Exome Sequencing – Research and Clinical Applications Karthik Kota, David Jenkins, John Seed, Justin Johnson

November 2012

EdgeBio 201 Perry Parkway, Suite 5 Gaithersburg, MD 20877 (800) 326-2685 www.edgebio.com 1

Exome Sequencing – Research and Clinical Applications Recent developments in next generation sequencing (NGS) technologies are rapidly changing the scope of genetics, genomics and even medicine. There is widespread interest in both the research and clinical settings to apply NGS platforms to selected regions of the human genome. Sequencing of targeted regions or coding regions of the human genome, called the exome has lead to the discovery of relevant and interesting genetic variations and is swiftly replacing other approaches that focus on specific regions for gene discovery and clinical testing. There are several commercially available capture kits that focus on different target regions, so care must be taken when choosing the appropriate kit for a specific experiment [ Refer to EdgeBio Targeted Sequencing V - Exome Sequencing Cheat Sheet white paper]. Genetic discoveries have been informative in basic disease mechanism research to model and understand the causes of several diseases. Even though traditional positional cloning approaches have resulted in the identification of causal mutations for several diseases, many complex family associated diseases cannot be fully understood with these techniques alone1. Exome sequencing utilizes DNA-enrichment methods and massively parallel nucleotide sequencing to identify all of the protein-coding variants in the genome. By limiting the scope of the experiment to the protein-coding sequences, only about 5% of the human genome is sequenced. Together with growing public databases of known variants, exome sequencing allows for identification of genetic mutations and risk factors in samples that were deemed insufficiently informative for previous genetic studies. Not only does exome sequencing enable identification of mutations in families that were undetectable with linkage disequilibrium and positional cloning methods, but it is also more rapid and less expensive. The use of exome sequencing has been successful in the characterization of many rare diseases2. A few such cases are described in this article.

Exome Sequencing to identify and treat unknown diseases in 4 year olds In 2011, a team of scientists and physicians at the Medical College of Wisconsin and Children's Hospital of Wisconsin used exome sequencing to help identify and treat3 an unknown intestinal disorder in a 4 year old boy, Nicolos. Nicolos had been subject to prolonged medical examination and several surgeries, but the doctors were unable to pinpoint a specific disease. Exome sequencing was considered after traditional methods were unable to diagnose Nicolos’ disease. After exome sequencing and analysis, the physicians were able to develop a treatment plan using a cord blood transplant that was able to stop the course of the disease. Realizing the potential of genomic sequencing in establishing the correct diagnosis for patients and improving the overall outcome, The Medical College of Wisconsin and Children's Hospital of Wisconsin are developing a new strategy with formal policies and procedures for all future cases in which genetic sequencing will be used as a diagnostic tool. In another similar case4 , the nonprofit organization Rare Genomics Institute was able to facilitate exome sequencing of another 4-year old, Maya Neider. Maya suffered from global developmental delays which were believed to be genetic and might have occurred sometime during fetal development and early childhood. Her condition has caused impairment in her speech, hearing, and running. Traditional diagnosis along with several operations had not lead to a definite explanation of her condition. RGI’s initiative to cure rare genetic diseases helped raise the cost of exome sequencing within 6 hours using crowdfunding by 50 donors from across the country. The sequencing and data analysis 2

performed by Yale researchers suggested a spontaneous mutation in a gene not previously known to affect development in children. Although the mutation may not be the only cause of Maya’s conditions, this story has provided more insights for the doctors to understand her disease and could present myriad opportunities to address future cases like this. Maya is one of ten children with rare diseases who are part of RGI’s pilot program to test out the effectiveness of the crowdsourced funding for sequencing.

Exome sequencing identifies the cause of a Mendelian disorder

Figure 1: Exome Sequencing and filtering strategy Nature Genetics 42, 13–14 (2010)

Many studies have recently been published that show the application of exome sequencing to the discovery of the basis of rare Mendelian disorders. A recent study combined whole exome sequencing with a filtering methodology to identify the underlying gene causing a Mendelian disorder. The study also correctly identified a gene previously known to be responsible for the Freeman-Sheldon syndrome5. The researchers used extensive parallel sequencing of exomes from a short set of affected individuals and then filtered out benign and unrelated variants to arrive at a final set of causal variants. Researchers using this strategy reported a causal variant in an uncharacterized Mendelian disorder, Miller syndrome. Miller syndrome, also known as postaxial acrofacial dysostosis, is a rare malformation syndrome with symptoms including cleft palate, absent digits and ocular anomalies among others. The identification of the gene mutated in such disorders will allow for improved diagnosis and act as a starting point for biological investigations. The real advantage of these studies is to show that this approach can be used to successfully characterize the genetic basis of rare monogenic disorders. In another study6, researchers sequenced the exomes of four individuals including two siblings with Miller syndrome. The researchers used exome sequencing to cover the target regions at an average coverage of 40x. A stepwise filtering approach was used to screen the identified variants in order to select those likely to be associated with the disorder (Figure 1). Initially the researchers screened for genes that contained non-synonymous variants, splice site mutations and coding indels. The common 3

