High throughput, massively parallel DNA sequencing provides a powerful technology to study the human genome and to identify variations in DNA that cause disease. Sequencing the protein coding region of the genome (`whole-exome sequencing') is a cost effective method to search the part of the genome that is most likely to harbor disease related mutations.

We developed software methods to process sequencing data and to annotate variants with data on genes, function, conservation, expression, diseases, pathways, and protein structure. We applied whole-exome sequencing to search for the molecular basis of disease in three projects: 1) a cohort of patients with congenital diarrheal disorders (CDDs); 2) a cohort of patients with congenital chronic intestinal pseudo-obstruction (CIPO) or the related disease, megacystis-microcolon-intestinal hypoperistalsis syndrome (MMIH); and 3) four siblings with infantile pontocerebellar hypoplasia and spinal motor neuron degeneration.

We sequenced 45 probands from diverse ethnic backgrounds who were diagnosed with a variety of CDDs of probable, but unknown genetic cause. Patients had been diagnosed with generalized malabsorptive diarrhea, selective nutrient malabsorption, secretory diarrhea, and infantile IBD. We found homozygous or compound heterozygous mutations, 25 of them novel, in genes known to be associated with CDDs in 27 cases (60%). The genes implicated were ADAM17, DGAT1, EPCAM, IL10RA, MALT1, MYO5B, NEUROG3, PCSK1, SI, SKIV2L, SLC26A3, and SLC5A.

With whole-exome sequencing in a cohort of 20 patients with congenital CIPO or MMIH, we identified a subset of 10 cases with potentially damaging de-novo dominant acting mutations at highly conserved loci in the ACTG2 gene, encoding actin, gamma-enteric smooth muscle precursor, a protein essential to the functioning of muscle cells in the intestinal wall.

By exome sequencing, we discovered rare recessive mutations in EXOSC3 (encoding exosome component 3) that were responsible for pontocerebellar hypoplasia and spinal motor neuron degeneration in the four probands, and identified identical and additional novel mutations in a large percentage of other children with the same disorder.

In conclusion, we demonstrated that whole-exome sequencing is an effective approach for the identification of casual mutations in that may escape detection with standard practice involving a complex diagnostic workup and targeted gene sequencing.

Over the past decade, decreases in the cost of DNA sequencing has allowed for a surge in the amount of data being generated. This has led to the discovery of genes causal for hundreds of Mendelian disorders and genes associated with many complex disorders. Given the opportunity to use sequencing data to tackle neuropsychiatric diseases with a complex genetic architecture, I take a data-first approach to study two diseases, bipolar disorder and autism spectrum disorder (ASD). Combining information gleaned from next-gen sequencing (NGS) data with the latest analytic methods shines new light on the biology of these diseases.

In the first part of this dissertation, I present a whole-exome analysis of nine affected individuals from four families in which bipolar disorder was transmitted over several generations, and six unrelated, affected individuals. Our results demonstrate the genetic heterogeneity of bipolar disorder and provide support for rare-variant oligogenic disease model. In the second part, I present an approach to identify rare variants associated with autism spectrum disorder (ASD) in a whole-genome sequencing study of 71 individuals diagnosed with ASD and their family members. I demonstrate that by incorporating knowledge of population-wide variant frequencies to analyses of NGS data and taking an approach sensitive to complex family structures, as opposed to utilizing only case-control or trio data, one can identify patterns that would otherwise have been missed and thus gain novel insights into disease etiology.

Finally, I present a mutational burden dataset called SORVA (Significance Of Rare VAriants), which is useful in vetting candidate variants and genes from NGS studies. In effect, my studies of complex disorders using next-gen sequencing show the field is constantly evolving with new computational approaches allowing for many advances being made in the areas of psychiatric disorders and ASD, in particular.

Background

Gastric adenocarcinoma is a rare diagnosis in childhood. A 14-year old male patient presented with metastatic gastric adenocarcinoma, and a strong family history of colon cancer. Clinical sequencing of FAP and APC1 were negative. Whole exome sequencing was used to capture the majority of protein-coding regions for the identification of single-nucleotide variants, small insertion/deletions, and copy number abnormalities in the patient's germline as well as primary tumor.

Materials and Methods

DNA was extracted from the patient's blood, primary tumor, and the unaffected mother's blood. DNA libraries were constructed and sequenced on Illumina HiSeq2000. Data were analyzed with the Genome Analysis Toolkit. Variants were annotated using an in-house Ensembl-based program. Copy number was assessed using ExomeCNV.

Results

Each sample was sequenced to a mean depth of coverage of greater than 120x. A rare non-synonymous coding SNV in TP53 was identified in the germline. There were 10 somatic cancer protein-damaging variants. ExomeCNV comparing tumor to the patient's germline, identified abnormal copy number, spanning 6,946 genes.

Conclusion

We present an unusual case of Li-Fraumeni detected by whole exome sequencing. There were also likely driver somatic mutations in the gastric adenocarcinoma. These results highlight the need for more thorough and broad scale germline and cancer analyses to accurately inform

patients of inherited risk to cancer and to identify somatic mutations.

Duchenne muscular dystrophy (DMD) is a progressive muscle wasting disease caused by mutations in the X-linked DMD gene. These mutations result in a loss of expression of the dystrophin protein, which provides structural support to the muscle cell membrane and protects it from contraction induced damage. Without dystrophin, a cyclical process of skeletal muscle degeneration and regeneration occurs until it is ultimately replaced with adipose cells and fibrotic tissue, rendering it non-functional. The most critical dystrophin domains are the N-terminus, which binds to cytoplasmic filamentous actin and connects it to the extracellular matrix through C-terminus binding to β-dystroglycan, and by extension the associated dystrophin-glycoprotein complex. An allelic disease, Becker muscular dystrophy, is caused by in-frame mutations within the dispensable central domain and manifests as a less severe phenotype.

The most promising DMD therapy is antisense oligonucleotide approaches in which single exons adjacent to the mutation are targeted for pre-mRNA removal by the splicing machinery to restore the mRNA reading frame. This has been successful in cell culture, animal models, and now has shown promise in clinical trials. However, rescued dystrophin protein levels remain variable suggesting the need for improvements in exon skipping efficiencies. We performed a high-throughput screen to identify and repurpose FDA approved drugs that potentiate this antisense exon skipping strategy. We found that one drug, dantrolene, enhanced exon skipping, rescued dystrophin protein levels, and improved overall function in a DMD mouse model. In reprogrammed patient myotubes, dantrolene, and other Ryanodine Receptor (RyR1) antagonists increased DMD exon 51 skipping, suggesting the RyR1 calcium channel as the relevant molecular target. An independent high-throughput screen found a cohort of small molecules with similar 2-D structures that enhanced DMD exon 51 skipping and shared a known protein target, calmodulin. Identified AO potentiating small molecules bind and inhibit either calcium channels (RyR1) or calcium regulatory proteins (Calmodulin) suggesting the importance of Ca2+ regulation and its impact on Ca2+-binding proteins in directing exon skipping activity

Antisense-based exon skipping is highly sequence specific and benefits from the availability of diverse genetic mutations for the evaluation of potential therapies or molecular mechanisms. We have collected and derived 49 DMD patient fibroblasts and have reprogrammed a subset into muscle cells as a resource for the research community. In summary, we have established a research environment facilitating the discovery of novel therapies for the treatment of DMD, and have identified Ca2+ modulation as a key regulator of antisense mediated splicing activity.

Background

Case presentation

Conclusion