Portal de Periódicos da CAPES

This Month in Genetics

2016; Elsevier BV; Volume: 99; Issue: 6 Linguagem: Inglês

10.1016/j.ajhg.2016.11.003

1537-6605

Kathryn B. Garber,

Dermatoglyphics and Human Traits

Two of the things that creep me out the most in this world are snakes and rodents. Kvon et al. have now created a transgenic mouse that is the stuff of my nightmares, but I have to say it is one of the most fascinating papers I have read in ages. The group zeroed in on an enhancer sequence that controls expression of sonic hedgehog (Shh) from a distance of a million bases away. Comparisons of a variety of limbed species indicated that this sequence is conserved but that the constraint is lost in snake species that completely lack limb structures in their skeletons. To confirm the importance of this enhancer to limb development, the authors replaced the sequence in mice with the orthologous region from a variety of species ranging from coelacanth to human; in most cases, the expression of Shh in the limbbuds of mouse embryos was identical. In contrast, the enhancer from cobra and some other snakes did not serve as an enhancer for Shh in the mouse embryos, and the mice with this sequence were “serpentized,” meaning they had no limbs. The critical enhancer could be boiled down to 17 bp of highly conserved sequence that, when reintroduced into the snake enhancer sequence, restored limb development to the snake-like mice that are now inhabiting my dreams. Kvon et al. (2016). Cell 167, 633–642.e11. Although it is one of the most common autosomal-dominant conditions in humans, familial hypercholesterolemia (FH) often goes unrecognized. This is particularly disappointing given that treatment improves outcomes. Because of the under-diagnosis, many have advocated for cascade screening within families, wherein disease in an index subject triggers additional genetic testing or cholesterol screening within the family. Usually, this index subject is an adult. Wald et al. explored the idea that they might have better screening success if they instead started with kids. In a population of more than 10,000 children in Britain, they piggy-backed a cholesterol screen and DNA testing onto a scheduled childhood immunization visit. According to their definition of a positive screen, they found an overall FH prevalence twice as high as usually quoted: 1 in 273 instead of 1 in 500, including 17 kids who had an FH mutation but did not have total cholesterol above their screen cutoff. One advantage of the approach is that parents of babies tend to be fairly young, and treatment for FH could have a bigger impact when started early. Indeed, the parents of the affected children in this study were subsequently screened, and of the 28 considered positive, none were receiving treatment with statins when they were identified, but 90% did after the screen. Wald et al. (2016). N. Engl. J. Med. 375, 1628–1637. It’s easy to reduce chromosomal duplications to the same effect—a potential increase in gene dosage. Lacking a gene, we often don’t know what to make of these copy-number variants. In the process of dissecting the duplications in one particular genomic region, Franke et al. showed that they could affect chromatin structure in very different ways by adding or blocking a regulatory unit. They started with the observation that different duplications in the SOX9 region could yield strikingly different phenotypes, from sex reversal to limb malformations to no phenotype. Using assessments of chromatin conformation coupled with mouse models, they realized that the key to these differences was the effect of each duplication on the discrete regulatory units in the region, which are called topologically associated domains, or TADs. As it turns out, these TADs have internal stability and can be shuffled around as a result of the duplications, thereby having the potential to create new regulatory genomic regions that the authors dub neo-TADs. The effect of each duplication depended on it size and position within the regulatory domain. The duplication with no phenotypic effect actually spanned from one TAD to the next and created a novel TAD that had distinct boundaries, isolating its effects from affecting the rest of the region. Franke et al. (2016). Nature 538, 265–269. A big challenge with genomic data is the difficulty of sorting through the reams of data to focus attention on the variants most likely to be relevant. Several bioinformatics programs use conservation and the predicted effect of a substitution on the protein to assess the likelihood that a variant is detrimental, but most of these are notoriously bad with their predictions. Jagadeesh et al. developed a new approach called the Mendelian Clinically Applicable Pathogenicity (M-CAP) score, which incorporates several previously available variant scoring methods as well as multiple new features. They trained the method by using millions of benign variants from the ExAC Browser and thousands of pathogenic variants from HGMD. Application to clinical exomes suggested that M-CAP can correctly filter out 60% of rare, missense variants while retaining most known pathogenic variants in the data. Jagadeesh et al. (2016). Nat. Genet. Published online October 24, 2016. . The drug ataluren is in preclinical and clinical trials for the treatment of a variety of disorders caused by nonsense variants. Although it has been unclear how, ataluren allows cells to blow through these premature stop codons, thereby allowing production of a full-length protein. In a variety of mammalian and yeast models, Roy et al. carefully dissected the protein products in cells treated with ataluren and found that the cells all seemed to use similar near-cognate tRNAs that recognized the stop codons, which allowed incorporation of an amino acid at the truncating codon rather than termination of translation. The preferred tRNAs suggest that nonstandard base pairing occurs with the codon. Further supporting a direct mode of action of ataluren on the ribosome, increasing amounts of tobramycin, a drug that interacts with the ribosomal A site, interfered with ataluren activity. Roy et al. (2016). Proc. Natl. Acad. Sci. USA 113, 12508–12513. Methods of uncovering genetic associations most commonly seek variation that plays a role in a single phenotype at a time. We know that certain genetic variants are actually important for the regulatory control of multiple genes and could affect multiple phenotypes at once, but we miss these important regulatory elements when we use the single-phenotype-based approaches. Some groups have developed ways of incorporating information on multiple phenotypes at once, but most do not control for population structure, a known source of false positives in genetic association studies. Joo et al. recently published a method they call GAMMA, which does both: it is applicable to multiple phenotypes simultaneously, and it corrects for population structure. In simulated and real data, GAMMA identified genetic associations and apparently did so with fewer false positives due to population structure than did a competing approach. Joo et al. (2016). Genetics. Published online October 21, 2016. .