Bernie Devlin

Research Projects - Computational Genetics Program - University of Pittsburgh

About The Lab

Publications

Software Distribution

Lab Members

Research Projects

In the News

Some contribution to Science (with selected publications) Below are some of "my" contributions to science. Like all scientific ventures, they result not just from the teams represented in the author lists but from the foundational research of the corpus of scientists. I owe much to my many grand mentors and colleagues, many represented in the author lists below. I want to acknowledge one mentor who is not on the lists below, namely Andrew G. Stephenson, my PhD advisor. Andy took a huge gamble on a not-ready-for-prime-time graduate student and helped me become far better than I was. Together we produced some keen research quantifying plant reproductive strategies, which led to my post-doctoral work on gene flow, which somehow led to statistical and human genetics, and so on.
Contents

1. Autism Genetics
2. Ancestry analysis
3. Cloning a gene affecting head and neck tumors
4. Genomic Control
5. Scientific basis for "The Bell Curve" and the heritability of IQ
6. DNA Forensics
7.Gene Flow

1. Autism Genetics: We have made substantial and fundamental contributions to the understanding of the genetic basis of autism by analysis of data involving copy number variants, common variants and rare sequence variants.

a. Autism Genome Project Consortium, with Scherer & Devlin corresponding authors. Mapping autism risk loci using genetic linkage and chromosomal rearrangements. Nature Genetics, 39, 319-328, 2007.
b. Glessner JT, Wang K, Cai G, ..., Devlin B, Schellenberg GD, Hakonarson H. Autism genome-wide copy number variation reveals ubiquitin and neuronal genes. Nature 459:569-573, 2009. PMCID: PMC2925224
c. Sanders SJ, Ercan-Sencicek AG, Hus V, ... Roeder K, Devlin B, State MW. Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism. Neuron 70:863-885, 2011. PMCID: PMC3939065
d. Anney R, Klei L, Pinto D, ..., Vieland VJ, Hakonarson H, Devlin B. Individual common variants exert weak effects on the risk for autism spectrum disorders. Human Molecular Genetics 21:4781-4792, 2012. PMID: 22843504
e. Sanders SJ, Murtha MT, Gupta AR, Murdoch JD, Raubeson MJ, ..., Günel M, Roeder K, Geschwind D, Devlin B, State MW (2012) Disruptive de novo point mutations, revealed by whole-exome sequencing, are strongly associated with Autism Spectrum Disorders. Nature 485:237-241. PMID: 22495306
f. Neale BM, Kou Y, Liu L, Ma'ayan A, Samocha KE, ..., Devlin B, Gibbs RA, Roeder K, Schellenberg GD, Sutcliffe JS, Daly MJ (2012) Patterns and rates of exonic de novo mutations in autism spectrum disorders. Nature 485:242-245. PMID: 22495311
g. Liu L, Lei J, Sanders SJ, Willsey AJ, Kou Y, Cicek AE, Klei L, Lu C, He X, Li M, Muhle RA, Ma'ayan A, Noonan JP, Sestan N, McFadden KA, State MW, Buxbaum JD, Devlin B, Roeder K. DAWN: a framework to identify autism genes and subnetworks using gene expression and genetics. Molecular Autism 5:22, 2014. PMID: 24602502
h. Gaugler T, Klei L, Sanders SJ, ..., Devlin B, Roeder K, Buxbaum JD. Most genetic risk for autism resides with common variation. Nature Genetics 46:881-885, 2014. PMCID: PMC4137411
i. De Rubeis S, He X, Goldberg AP, Poultney CS, Samocha K, Ercument Cicek A, ..., Zwick ME, Barrett JC, Cutler DJ, Roeder K, Devlin B, Daly MJ, Buxbaum JD (2014) Synaptic, transcriptional and chromatin genes disrupted in autism. Nature 515:209-215. PMID: 25363760

2. Ancestry analysis: Since the introduction of GC more elegant ways of controlling for population structure have been developed. We contributed to these approaches in fundamental ways by showing how genetic matching can be effective as a design tool and how the connections between multidimensional scaling and spectral graph theory, specifically a spectral embedding derived from the normalized Laplacian of a graph, can produce more meaningful delineation of ancestry than by using Principal Component Analysis.

a. Luca D, Ringquist R, Klei L, Lee AB, Gieger C, Wichmann HE, Schreiber S, Krawczak M, Lu Y, Styche A, Devlin B, Roeder K, Trucco M. On the use of general control samples for genome-wide association studies: genetic matching highlights causal variants. American Journal of Human Genetics 82, 453-463, 2008. PMCID: PMC2427172
b. Lee AB, Luca D, Klei L, Devlin B, Roeder K. Discovering genetic ancestry using spectral graph theory. Genetic Epidemiology 34:51-59, 2009. PMID: 19455578

