genomic and bioinformatics resources to systematically search for and catalog these genes and their mutant phenotypes.

DESCRIPTION (provided by applicant): The long term goal of our studies is to identify genetic factors that underlie molecular events involved in the regulation of rest:activity behavior in the mouse. We employed forward genetics approach to identify single gene mutations that cause gross changes in activity levels or organization of rest:activity cycles. These changes may be due to general metabolic or developmental defects, or caused by anomalies in neurobiological processes, such as the circadian system and regulation of sleep. We have established an integrated and nested phenotypic protocol, which will allow us to characterize novel mutations on several levels; molecular, neuropathological, electrophysiological and behavioral. Our aims are: Aim 1) Genetic characterization, mapping and positional cloning of rest/activity mutants, identified by random ENU (N-ethyl-N-nitrosourea) mutagenesis; 1) Aim 2) Characterization of sleep patterns in selected rest:activity mutants; Aim 3) Determination of the specificity of rest:activity disturbances; Aim 4) Microarray analysis will be used to define downstream pathways disrupted by the mutant gene. Initially, this project includes genetic and phenotypic characterization of two mutations, one with an effect on circadian period (Rooster), and second, Bedlam, associated with a decreased amplitude of the circadian rhythms. Our hypothesis is that a subset of rest:activity mutations will uncover novel genes involved in the regulation of sleep and their interaction with the other neurobiological processes, such as those that underlie learning and memory or circadian system. Human orthologs of loci defined by these single gene mutations may represent additive or interactive contributions to the polygenic component of inherited psychiatric and sleep disorders.

[unreadable] DESCRIPTION (provided by applicant): Some of the most prominent and characteristic symptoms of recurrent affective illnesses, e.g., bipolar and unipolar disorders, include dramatic disturbances in the sleep:wake or rest:activity cycle. It has been shown that manipulations of rest:activity cycles can profoundly change the course of clinical manifestations of these illnesses. These findings suggest that genetic predisposition to alterations in sleep:wake or rest:activity cycles represent endophenotypes for recurrent affective disorders. Based on recent findings that gene expression of 5-10% of genes in the genome are controlled by the clock (Akhtar et al., 2002; Panda et al., 2002), we propose that clock genes and/or genes that are controlled by the clock may underlie polygenic inheritance of the bipolar disorder. Recent findings that cultured skin fibroblasts, following a serum shock (50% fetal bovine serum) exhibit synchronized cycling in the expression of core clock genes (Balsalobre et al., 1998; Nagoshi et al., 2004) provide a new avenue to investigate the link between bipolar illness and the circadian system. The purpose of this application is to explore the possibility of using fibroblasts derived from skin biopsies of bipolar patients to investigate their intrinsic circadian clock. Specifically, our aims are: Aim 1. To develop a high-throughput method for monitoring circadian rhythms in human cultured fibroblasts to examine cycling expression of core clock and clock-controlled genes by Real-time quantitative PCR and bioluminescence recording; and Aim 2. To apply the method to examine if the core clock machinery is involved in the pathophysiology of bipolar disorder. In this aim will involve analysis of fibroblasts cell-lines available through the Coriell Cell Repositories (Important note: The identity of the specific individuals that were the source of the original cells is not known to the repository).This highly exploratory project has a tremendous potential for accelerating our understanding of the genetic basis of an important mental disorder. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Bipolar affective disorder is a life-long, often recurring, chronic mental illness which typically appears in young adulthood and is characterized by fluctuations in mood, including recurrent episodes of mania or hypomania and depression. Lifetime prevalence estimates for bipolar disorders are 1 - 3.7%, and tragically, approximately 10-15% of affected individuals die of suicide. A long term goal of our studies is to identify the molecular events that underlie bipolar disorder. To complement ongoing large-scale genome-wide association studies, we focus on an exceptionally large Amish family with high incidence/prevalence and risk of developing bipolar disorder to identify common and rare variants, single nucleotide polymorphisms and structural variants associated with disease susceptibility. We propose to combine high-density SNP genotyping (on all individuals) with next- generation sequencing (on a selected subset) to extract maximum value from the complete understanding of genetic variation in this genetic isolate. To achieve this goal, we propose a three- step strategy: a) to establish a high-density genotype map for all 450 well-phenotyped family members in the pedigree segregating bipolar disorder using Illumina Omni2.5-Quad arrays; b) to generate, in an unbiased way, a full spectrum of genetic variants by whole genome sequencing of 60 individuals (20 parent-child trios), from genomically defined subfamilies and with different disease status (affected and unaffected); and c) to infer a spectrum of identified mutations (rare single nucleotide polymorphisms and structural variants) to the entire pedigree, specifically to unsequenced family members that harbor overlapping haplotypes, (Li et al., 2009; Howie et al., 2010). This variant imputation will use correlation between SNP markers on the genotype platform available for all 450 subjects and SNPs identified by deep sequencing, to predict positions and genotypes of novel common and rare variants. Functional annotation of risk alleles (single variants and combinations of alleles) in the large number of affected and unaffected family members, combined with the variation identified through the 1000 Genome Project, should accelerate identification of disease genes. This project will use and further develop high throughput genomic approaches for a combined analysis of genotypes and whole genome sequence that should be applicable to the analysis of other psychiatric disorders and studies of large families. The identification of etiological basis of bipolar disorder in a genetic isolate will shed light on the gene pathways that are involved in this disorder in the general population.

