Genomic disorders: complexity at multiple levels

A report on the 7th Wellcome Trust Genomic Disorders conference held at the Homerton College, Cambridge, UK, April 10-12, 2013.


Sixty years of DNA and the Cambridge connection
Th e race to discover the structure of DNA was brought to life by Jim Lupski (Baylor College of Medicine, USA) and Helen Firth (Wellcome Trust Sanger Institute, UK) at the outset of the meeting. Lupski reviewed the circumstances at that time, explained the competition from Linus Pauling's group and described the Nature papers that report the structure of DNA. Firth gave a wonderful overview and a virtual tour of Cambridge, illustrating the links between the pioneers of molecular biology and various colleges and laboratories in Cambridge, the BRCA2 sequence stripes along a cycle path in Cambridge shire, and the history and evolution of genomics in Cambridge. In a lighter vein, Firth also showed images of the Eagle pub where Watson and Crick, 60 years ago, interrupted customers to announce that they had discovered the secret of life!

Genome sequencing and annotation
Our ability to understand the genome has increased by leaps and bounds in tandem with our increased ability to sequence DNA accurately at high speed. Yet, as pointed out by Richard Durbin (Wellcome Trust Sanger Institute, UK), we do not completely comprehend the function of genes and a vast amount of biology is yet to be understood. However, our knowledge of the genome has improved, with some surprising insights from results of the Encyclopedia of DNA Elements (ENCODE) project. Th e ENCODE project involves an international consortium of investigators studying the functional elements in the human genome, including those regulating expression and activity of genes. Th e ENCODE project coordinator, Ewan Birney (European Molecular Biology Laboratory -European Bioinformatics Institute, UK), noted one interesting outcome from this project: about 99% of the genome is within 1.7 kb of a biochemical event (such as RNA activity, protein binding or a specifi c chromatin structure) and about 95% of the genome is within 10 kb of a functional sequence motif or a regu latory footprint.
Lupski then returned for a second presentation, this time focusing on his own research on copy number variation, individual genomes, and challenges associated with the interpretation of genetic fi ndings in a clinical setting. He argued that most sites of variation that contribute to disease are rare (<0.1% frequency in the population), recent in origin within a family lineage, and under strong negative selection. A number of large-scale sequencing studies were presented and most of these talks centered on the interpretation of low coverage sequencing data to understand the etiology of complex disease. Gonçalo Abecasis (Center for Statistical Genetics, University of Michigan, USA) spoke about overlapping hits in genome-wide association studies relevant to comorbid cardiovascular and retinal degeneration pheno types, suggesting how disparate disorders can have similar genetic etiologies. Nicole Soranzo (Wellcome Trust Sanger Institute, UK) presented the objectives of the multi-center UK10K project. The UK10K project has now sequenced to six times coverage about 4,000 individuals from deeply phenotyped cohorts. Preliminary results presented by Soranzo showed a high (>90%) concordance for rare variants compared with the 1000 Genomes data. Chris Tyler-Smith (Wellcome Trust Sanger Institute, UK) estimated that each individual carries approximately 100 loss-of-function variants, but cautioned that population distribution and haplotype background must be considered before labeling a genetic change as pathogenic.
Exome-sequencing data presented by several groups estimated that a diagnosis could be made in 30% of cases. Even so, it is clear that a large portion of the genetic basis of human disease is yet to be uncovered. Some of it might be accounted for by gene interactions and the effect of epigenetic and regulatory elements. Ben Lehner (European Molecular Biology Laboratory -Centre de Regulació Genòmica, Spain) presented exciting results from Caenorhabditis elegans experiments showing how some gene mutations modify the effect of mutations of other genes. His captivating presentation featured time-lapse movies profiling expression patterns of tbx-8 in response to mutations in tbx-9, an ancestral gene duplicate, or changes in the molecular chaperone heat shock protein 90.
Talks by Wolf Reik (Babraham Institute, UK) and John Mattick (Garvan Institute of Medical Research, Australia) outlined how the human genome and epigenome can be altered to express genes at different stages of development in a temporal, spatial and parent-of-origin-specific manner. Reik's presentation was on DNA methylation dynamics during development; he noted that the rate of erasure of methylation marks is the same in males and females but reprogramming is faster in males than females. Mattick's talk was about the 'information that programs our complexity' . He illustrated this concept using long non-coding RNA, which add multiple orders of complexity to the genome, but also impressed on us the ubiquitous role of non-coding RNAs in human development and disease.

Making sense of the sequence and then what?
Today's genomic science is fueled by genome sequencing efforts, but interpreting which variants contribute a causative role in disease has been challenging. JT den Dunnen (Leiden Genome Technology, Netherlands) and Jawahar Swaminathan (Wellcome Trust Sanger Institute, UK) discussed systematic approaches to the generation of repositories to store, visualize and interpret data. The ethics of the centralization of data, the need for consent and the sensitivity of information were also discussed. Les Biesecker (National Institutes of Health, USA) presented the ClinSeq study, an excellent predictivemedicine model for iterative hypothesis-generating research where participants can be re-contacted, retested, phenotyped, and results returned. Robert Green (Brigham and Women's Hospital, USA) spoke about the disclosure of primary and incidental risk information, a new challenge encountered in genomic medicine. Han Brunner (Radboud University, Netherlands) stimulated post-session discussions centered on attributing causality, phenotyping, and integration of genomic science into clinical medicine.

Conclusions
It is becoming increasingly clear that the practice of genomics requires concerted collaborative effort from clinicians, molecular biologists, and computational scientists to address the current challenges in the discovery, interpretation, and application of genomic data to health care. This conference was a one-stop-shop for everyone who deals with sequence; sessions comprised a motley crew of speakers from related yet diverse interests. It was a great conference, linking the past 60 years to the present and to a promising future for genomic medicine.