NEW YORK (GenomeWeb News) – A 23andMe team this week reported findings from a genome-wide association study of hypothyroidism in PLoS ONE.
Using data for 3,736 individuals included in the 23andMe database with self-reported hypothyroidism and more than 35,500 unaffected controls, the investigators found five genome-wide significant associations with hypothyroidism. Three of the five variants detected — including a SNP in the human leukocyte antigen class I region — had previously been linked to autoimmune conditions. Another turned up near the immune gene VAV3, and the fifth fell near FOXE1, a thyroid development gene implicated in hypothyroidism before.
A smattering of other SNPs showed suggestive associations in the study, including variants in the HLA class II region and genes believe to influence thyroid stimulating hormone levels.
"These findings shed new light on the biology of hypothyroidism, showing that genes involved with both thyroid function and immune response impact this disease," corresponding author Nicholas Eriksson, said in a statement. "This demonstrates our ability to harness the power of the enormous 23andMe genetic database to further medical research."
In Nature Genetics, investigators from Mount Sinai School of Medicine, the University of Washington, and Rosetta Inpharmatics describe their computational strategy for coming up with SNP genotype information using RNA transcript data.
The team generated gene expression and genotype profiles for liver and/or fat tissue samples from hundreds of European-American individuals. Using expression quantitative trait loci information from these samples, they then developed an algorithm that could decipher SNP genotypes at 1,000 sites in the genome based on RNA transcript levels. The technique could reliably identify individuals from their gene expression profiles between about 90 and 99 percent of the time, researchers reported, depending on whether the expression data was assessed using eQTL information ascertained for the same tissue type or for a different tissue type.
"By observing RNA levels in a given tissue, we can infer a genotypic barcode that uniquely tags an individual in ways that enables matching the individual to an independently derived DNA sample," co-corresponding author Eric Schadt, genetics and genomics chair at Mount Sinai School of Medicine and director of its Institute for Genomics and Multiscale Biology, said in a statement.
Researchers from the US, Italy, and Austria used Affymetrix SNP 6.0 arrays to track down prostate cancer-associated copy number alterations on chromosomes 12 and 15 — work that they describe online in the Proceedings of the National Academy of Sciences.
The team tested blood samples from more than 1,900 men participating in the Tyrol Prostate Specific Antigen Screening study. Of these, 867 men had been diagnosed with prostate cancer and 1,036 had not, though all of the participants had higher-than-usual prostate specific antigen levels.
From the thousands of CNVs detected in case and control genomes, researchers whittled their list of candidate variants down to two CNVs: one deletion affecting the glycosyltransferase gene MGAT4C on chromosome 12 and another involving a non-coding site on chromosome 15. After verifying the associations in hundreds more prostate cancer cases and controls, investigators went on to explore some of the functional consequences of these risk deletions.
Also in PNAS, a look at ultraconserved elements in plant and animal genomes from a University of Missouri and University of California, Berkeley team that developed a new alignment-free computational method to catalog so-called long interspersed multispecies elements, or LIMEs, in half a dozen animal genomes and as many plant genomes.
In the human, macaque, mouse, rat, chicken, and dog genomes, investigators found that most stretches of identical, ultraconserved DNA that were shared between two or more animal genomes were also syntenic, turning up in the same place in the genome. On the other hand, in the six plant genomes examined, the team saw only non-syntenic LIMEs. The plant LIMEs also had slightly different complexity and composition patterns than those detected in animals, prompting speculation about the distinct mechanisms that produced these ultraconserved sequences in plant compared to animal genomes.
"[D]espite the common presence of LIMEs in both animal and plant lineages, the evolutionary processes involved in the creation and maintenance of these elements differ in the two groups," University of Missouri at Columbia computer science and informatics researcher Dmitry Korkin, the study's corresponding author, and colleagues wrote, "and are likely attributable to several mechanisms, including transfer of genetic material from organellar to nuclear genomes, de novo sequence manufacturing, and purifying selection."
As part of a special section on computational biology in Science this week, a trio of researchers from the University of California at Santa Cruz reflect on the computational strategies that are being used to integrate various forms of genomic information — from assembling aligning genome sequences to interpreting genetic, epigenetic, and proteomic data for individuals and populations.
"As genomic sequencing projects attempt ever more ambitious integration of genetic, molecular, and phenotypic information, a specialization of genomics has emerged, embodied in the sub-discipline of computational genomics," corresponding author David Haussler and colleagues wrote.
"Models inherited from population genetics, phylogenetics, and human disease genetics merge with those from graph theory, statistics, signal processing, and computer science to provide a rich quantitative foundation for genomics that can only be realized with the aid of a computer."
Genomics In The Journals is a weekly feature pointing readers to select, recently published articles involving genomics and related research.