I’m an avid supporter of large-scale genomics studies, which have yielded some of the most important resources in human genetics and cancer genomics. With my new focus on inherited disorders that affect children, I have a new appreciation for initiatives that tackle not-so-common diseases.

One such effort is the UK-based Deciphering Developmental Disorders (DDD) study, which recruited 13,000 patients with undiagnosed developmental disorders from 2011 to 2015. Their penultimate findings, published earlier this year in Nature, have implicated 14 new candidate genes for DDs and provided some fascinating insights into their genetic underpinnings. Yet I’m also discussing this publication because it illustrates some important considerations for sequencing studies of genetic disorders.

The DDD study performed exome sequencing on 4,293 families, predominantly as classic trios of an affected individual and his or her parents. Most affecteds (81%) had been screened for large duplications and deletions, which is typically done by array and captures events measuring hundreds of thousands or millions of base pairs. The authors also collected the family history and detailed clinical information (e.g. growth measurements, developmental milestones, and HPO terms).

De Novo Mutation Calling

The authors placed a primary emphasis on de novo mutations (DNMs) present in the affected child but not his or her parents. Well-established de novo mutation rates in humans predict that newborns harbor, on average, 0.5 DNMs in coding sequences. The authors used DeNovoGear and identified ~1.95 coding DNMs per proband. They acknowledge that this is too high, but argue that they were aiming for sensitivity and can address the lower specificity by independent validation.

Their filtering strategy for candidate DNMs appears technically sound. Using a set of 1,133 trios for which all candidate DNMs underwent validation (3730 sequencing), the authors chose a posterior probability threshold leaning towards sensitivity (see right). They also applied some common sense filters:

• Requiring that GATK call the variant in the child but neither parent
• Removing DNMs with strand bias (Fisher p<10^-3) in supporting reads
• Removing DNMs at multiallelic sites or with 1+ supporting reads in both parents

If we assume a conservative validation rate of 80%, that would yield about 1.5 DNMs per proband (a 3-fold enrichment over expectation). Yet this may be the true underlying biology: by enrolling DD patients with predominantly healthy parents, they’ve almost certainly enriched their cohort for individuals harboring coding DNMs.

DD Gene Discovery

Of the 4,293 individuals in this study, 23% had “probably pathogenic” missense or truncating DNMs within a clinically curated set of dominant DD genes. To systematically search for frequently mutated genes in DDs, the authors combined their 8,361 candidate DNMs with 4,224 DNMs (3,287 individuals) from 13 published sequencing studies. Some 93 genes achieved genome-wide significance, of which 80 had previous evidence of association with developmental disorders.

The authors performed some interesting analyses that I won’t cover in-depth here, such as:

• Creating composite face images from clinical photos of patients harboring mutations in the same gene to highlight common facial dysmorphology. It’s essentially the converse of the Face2Gene project. Honestly, the composites all look the same to my untrained eye; I find the example photos in Smith’s to be more useful.
• Integrating phenotypic data into disease gene association. Combining genetic evidence with HPO term similarity increase the significance for some genes, but decreased significance for a larger number of genes that are associated with DD and broader phenotypes.

About 20% of individuals in the cohort had clinical evidence of seizures and/or epilepsy, so the authors compared the analysis of this subgroup compared to the entire cohort for seizure-associated genes. Fifteen such genes achieved significance in both the whole cohort and the clinical seizure subset, but another 9 genes were significant only in the cohort analysis.

Furthermore, of the 285 individuals with pathogenic DNMs in seizure-associated genes, only about half (56%) had clinical evidence of seizures. The authors conclude that a large sample size easily outweighs phenotypic similarity for achieving statistical power to discover new disease genes.

Exome or Genome Sequencing?