variants were then filtered out by comparing the four exomes to eight control HapMap individuals and to the dbSNP database. Finally, the researchers removed any benign variants predicted by PolyPhen. Both a dominant and recessive model of inheritance was tested, but Millers syndrome was determined to follow the recessive model which met the researcher’s expectations. Eight candidate genes were identified under the dominant model and one DHODH gene was identified under a recessive model. The four individuals with Miller syndrome were found to have six rare variants in DHODH. The excitement surrounding this work emphasizes the utility of exome sequencing in identifying genes associated with Mendelian disorders. A 2012 study7 by Duke scientists highlighed the success rates of NGS in a clinical setting along with some key challenges. The authors presented a study of whole exome sequencing of twelve parent-child trios in which the children had a combination of congenital anomalies, facial dimorphisms and/or intellectual disabilities. The anomalies were determined to be due to genetic conditions. In all cases both parents had no history of any prior genetic abnormality. Unlike many other studies the researchers did not seek patients with similar phenotypes. Instead, the researchers followed a series of carefully sketched predetermined criteria that were approved by the Duke Review board. The screens were designed to identify influential genotypes that may be responsible for the child’s condition and prioritized them into categories such as recessive variants, X-linked variants, putative de novo variants and compound heterozygotes. Five genes -TCF4, EFTUD2, SCN2A, NGY1 and SMAD4, were found to be associated with a disease in the study. This resulted in a likely genetic diagnosis for six of the twelve families. Particularly interesting was the observation of simultaneous occurrences of two mutations in the same gene, EFTUD2. This scenario was found to occur in two unrelated family trios with good statistical significance, suggesting the possibility of a causal Mendelian disease gene that was agnostic at the time of the study.

Role of exome sequencing in Neurodevelopmental Disease Another study8 focused on the advances of exome sequencing by identifying novel genes in 118 subjects who were diagnosed with neuro-developmental disease from birth. None of these diseases could be associated with previously known causes. Several genes were discovered such as EXOC8 associated with Joubert syndrome, GFM2 for microcephaly, and 22 other genes that were not previously associated with any other diseases. Exome sequencing helped identify mutations in ten subjects in the study cohort that were known to cause a different disease from the earlier diagnosis. These mutations were then confirmed to be responsible for the disease, which helped change the original diagnosis and alter patient care. This research further establishes exome sequencing as a useful resource for genetic discovery. Several other rare genetic diseases that were caused by a genetic mutation were also identified and published in the recent years through exome sequencing. These cases include severe brain malformations (WDR62)9 , recessively or dominantly inherited autoimmune lymphoproliferative syndrome (FADD)10 , glucose homeostasis and adiposity (ADIPOQ)11 , fatal classic Kaposi sarcoma (STIM1)12 , Sensenbrenner syndrome (WDR35)13, complex I deficiency (ACAD9)14, Brown-Vialetto-van Laere syndrome (C200rf54)15, familial ALS (VCP)16 , hyperphosphatasia mental retardation (PIGV)17, Fowler syndrome (FLVCR2)18 , familial combined hyperlipidemia (ANGPTL3)19, Kabuki syndrome (MLL2) 20 , non-syndromic hearing loss (GPSM2)21 and spinocerebellar ataxias (TGM6)22.

4

Limitations Although there are many examples where the cause of a rare genetic disease was identified rapidly with the help of CLIA-certified exome sequencing centers, there are limitations to this type of analysis as well. Exome sequencing can only identify variants within the regions included in the exome capture kit. Any disease causing or relevant variants outside of the capture region cannot be detected. Additionally, repeats present in the target regions may be difficult to cover at sufficient coverage to confidently call variants. Exome sequencing also struggles to cover target regions falling towards the ends of chromosomes and centromeres23. Additionally, mitochondrial genes are not included in exome sequencing kits. Another limitation of exome sequencing is that every capture kit available for purchase has major target regions unique to that kit. Researchers need to be aware if the regions in which they are interested are sufficiently covered by their chosen capture kit. Exome kits are also not a complete solution, as not all genes are included in the target regions. For instance, mutations in some microRNAs are known to cause cancers such as in the pancrease, but the more than 1,000 microRNAs in the human genome are not included in any commercially available exome kit. A few medically important genes such as CACNA1C, a biopolar disorder causing gene candidate, and MLL2, a gene implicated in leukemia, are not targeted by popular kits because they are not yet annotated as consensus coding sequence (CCDS)24. Finally, exome sequencing cannot be used to identify structural variants. Despite these limitations, exome sequencing is being widely adopted by many researchers. An abundance of mutation discovery is expected to enable the discovery of not only rare causal variants, but also protein-coding risk variants. This method has application in both the research and clinical arenas and sets the scene for the use of whole-genome sequencing. Researchers interested in exome sequencing should invest time to understand the limitations of this technology. As the significance and success of exome sequencing is increasing rapidly in clinical studies, it is important to begin to implement this technology to advance patient health, improve medical care and explore challenges associated with the discovery causal variants.

