In summary, data from the UCSF Family study did not reveal evidence for genome-wide significance for linkage (LOD > 3.0) to the DSM-IV cannabis dependence diagnosis, however, linkage analyses of cannabis “craving” and the cannabis withdrawal symptom of “nervous, tense, restless or irritable” revealed five sites with LOD scores over 3.0 on chromosomes 1, 3, 6, 7, 9. These results have identified new regions of the genome associated with cannabis use phenotypes as well as corroborated the importance of several chromosome regions highlighted in previous linkage analyses for other substance dependence phenotypes. The results of this study should, however, be interpreted in the context of several limitations. First, multiple phenotypes were tested in the present study, thus increasing the possibility that a false positive result was interpreted as substantive. To protect against this possibility, empirically-derived p-values were reported for each linkage signal allowing for the objective evaluation of the linkage evidence. Second, the findings may not generalize to other large studies that were not ascertained for alcoholic probands. Third, only retrospective and cross-sectional data on cannabis use and use disorders