DESCRIPTION (provided by applicant): The NIMH Research Domain Criteria (RDoC) Project provides a framework for the integration of basic and translational research, by defining basic dimensions of function that cut across disorders. Circadian and sleep disturbances, observed in several highly incapacitating and debilitating mental disorders, represent the key constructs within the Arousal and Regulatory system RDoC Domain. The study of circadian and sleep disturbances in psychiatric populations is difficult and contradictory results between numerous small scale studies may be due to a high degree of genetic and phenotypic heterogeneity. Our long-term aim is to evaluate the feasibility of applying actigraphy for large-scale genetic studies across traditional diagnostic categories. To facilitate these studies, we need to evaluate novel methods for large scale, longitudinal profiling of activity patterns. Our Aims are: Aim 1. To identify actigraphy devices and establish protocols suitable for concurrent and longitudinal ascertainment of activity and sleep: wake patterns and mood symptoms in 100 subjects recruited from an outpatient mood disorders treatment program; and Aim 2. To test heritability of an array of quantitative parameters collected using actigraphy devices (including activity levels, timing of activity and rest onsets, amplitude, organization of activity-rest cycles) and self- reported mood symptoms in a set of 250 twin pairs from the Pennsylvania Twin Registry. The comparison of individual difference measures and heritability of parameters related to mood, sleep: wake or rest: activity patterns in subjects presenting at the mood clinic and in a set of twins represents will provide important preliminary insights into parameter selection for large scale genetic studies across several disorders. Finally, the RDoC initiative is new and will require an integration of approaches across fields. This proposal will provide an opportunity for the interdisciplinary team (geneticists, clinical psychologists, psychiatrists and statisticians) t initiate RDoC-based phenotyping relevant to sleep and circadian disturbances.

? DESCRIPTION (provided by applicant): The goal of the Penn Computational Genomics Training Grant (PENN CG-TG) Program is to train the next generation of quantitative genomic scientists who will develop new algorithms and quantitative models to address biomedical problems using genomic technologies. Recent developments in genomics of dynamic functional data as well as the $1,000 genome next-generation sequencing are accelerating the need for well-trained computational genomicists. Penn has trained computational genomicists since 1994 and created an independent PhD degree program since 2001. Leveraging extensive experience and resources of Penn's research and training programs, PENN CG-TG will train 8 pre-doctoral students in year 3 and 4 of their PhD program, supporting their training with core courses, seminars, mentoring, and symposiums. Students will learn foundational knowledge to understand algorithms and modeling at a deep level learn biology background to generate computational models of key biomedical problems, learn to communicate and disseminate quantitative material, and understand the importance of provenance and integrity in large-scale data analysis.

SUMMARY The objective of the Diversity Action Plan at the University of Pennsylvania (Penn) Genomic Program (DAPPG) is to guide undergraduate and recent college graduates from underrepresented (UR) groups into graduate school to pursue genomics research. The Diversity Action Plan will be developed and implemented in partnership with NHGRI T32 training programs in Computational Genomics, Genomic Medicine, and the Ethical, legal and social implications of genetics and genomics as well as a research program in sub-cellular genomics. This program especially emphasizes vertical integration of regional training, recognizing the geographic mobility restrictions common to many socio-economically disadvantaged UR groups. At least 10 UR scholars (7-10 summer students and 3 post-baccalaureate students), from the Greater Philadelphia area along with additional candidates from across the country, will be identified each year who are interested in genomics and genomic medicine research but lack the experience or expertise necessary for graduate or medical school training. Each scholar will be matched with a research mentor among the graduate training faculty as well as PhD/Postdoctoral level mentors from the T32 programs and provided with a significant one to two-year research project. An Individual Student Development Plan (IDP) is developed for each scholar to ensure that the scholar undertakes training appropriate to his or her own scientific needs and interests, including graduate or advanced undergraduate coursework. Scholars will also develop the skills necessary for success in graduate school by completing workshops in genomics and computational biology, grant writing, the responsible conduct of research, critical analysis of scientific literature, and oral and written presentation skills. Scholars will also participate in a one-on-one writing workshop with a professional writing instructor. Scholars will meet weekly as a group to discuss scientific journal articles and their own research, in order to further develop their skills in the critical evaluation of research and the delivery of scientific presentations, as well as to increase their exposure to research outside their own research groups. In addition, scholars will receive advising for the graduate school application process, including selecting programs, writing application essays, and practicing interviews. If necessary, scholars will take a GRE or MCAT preparation course. Scholars will also participate in various lunches and seminars with Penn faculty, postdoctoral fellows and graduate students who share their training experiences, on-going research, and academic paths. The ultimate measure of the program's success is the percentage of scholars who are admitted to PhD or MD-PhD training programs and pursue a research career. !

