This region of chromosome 2 is home toGAD1(glutamatedecarboxylase 1, gene map locus 2q31), which catalyzes theconversion of glutamic acid to gamma-aminobutyric acid(GABA), the major inhibitory neurotransmitter in the verte-bral central nervous system. It has been associated with alco-hol dependence in Han Taiwanese males (Loh et al., 2006). Inthe IASPSAD sample, no association was found with depen-dence, but significant associations were found with initial sen-sitivity to alcohol and to age at onset of dependence (Kuoet al., 2008). Kuo and colleagues (2008) proposed that theunderlying pathophysiology regulated byGAD1may be moredirectly related to the component processes of dependence,than to the disorder itself.A small linkage peak (LOD 1.6) on chromosome 3 waslocated in the same region as linkage found previously in thesame sample for alcohol consumption (LOD 2.7) (Hansellet al., 2009). In addition, a minor peak (LOD 1.0) wasobserved in the region encompassing the gamma-aminobutyricacid receptor A subunit 2 (GABRA2) gene. This gene andothers in its cluster on chromosome 4 (GABRA4,GABRB1,GABRAG1) have been found to be associated with alcoholdependence (as well as other substance use disorders andpsychopathology) across several independent studies (Covaultet al.,2004;Edenberget al.,2004;Fehret al.,2006;Lappalainen et al., 2005; Lind et al., 2008; Matthews et al.,2007; Soyka et al., 2008).Somelimitationsneedtobeconsidered.First,theseresults were not genomewide significant. However, the likeli-hood that these are false positives is somewhat diminishedas the results replicate linkage found in these regions by pre-vious studies. Further, quantitative trait loci of small effectare to be expected given that an alcohol dependence symp-tom score is likely to be inﬂuenced by many genes of smalleffect, as has been found for numerous traits in genomewideassociation studies (Wray et al., 2007). Larger linkage sig-nals, as have been reported previously, may simply reﬂectthe high variance of effect size estimates found with smallersamples. Linkage differences related to sample variationwere found in the current study, but are not sufficientlylarge for meaningful interpretation. A further limitation isthat the component samples were collected for heteroge-neous purposes and may not be entirely representative ofthe adult Australian population. Notwithstanding these limi-tations, our analyses demonstrate the utility of a communitysample, results of which are easily generalizable, for linkageanalyses.Thechromosomalregionsidentifiedprovideafocus for future gene-mapping efforts.ACKNOWLEDGMENTSWe thank all participating families, the project coordina-tors and interviewers under the supervision of Dixie Statham,the data managers under the supervision of John Pearson andDavid Smyth, and laboratory personnel under the supervisionof Megan Campbell and Anjali Henders. For genome scansof twins and siblings, we acknowledge and thank the Mam-malian Genotyping Service, Marshfield WI (Director: Dr.