References 1

Exome sequencing: a transformative technology. Singleton B. A. The Lancet Neurology, Volume 10, Issue 10, Pages 942-946 2

Exome sequencing makes medical genomics a reality. Biesecker L.G. Nature Genetics 42, 13–14 (2010) doi:10.1038/ng0110-13 3

Making a Definitive Diagnosis: Successful clinical application of whole exome sequencing in a child with intractable inflammatory bowel disease. Worthey E., Mayer A.N. et. al. 4

Neider’s blog http://niederfamily.blogspot.com/2010/06/uncommon-sense-title.html

5

Targeted capture and massively parallel sequencing of 12 human exomes. Ng S.B., Turner E.H.,et. al.

6

Exome sequencing identifies the cause of a mendelian disorder. Ng, S. B., K. J. Buckingham, et al. Nature Genetics 42(1): 30-5. 7

Clinical application of exome sequencing in undiagnosed genetic conditions. Need A.C., Shashi V., et. al. J Med Genet doi:10.1136/jmedgenet-2012-100819

5

8

Exome sequencing can improve diagnosis and alter patient management. Dixon-Salazar TJ, et al. Transl. Med. 4, 138ra78 (2012). 9

Whole-exome sequencing identifies recessive WDR62 mutations in severe brain malformations. Bilguvar, K., A. K. Ozturk, et al. Nature 467(7312): 207-10. 10

Whole-exome-sequencing-based discovery of human FADD deficiency. Bolze, A., M. Byun, et al. The American Journal of Human Genetics 87(6): 873-81. 11

Molecular basis of a linkage peak: exome sequencing and family-based analysis identify a rare genetic variant in the ADIPOQ gene in the IRAS Family Study. Bowden, D. W., S. S. An, et al. Hum Mol Genet 19(20): 4112-20. 12

Whole-exome sequencing-based discovery of STIM1 deficiency in a child with fatal classic Kaposi sarcoma. Byun, M., A. Abhyankar, et al. J Exp Med 207(11): 2307-12. 13

Exome sequencing identifies WDR35 variants involved in Sensenbrenner syndrome. Gilissen, C., H. H. Arts, et al. The American Journal of Human Genetics 87(3): 418-23. 14

Exome sequencing identifies ACAD9 mutations as a cause of complex I deficiency. Haack, T. B., K. Danhauser, et al. Nat Genet 42(12): 1131-4. 15

Exome sequencing in Brown-Vialetto-van Laere syndrome. Johnson, J. O., J. R. Gibbs, et al. The American Journal of Human Genetics 87(4): 567-9; author reply 569-70. 16

Exome sequencing reveals VCP mutations as a cause of familial ALS. Johnson, J. O., J. Mandrioli, et al. Neuron 68(5): 857-64. 17

Identity-by-descent filtering of exome sequence data identifies PIGV mutations in hyperphosphatasia mental retardation syndrome. Krawitz, P. M., M. R. Schweiger, et al. Nat Genetics 42(10): 827-9. 18

Unexpected allelic heterogeneity and spectrum of mutations in Fowler syndrome revealed by next-generation exome sequencing. Lalonde, E., S. Albrecht, et al. Hum Mutat 31(8): 918-23. 19

Mutations in the ANGPTL3 gene and familial combined hypolipidemia: a clinical and biochemical characterization. Minicocci I., Montali A., Robciuc M.R., et al. J Clin Endocrinol Metab. 20

Exome sequencing identifies MLL2 mutations as a cause of Kabuki syndrome. Ng, S. B., A. W. Bigham, et al. Nat Genet 42(9): 790-3. 21

Whole exome sequencing and homozygosity mapping identify mutation in the cell polarity protein GPSM2 as the cause of nonsyndromic hearing loss DFNB82. Walsh, T., H. Shahin, et al. The American Journal of Human Genetics 87(1): 90-4. 22

TGM6 identified as a novel causative gene of spinocerebellar ataxias using exome sequencing. Wang, J. L., X. Yang, et al. Brain 133(Pt 12): 3510-8. 23

Scientific American blogpost http://blogs.scientificamerican.com/guest-blog/2012/05/16/10-things-exomesequencing-cant-do-but-why-its-still-powerful/ 24

A comparative analysis of exome capture. Parla, J.S, Iossifov Ivan, et al. Genome Biology , 12:R97

6