doi:10.1086/303104 Am. J. Hum. Genet. 67:i–ii, 2000 i This Month in the Journal This month in the Journal, we present a review article by Karl Pfeifer on the mechanisms of genomic imprinting (p. 777). This article outlines current knowledge on the imprinting process, with emphasis on both the human 11p15.5 region, which is associated with Beckwith-Wie- demann syndrome and Wilms tumor, and the 15q11- q13 region, which is associated with the Prader-Willi and Angelman syndromes. Dr. Pfeifer explains in detail how a cell might use a methylation imprint as both a positive and a negative signal for transcription. Also included in this issue are two invited editorials. In the first (p. 788), Stephen Mount cautions that, even as mapping of the human genome nears completion, researchers are still limited in their ability to predict gene structure. Especially as gene annotation becomes more automated, researchers will need to be aware of the ex- ceptions to the rules for genetic-structure predictions. In the second invited editorial (p. 793), Rando Allikmets discusses the role of the ABC transporter ABCR in ret- inal disease. ABCR mutations lead to a clinically het- erogeneous array of phenotypes, including Stargardt disease, autosomal recessive retinitis pigmentosa, and autosomal recessive cone-rod dystrophy. Dr. Allikmets outlines current knowledge on the role of ABCR in a variety of retinal disorders and highlights the work of both Rivera et al. (p. 800), who find that mutations in ABCR are the major cause of autosomal recessive cone- rod dystrophy, and Maugeri et al. (p. 960), who provide further evidence that sequence variation in ABCR is as- sociated with age-related macular degeneration. TNNT1 Gene Mutation in Nemaline Myopathy, by Johnston et al. (p. 814) Amish nemaline myopathy is an autosomal recessive disorder that is very common in the Old Order Amish, with an incidence of ∼1/500. This muscular disorder presents as tremors in infants. Progressive muscle atro- phy and proximal contractures develop and eventually lead to respiratory insufficiency and death at an early age. Mutations in the genes for three sarcomeric pro- teins—nebulin, a-tropomyosin, and a-actin—have been associated with nemaline myopathy. To identify the gene involved in the unique Amish form of nemaline myopa- thy, Johnston et al. performed a genome scan and iden- tified, on chromosome 19, a 2-cM region of interest that included the gene for the sarcomeric protein troponin T. Sequencing of this gene revealed a truncating mutation, E180X, in an affected individual. Although it is not clear � 2000 by The American Society of Human Genetics. All rights reserved. 0002-9297/2000/6704-0001$02.00 whether this mutation would result in nonsense-medi- ated decay, protein instability, or defective protein ac- tivity, loss of the C terminus of the protein would re- move the sites of interaction with other components of the troponin complex and would disrupt the coupling of excitation and contraction in muscle fibers. Linkage Disequilibrium in the NF1 Gene Region, by Eisenbarth et al. (p. 873) Linkage disequilibrium (LD) is not evenly distributed throughout the human genome, but the reason for this lack of uniformity is unclear. Since the extent of LD has a major impact on genetic-association studies, a greater understanding of the LD distribution would allow more- efficient gene mapping. Eisenbarth et al. have examined LD in the region of the NF1 gene on chromosome 17q11.2, an area that shows variation in the extent of LD. They have been able to correlate differences in LD with the DNA composition in this region. More specif- ically, the recombination fraction begins to increase pre- cisely at an isochore-transition boundary. Because iso- chore class is defined by GC content, the change in recombination is associated with a change in the general DNA composition; a higher recombination fraction is associated with a higher GC content. As the authors point out, if this correlation between LD and isochore structure is a general phenomenon, it will have an impact on the marker density required for genetic-association studies. Markers could be concentrated in the regions with lower LD (high-GC content), thereby increasing the efficiency of these studies. APOE Haplotype Variation, by Fullerton et al. (p. 881) All populations studied to date are polymorphic at the APOE locus, which encodes apolipoprotein E. The three common alleles of APOE—e2, e3, and e 4—vary in their associated risk for both cardiovascular and Alzheimer disease. Individuals possessing an e4 allele have an in- creased risk of both diseases, whereas e2 is protective. More-complete examination of the alleles of APOE has demonstrated variation within each common isoform, and these data led Fullerton et al. to study further var- iation of APOE. Through careful haplotype analysis, these authors discovered considerable variation in the population-specific distribution of APOE haplotypes. This variation provides a potential explanation for the interpopulation differences in association of APOE with disease risk. On the basis of the haplotypes, reduced- median networks were created to study the history of Am. J. Hum. Genet. 67:i–ii, 2000 ii variation at the APOE locus. It appears that e4 is the ancestral allele but that e3 has risen in frequency during the past 200,000 years. Fullerton et al. hypothesize that this change in allele frequency, as well as a reduction in APOE variation, result from the rise of an advantageous mutation that is present in the e2 and e3 alleles. Admixture-Generated Linkage Disequilibrium, by Wilson and Goldstein (p. 926) The study by Wilson and Goldstein provides an impor- tant bit of information on the relationship between LD and admixture. Ultimately, this provides data on the number of markers necessary for genomewide associa- tion studies, an important unanswered question. It has been believed for quite some time that admixture creates LD, and this study is the first to conclusively prove this fact. Using the Lemba population from South Africa, which is a Bantu-Semitic hybrid population, these au- thors report elevated LD at large genetic distances, as well as increased LD at most genetic distances. This is in contrast to the situation in the putative parental pop- ulations, Bantu and Ashkenazi Jews, where LD extends only 0–6 cM. Analysis of allele-frequency differentials between the parent populations shows that, in the Lemba, there is a positive correlation between the dif- ferentials and LD, and this demonstrates that LD is gen- erated by admixture. Because LD measurements in dif- ferent populations and different genetic regions have shown considerable variability, researchers should no longer generalize LD across the genome and between populations. Population-specific measurements of LD, such as this one, will allow more-efficient use of markers in genomewide association studies, and they will also help us to understand better the factors, such as admix- ture, that affect LD variation. Report (Mutational Hotspots in mtDNA), by Stoneking (p. 1029) In attempt to resolve the controversy concerning the or- igin of the hypervariable sites in human mtDNA, Mark Stoneking has analyzed the evolutionary rates for new mutations at these sites. Although the hypervariable sites have generally been hypothesized to be mutational hot- spots, an alternative theory has been proposed—that is, that they may have resulted from ancient mutations that were moved to various mtDNA lineages via recombi- nation. Stoneking has found that new mtDNA mutations occur preferentially at the hypervariable sites, support- ing the hypothesis that the sites are mutational hotspots. However, the molecular basis for the increased mutation rate at these sites is unknown. KATHRYN BEAUREGARD Editorial Fellow