Didactic Program

University of Chicago Training Program in Psychiatric Genetics
Didactic Program

Excerpted from the NRSA Institutional Training Fellowship Award application

Executive Committee: Profs. Elliot S. Gershon (PI), Nancy J. Cox (Co-PI), David Ledbetter, Edwin Cook, Martha McClintock

This postdoctoral training program is focused on interdisciplinary approaches to genetics of complex inheritance psychiatric disease, and to basic research and technology development that may prove pertinent. Because the program includes what could be considered translational research for which physician-scientists are particularly well suited, we expect up to one half of the program participants to be M.D.s or M.D./Ph.D.s, while the remainder will be Ph.Ds.

Based on our experiences with and perceptions of other training programs within the University of Chicago and at other universities, we anticipate that trainees entering this program will have a broad range of experience, background and research interests. For example, we expect to attract psychiatrists who may have considerable experience in diagnostic assessment, but desire more training in molecular biology and/or genetic analysis. We also expect to attract individuals who have experience and background in molecular biology, but no experience in genetic studies of complex disorders or psychiatric diagnosis. Similarly, we may well attract individuals with quantitative backgrounds who have no specific experience in genetic studies or psychiatric disorders. Thus, we will not be surprised to draw trainees from psychiatry, molecular biology, anthropology, physics, mathematics, statistics, genetics and other medical subspecialties. Although all trainees will be expected to complete our core course requirements, we will also be able to provide additional course work for those who need background in, for example, basic statistics or genetics, as well as in advanced topics such as molecular genetics and statistical genetics. Thus, we expect each trainee to develop, in consultation with their faculty mentor and the executive committee, an essentially unique training program that combines formal course work, clinical education, and research activities that builds on their previous education and their current research interests.

The duration of the program will generally be 3 years, but may be decremented by 1 or 2 years on a case-by-case basis for unusually qualified and experienced applicants. The decision to allow this will be made after recommendation by the training mentor to the executive committee, and review of the proposal by the Executive Committee. This will occur either before the candidate is admitted to the program, or before the end of the first year. In any case, the duration of the training program will be in whole years.

(a) Didactic Teaching and Examples of Related Potential Research Projects

Although the overall course work required for each trainee will be decided on a case-by-case basis with the trainee and executive committee, the following four courses will generally be required for each trainee in their first two years, with at least two to be completed the first year:

1) Multidisciplinary Approaches to Psychiatric and Behavioral Genetics (directed by Drs. Gershon and Cox), is outlined below. It will be offered yearly, and all trainees will be expected to attend seminars during their first year, and to present their research in the trainee presentations that will be given at the end of each year. Later year trainees will also present their work at the end of each academic year. The course will be advertised throughout the university and it is likely that other students, post-docs and faculty will attend at least some lectures.

I. Perspectives on the genetics of human behavior

II. Behavioral assessment and epidemiology

III. Genomics and Transgenics

IV. Overview of current understanding of genetics and epidemiology of psychiatric disorders and behavioral traits

V. Trainee Presentations. Each trainee, at the end of the formal didactic part of the course, will present a seminar on her or his planned work. These seminars would begin in the late spring of the academic year, so the trainee will have been in the program for at least 9 months.

2) Systematic Diagnosis of Psychiatric Disorders , directed by Elliot Gershon, will be given yearly, with all trainees expected to attend the course during their first year. The course is designed to give a minimal understanding of the diagnostic process for research, including research conducted by non-clinicians. A more extensive training program in diagnosis is offered to participants by specific research projects within the Department.

I. Introduction – definition, operational diagnostic criteria and their validation
II. Data Collection – instruments, algorithms, rating scales
III. Nosology – categorical logic, diagnostic classification
IV. Reliability and Validity – theory, statistical methods, problems
V. Conclusion – controversial issues and future directions

3) Essentials of Patient-Oriented Research Course , from the Clinical Research Training Program of the Department of Health Studies. This course is offered yearly, with key faculty including Fredric Coe, M.D., Murray Favus, M.D. and Ronald Thisted, Ph.D. of the Department of Health Studies. Major topics include:

I. History of modern clinical research
II. Hypothesis generation and experimental design
III. Statistics for medical researchers
IV. Conforming to law and society
V. Economic and financial issues
VI. Support systems
VII. Grant management
VIII. Data management
IX. Reporting research results

4) Human Genetics II: Introductory Statistical Genetics (University course offered by Dr. Nancy Cox and Dr. Judith Badner)

Students completing this course will understand genetic models for complex human disorders and quantitative traits. They will be able to conduct parametric and non-parametric linkage analyses, as well as linkage disequilibrium mapping using transmission/disequilibrium tests (TDT) and decay of haplotype sharing (DHS). Students will have a thorough understanding of the assumptions, robustness and limitations of these approaches, and an appreciation for the complexities of interpreting studies using these approaches. The course will include computational labs, and will be taught in a computational lab. The final exam will be a take home exam, in which students receive a data set for analysis (which will require using virtually all of the techniques learned in the course) and interpretation.

