Workshop: From the coding to the noncoding genome and its implications

2012-01-19: University of Copenhagen, at 12:30, lecture room A2-83.01, Thorvaldsensvej 40 at Frederiksberg campus.

Registration is not necessary. Refreshments, coffee and sandwiches (first come, first served) will be served.


12:30-12:35 Welcome
Jan Gorodkin, University of Copenhagen.

12:35-13:20 Identification of Type 1 diabetes-associated DNA methylation variable positions
Bernhard Boehm, University of Ulm

DNA methylation is a heritable epigenetic modification process whereby CpG dinucleotides are methylated at the C5 position of cytosine. Changes in the methylation pattern can result in gene expression differences. We hypothesized that epigenetic variation could play a role in the aetiology of T1D.  DNA methylation profiles from T1D-discordant MZ twin pairs and from singletons before and at T1D diagnosis were generated using methylation array technology. Epigenetic differences were found in HLA genes, as well as genes that regulate inflammation and apoptosis. Temporal origins of T1D-associated methylation variable positions suggest that these variations may arise very early in the etiological process of T1D.

13:20-14:05 High-throughput methods for determining the specificity of transcription factors
Gary Stormo, Washington University Medical School

Several high-throughput methods have been developed in recent years for determining the specificity of transcription factors. This has greatly increased our knowledge of specificity for many transcription factor families. It allows us to quantitatively assess the complexity of the models needed to accurately predict binding affinities which further allows us to gain insight from the discrepancies between in vitro binding specificities and in vivo location analyses. It also also for the development of more accurate recognition models, facilitating the design of proteins with novel specificity.

14:05-14:30 Break

14:30-15:15 Structural Evolution of Long Non-Coding RNA
Peter Stadler, University of Leipzig

Several recent studies have demonstrated that there are more than 10000 well-defined mRNA-like long non-coding RNAs produced from the human genome. As a group, they have been found to be under significant but weak stabilizing selection. Their evolutionary age, however, remains uncertain. We use conservation patterns of splice sites to show that more than 50% of human linc RNAs originated in the stem lineage of placental mammals and a significant fraction is even considerable older. At the same time, even ancient lncRNAs exhibit a high level of turnover of individual splice sites, thereby complication considerably the comparative analysis of their sequences.

15:15-16:00 The hidden layer of RNA regulation in development and cognition
John Mattick, The University of Queensland

It appears that the genetic programming of complex organisms has been misunderstood because of the assumption that most genetic information is transacted by proteins.  The emerging evidence suggests that RNA is the computational engine of development, by directing generic effector proteins to their sites of action, and the substrate for gene-environment interactions, especially in the brain.

16:00-16:30 Chromosomal breakpoints targets all functional regions of the human genome
Niels Tommerup, University of Copenhagen

In the first unbiased long-term clinical follow-up of prenatally diagnosed carriers of de novo balanced chromosomal rearrangements, we observed a significant (~20%) morbidity in carriers, a 2-3 fold increase compared to previous estimates. The morbidity exclusively involved neurodevelopmental, cognitive, and psychiatric disorders. By mate-pair sequencing we identify and confirm some of the targets for these disorders, including coding genes like NPAS3, but also long (intergenic) non-coding RNA genes (lincRNA/lncRNA), exemplified by the canonical lincRNA HOTAIR, as well as unannotated transcriptional regions identified by RNA-Seq. Our study illustrates that a systematic mapping of chromosomal breakpoints by mate-pair sequencing would have the power to saturate a wide range of functional compartments of the human genome with mutations, with or without associated phenotypes.