3. Cloning a gene affecting head and neck tumors: Initiated by Dr. Charles Richard, the goal of this project was to find one or more mutations causing paragangliomas (PGL). I took on the project after Dr. Richard accepted a job in industry. With post-doc Bora Baysal, who performed the bulk of the early PGL work for his Ph.D. thesis, and within two years, the gene PGL1 was cloned and those results were reported in Science. The finding was a breakthrough, leading not only to the cloning of other genes for PGL, but also to a deeper understanding of much more common and more lethal pheochromocytomas.

a. Baysal, B.E., R.E. Ferrell, J.E. Willett-Brozick, E.C. Lawrence, D. Myssiorek, A. Bosch, A. van der Mey, P.E.M. Taschner, W.S. Rubinstein, E.N. Myers, C.W. Richard III, C.J. Cornellisse, P. Devilee, B. Devlin, Mutations in SDHD, a Mitochondrial Complex II Gene, in Hereditary Paraganglioma, Science 287:848- 851, 2000.

4. Genomic Control: It has long been known that heterogeneity among populations, both genetically and phenotypically, gives rise to spurious associations between phenotype and genotype. An elegant solution to this confounding is to use family structure to control for heterogeneity. Because of the difficulty of accruing family samples, we developed an alternative that controls for confounding using the properties of the genome. Since the original paper by Devlin and Roeder, Genomic Control or GC is now recognized as a general approach to understand the problem of genotype/phenotype confounding.

a. Devlin B, Roeder K. Genomic control for association studies. Biometrics, 55:997-1004, 1999.

5. Scientific basis for "The Bell Curve" and the heritability of IQ: When Herrnstein and Murray's The Bell Curve appeared in late 1994, a group at Carnegie Mellon University (Bernie Devlin, Stephen Fienberg, Daniel Resnick and Kathryn Roeder) decided to investigate the roots of this controversial bestseller. This research led to an early report to the Carnegie Commission, co-editing a book on the subject ("Intelligence, Genes and Success: Scientists Respond to The Bell Curve"), and a publication in Nature in July of 1997 entitled The Heritability of IQ. It challenged much of the wisdom underlying previous analyses of IQ data by showing that there are other biologically plausible ways of modeling the data. In part, these results also undermined the "scientific" wisdom in The Bell Curve. More importantly our results pointed to prenatal factors as possible predictors of population variation in IQ, something that is gaining wider appreciation as years go by and for a wide array of human phenotypes.

a. Devlin B, Daniels M, Roeder K. The heritability of IQ. Nature 388:468-471, 1997.

6. DNA forensics: In 1990 the use of DNA hypervariable markers for forensic inference was in a precarious position, subject to grave concerns about the validity of its use in the courts. By rigorous statistical modeling of the processes generating these molecular data, a series of 11 papers was produced that highlighted and resolved the issues in this volatile field, virtually all of them appearing in high impact journals such as Science, Journal of the American Statistical Association and The American Journal of Human Genetics.

a. Devlin B, Risch N, Roeder K. Forensic tests and Hardy Weinberg equilibrium. Science 253:1039-1041, 1991.
b. Devlin B, Risch N, Roeder K. Forensic inference from DNA fingerprints. Journal of the American Statistical Association 87:337-350, 1992.
c. Risch N, Devlin B. On the probability of matching DNA fingerprints. Science 255:717-720, 1992.
d. Devlin B, Risch N, Roeder K. The statistical evaluation of DNA fingerprinting: Critique of the NRC report. Science 259:748-749, 837, 1993.

7. Gene flow: My early work on statistical methods to quantify gene flow was some of most fun I've had in research, in part because it opened up a whole new world for me in terms of collaboration among statisticians (i.e., Bruce Lindsay and Kathryn Roeder) and scientists (Norman Ellstrand). With Bruce and Kathryn we introduced mixture models as a tool for estimating the movement of genes in plant populations, specifically the success of pollen donors, from genetic markers realized from seeds and mature plants (6, 13, 16.) With Norm Ellstrand, my post-doc advisor, we applied these approaches to plant populations.

a. Devlin B, Roeder K, Ellstrand NC. Fractional paternity analysis: theoretical development and comparison to other methods. Theoretical and Applied Genetics 76:369-380, 1988.
b. Ellstrand NC, Devlin B, Marshall DL. Impact of spatial isolation on gene flow in wild radish. Proceedings National Academy of Sciences, USA 86:9044-9047, 1989.
c. Roeder K, Devlin B, Lindsay BG. Application of maximum likelihood methods to population genetic data for the estimation of individual fertilities. Biometrics 45:363-379, 1989.
d. Devlin B, Ellstrand NC. Variation in male and female fertility in wild radish, a hermaphrodite. American Naturalist 136:87-107, 1990.
e. Devlin B, Clegg J, Ellstrand N. The relationship between flower production, male fertility and male reproductive success in wild radish populations. Evolution 46:1030-1042, 1992.