1. Genetic models for human phenotypes – single locus, polygenic, and mixed models for dichotomous and quantitative phenotypes, classic threshold models, parameters and their relationships
2. More sophisticated genetic models – gene x gene interaction, gene x environment interaction, how do these models fit into the context of general “models” such as common disease/common variant or multi-equivalent risk model; implicit and explicit assumptions of models
3. Segregation analysis – historical overview, modern implementations, ascertainment issues, regressive models, path analysis and variance components analysis
4. Linkage analysis – classic parametric 2-point linkage analysis, introduction to multipoint analysis, comparison of algorithms for linkage analysis; MOD scores, likelihood ratio tests
5. Linkage analysis – non-parametric linkage analysis; complete discussion of impicit and explicit parameters, sensitivity of results to nuisance parameters
6. Computational lab on linkage analysis – FASTLINK for parametric analysis; likelihood ratio analyses using revision of ILINK; parametric and non-parametric analysis with GENEHUNTER, GENEHUNTER-PLUS, and ALLEGRO; ASPEX; ANALYZE
7. Linkage disequilibrium mapping – linkage disequilibrium, power to detect, theoretical and empirical data, transmission/disequilibrium test (TDT) and related approaches
8. Linkage disequilibrium mapping – decay of haplotype sharing, coalescent theory, haplotype estimation in related and unrelated individuals
9. Computational lab – TDT software including ASPEX, ANALYZE, TRANSMIT; DHS
10. MIDTERM EXAM
11. Case/control vs. family-based LD mapping methods – efficiency vs. quality of information, correction of test statistics vs. the identification of substructure; estimation of proportions of admixture
12. Computational lab – correction of case/control association test statistics, estimation of substructure; estimation of proportion of admixture (STRUCTURE, etc.)
13. Linkage analysis of quantitative phenotypes – Haseman and Elston regression and related approaches, path analytic approaches, variance components approaches, and the relationships among these
14. Computational lab on linkage analysis of quantitative phenotypes – SOLAR, GENEHUNTER.
15. One million ways to screw up data (and how to find and fix them all) – computational lab and lecture – relationship misspecification, genotyping error, random and non-random error
16. Choice of study population – myths and facts, isolated populations, admixed populations and mapping by admixture linkage disequilibrium – lab with empiric studies of LD using different approaches to estimating LD
17. Lecture/lab – linkage studies allowing for gene x gene and gene x environment interaction; theory and applications
18. Methods under development – genome-wide studies of interaction (beyond pairwise); distinguishing variants affecting susceptibility to disease in the context of positional cloning studies for complex phenotpes
19. Computational lab – accounting for the evidence for linkage; conditioning on SNP genotypes, approaches used in SOLAR, cluster approaches from AGILENT
20. Computational lab – informatics; BLAST, gene data bases, sequence data, establishing physical maps, doing sequence comparisons, homologies, transcription factor data bases
21. FINAL EXAM (take home)

In addition to the courses outlined above, there are a variety of courses offered at the University of Chicago that will be of potential interest to trainees in the Psychiatric Genetics Training Program, and could be made a part of an individualized training program. Among these courses are Human Genetics (HG 470, Ledbetter), Genetic Analysis (MGCB 314, Pruess), Molecular Biology I (MGCB 312, Rothman-Denes), and Human Variations and Disease (HG 469 DiRienzo).

(b) Examples of potential trainee projects related to the didactic curriculum:

Integration of ethical, diagnostic assessment, and family ascertainment components of the curriculum: develop a study of adult children of Alzheimer's patients, in which they are surveyed on attitudes towards, and desirability of, testing for Alzheimer's disease ApoE4-related susceptibility. Assessment of participants' current cognitive and ApoE statuses.

Integration of human and rodent behavioral, and statistical genetics components: study of known polymorphisms with individual differences in subjective, behavioral or physiological responses to a challenge dose of a psychoactive drug, such as nicotine. Mapping of a genetic locus for cocaine consumption in mice using an existing behavioral measure and mice who are the second generation (F2) of a cross between high and low cocaine consumption strains. Development and incorporation of a scale for intensity of mania as a quantitative trait in an association study of mania, or a genetic linkage study of bulimia.

Integration of molecular genetic and informatics components of curriculum: Integrated gene, transcript, STS, STR polymorphisms, and SNPs map of a genomic region of interest. In such a map, existing gaps in the finished sequence will be filled, orientation of the BAC scaffolds/contigs will be validated, and redundancies and positional errors in SNPs deposited in a national database such as dbSNP will be eliminated.

Clinical genetic investigation projects: Detection of microdeletions or insertions in regions where linkage has been reported in Bipolar illness or Schizophrenia. Pharmacogenetics of clinical response to serotonin-specific-response inhibitors, studying association of response with polymorphisms of the serotonin transporter (including newly discovered SNPs from mutational analysis of patients), serotonin receptor genes, and genes involved in related signal transduction events.

Integration of statistical analysis with already collected clinical and molecular data: analysis of SNP data generated in the context of positional cloning studies on complex disorders, including TDT and related approaches and decay of haplotype sharing (DHS); analysis of data collected in the context of genome-wide linkage screens for qualitative or quantitative traits; testing data generated for case-control association studies for population substructure; secondary studies on genome-wide linkage data designed to identify gene x gene and/or gene-environment interactions.

(c) Grant submission by the end of each trainee's program

Given the strength and multidisciplinary nature of the didactic education, and the research experience including exposure and assimilation into a network of investigators, the trainee is expected to prepare and submit his or her first application for independent support, as a K- award or an R01, by the end of the third year in the program. The mentor is expected to be supportive in the process. Because the projects will tend to be multidisciplinary in nature, we anticipate that multiple training faculty will support trainees in each grant preparation.

University of Chicago Training Program in Psychiatric Genetics Didactic Program

University of Chicago Training Program in Psychiatric Genetics
Didactic Program