Behavior Genetics, Vol. 31, No. 6, November 2001 (© 2001) 497 0001-8244/01/1100-0497$19.50/0 © 2001 Plenum Publishing Corporation A Genome-Wide Scan of 1842 DNA Markers for Allelic Associations With General Cognitive Ability: A Five-Stage Design Using DNA Pooling and Extreme Selected Groups Robert Plomin, 1,5 Linzy Hill, 1 Ian W. Craig, 1 Peter McGuffin, 1 Shaun Purcell, 1 Pak Sham, 1 David Lubinski, 2 Lee A. Thompson, 3 Paul J. Fisher, 4 Dragana Turic, 4 and Michael J. Owen 4 All measures of cognitive processes correlate moderately at the phenotypic level and correlate sub- stantially at the genetic level. General cognitive ability (g) refers to what diverse cognitive processes have in common. Our goal is to identify quantitative trait loci (QTLs) associated with high g com- pared with average g. In order to detect QTLs of small effect size, we used extreme selected sam- ples and a five-stage design with nominal alpha levels that permit false positive results in early stages but remove false positives in later stages. As a first step toward a systematic genome scan for allelic association, we used DNA pooling to screen 1842 simple sequence repeat (SSR) mark- ers approximately evenly spaced at 2 cM throughout the genome in a five-stage design: (1) case- control DNA pooling (101 cases with mean IQ of 136 and 101 controls with mean IQ of 100), (2) case-control DNA pooling (96 cases with IQ 160 and 100 controls with mean IQ of 102), (3) individual genotyping of Stage 1 sample, (4) individual genotyping of Stage 2 sample, (5) trans- mission disequilibrium test (TDT; 196 parent-child trios for offspring with IQ 160). The over- all Type I error rate is 0.000125, which robustly protects against false positive results. The num- bers of markers surviving each stage using a conservative allele-specific directional test were 108, 6, 4, 2, and 0, respectively, for the five stages. A genomic control test using DNA pooling sug- gested that the failure to replicate the positive case-control results in the TDT analysis was not due to ethnic stratification. Several markers that were close to significance at all stages are being investigated further. Relying on indirect association based on linkage disequilibrium between mark- ers and QTLs means that 100,000 markers may be needed to exclude QTL associations. Because power drops off precipitously for indirect association approaches when a marker is not close to the QTL, we are not planning to genotype additional SSR markers. Instead we are using the same design to screen markers such as cSNPs and SNPs in regulatory regions that are likely to include functional polymorphisms in which the marker can be presumed to be the QTL. KEY WORDS: Intelligence; genome scan; quantitative trait loci (QTLs); DNA pool; association. 1999a). One of the most consistent findings from indi- vidual differences research on human cognitive abili- ties and disabilities during the past century is that di- verse cognitive processes intercorrelate. Despite the diversity of cognitive tests, individuals who perform 1 Social, Genetic and Developmental Psychiatry Research Centre, Institute of Psychiatry, King’s College London, London, UK. 2 Department of Psychology and Human Development, Vanderbilt University, Nashville, TN, USA. 3 Department of Psychology, Case Western Reserve University, Cleveland, OH, USA. INTRODUCTION General cognitive ability (g) is a highly heritable quan- titative trait that varies from a low end of mild mental retardation to a high end of gifted individuals (Plomin, 4 Department of Psychological Medicine, University of Wales Col- lege of Medicine, Cardiff, UK. 5 To whom correspondence should be addressed. Social, Genetic and Developmental Psychiatry Research Centre, Institute of Psychia- try, King’s College London, London, UK SE5 8AF. e-mail r.plomin@iop.kcl.ac.uk