ABSTRACT Given the growing prevalence of autism spectrum disorder (ASD), there is an urgent need to better understand its etiology. Genetic variation identified through association and sequencing studies has provided valuable clues about the biological underpinnings of ASD, which is highly heritable. Our objective is to combine family-based genetic studies of ASD currently ongoing at the Autism Spectrum Program of Excellence at the University of Pennsylvania (Penn ASPE) with the Simons Simplex Collection (SSC) and SPARK genetic data to investigate fundamental mechanisms contributing to ASD risk, specifically in ASD subjects without intellectual disability (ASD w/o ID) and their families. New datasets, such as SSC, SRARK and our own ASPE collection, will allow us to genetically dissect polygenic risk burden and address a longstanding but unexplored hypothesis that assortative mating may contribute to ASD liability. If assortative mating is present, this affects a broad range of studies of genetics of ASD and will be important to consider in re-analysis of existing data and design of future studies. Having scientific evidence to support or refute these theories may be of immediate and direct value to the ASD community. We propose the following Specific Aims: Aim 1) To genetically dissect genomic features that differentiate ASD with and without ID. Genetic dissection of polygenic risk scores and rare or low frequency variants known to be associated with ASD will be further combined with the analysis of ancestry and trait-related assortative mating across a large number of ASD families (Aim 2). Aim 2) To characterize assortative mating in parents of probands with ASD. We hypothesize that assortative mating, at a trait-level and a genomic level, will be more prevalent in parents of ASD w/o ID probands. Correlation of the PRS and the degree of assortative mating in families with a rich collection of phenotypes and traits (in SPARK, SSC and ASPE) will reveal important insights into polygenic architecture and may provide critical mechanistic threads. The focus of this proposal on ASD w/o ID is novel because most previous ASD genetics findings have been in studies that recruited ASD probands with ID. Similarly, the role of ancestry and trait- based assortative mating in ASD families may reveal unique aspects of genetic architecture in some ASD families and thereby facilitate interpretation genetic findings. As most standard genetic analysis approaches assume random mating, if assortative mating on the basis of social responsiveness or other ASD-related traits is present, this has implications for a wide range of studies of ASD genetics.

Project Summary University of Pennsylvania has trained computational genomicists for the past 20 years supported by the NHGRI T32 program, training 52 predoctoral and 13 postdoctoral trainees the majority of whom have gone on to careers in research and development. Here we propose to continue our Computational Genomic Training program with eight predoctoral and two postdoctoral trainees focused on the theme of Data Science and Machine/Statistical Learning methods as applied to genomics data. Our training program concentrates on a rigorous course-based curriculum supported by courses in multiple graduate groups. In addition, our training also involves 13-16 hours of Responsible Conduct of Research (RCR) and Scientific Rigor and Reproducibility (SRR) training, Individual Development Plan, and utilization of Electronic Notebooks and code repositories. Research training is enhanced by a dual mentorship model whenever possible. Our program is supported by a greater genomics training program consisting of three NHGRI T32 programs in Computational Genomics (this program), Genomic Medicine, and ELSI, as well as a NHGRI R25 Diversity Action Plan (DAP). In particular, the DAP program recruits URM undergraduate and postbacc trainees focused on the Greater Philadelphia Area whose many institutions serve the urban URM population. This vertical regional integration will allow us to develop a regional pipeline of URM students with undergraduate research experience in genomics. Our program consists of 31 trainers of which 14 are female scientists and 2 are URM scientists. Twenty one of our 31 trainers have an active computational genomics research program. The expertise of the trainers span disease genomics, genomic technologies, multidimensional statistics, algorithms, data sciences, and machine learning. Our training environment is enhanced by key facilities including large biobanks, high-throughput genomics core, high-performance computing core, and a unique immersive data visualization facility. Penn overall hosts more than 60 NIH training programs with strong institutional administrative support for managing the training programs including an Office of Biomedical Postdoctoral Programs, Office of Diversity and Inclusion, combined Biomedical Graduate Studies, among others. Success of our training program will help train the next generation of genomic workforce in the skills and knowledge necessary to apply state-of-art computational techniques to genomics and develop new techniques for novel genomic data.