The authors felt that the large number of significant genes provided an opportunity to compare exome and genome sequencing, with respect to their power to identify genes associated at genome-wide significance. They laid out the following assumptions:

• Genome sequencing costs $1,000 per sample.
• Exome sequencing costs 30-40% of that cost per sample
• Genome sequencing uncovers 5% more coding variants
• Project budgets is $1 million, $2 million, or $3 million

Under these assumptions, the authors found that “exome sequencing detects more than twice as many genome-wide significant genes.” Well, sure: if you can sequence 3 times as many samples with 95% sensitivity, you obviously have far more statistical power. Sample numbers always trump slight increases in variant detection sensitivity.

But let’s not forget that this is a gene-centric analysis focused on coding DNMs, which are accessible to both exome and WGS strategies. It ignores the classes of pathogenic variants that are refractory to detection by exome sequencing, such as noncoding and structural variants. These are harder to find and to distinguish from benign variation than coding variants. But if we don’t look for them, we’ll never fully understand the genetic architecture of inherited diseases.

Previous Testing and the Falling Hit Rate

The authors also noted that, although their power calculations estimated they’d identify about half of all haploinsufficient DD-associate genes at genome-wide significance. Empirically, they identified slightly less than half (47%). They hypothesized that a priori genetic testing may have depleted the cohort of easy-to-diagnose disorders. Indeed, an assessment of the DNMs in the most clinically recognizable DD genes showed a three-fold lower enrichment for truncating mutations than other DD genes. When these disorders were removed, they identified 55% of all haploinsufficient genes. Right on the money.

This example illustrates an idea about the feedback loop between clinical diagnostic sequencing and research discovery sequencing. In the first several years after exome sequencing was applied to rare disease diagnosis in appreciable numbers, researchers gradually came to recognize that the industry-wide “hit rate” — the proportion of sequenced cases achieving a molecular diagnosis — is typically 25-35% across centers and studies. Clinical and diagnostic laboratories have often adopted this as an informal benchmark for the success of their pipelines.

As this study illustrates, the proportion of cases that achieve a molecular diagnosis depends considerably on the extent of genetic (and biochemical) testing already performed. For completely naive (untested) individuals, we expect that hit rate to increase over time as new assays, methods, and knowledge improves our ability to find the underlying cause of disease.

Yet the improvement and widespread adoption of genetic testing means that such naive samples are increasingly rare. Most patients with a rare disease, particularly a recognizable one, will undergo a barrage of gene or panel testing (and often a copy number array) as part of the diagnostic process. As we find and establish new disease genes, that barrage becomes ever more effective at providing diagnoses before exome or genome sequencing are necessary.

In summary, we should not be alarmed (or self-critical) if the “hit rate” of exome and genome sequencing studies decreases over time. That is the expected outcome. In settings like ours, where research aims to inform clinical care, it’s actually a sign that we’re moving in the right direction.

Among the many fields of research and medicine aided by next-generation sequencing, few have seen a greater impact than the study of rare inherited diseases. Hundreds (certainly) or thousands (probably) of new disease-gene relationships have been reported in the past several years, causing exponential growth in databases like HGMD and ClinVar. The growth is also evident from simply perusing the table of contents of peer-reviewed genetics journals, which typically feature at least a handful of new disease-gene reports in every issue.

In the early days of MassGenomics, I even tried to keep track of new gene discoveries enabled by exome sequencing, but that quickly became untenable. There’s no way to keep up. However, I think the occasional snapshot of recent discoveries can tell us a bit about how exome sequencing is being applied to rare disease research. I collected six such studies, three each from the journals of American and European human genetics (AJHG and EJHG).

I required that they be exome sequencing and convincingly implicate a single gene. From these, I summarized the phenotype, study design (i.e. who was sequenced), some notes on the analysis, the genetic finding(s), and the supporting evidence presented. Most of these were disorders with presumed recessive inheritance, but the first example is a fascinating exception.

Study 1: Gonadal Mosaicism in Intellectual Disability with Epilepsy

Ewans et al studied a family in which the parents (healthy) had six daughters, four of whom intellectual disability with epilepsy. That might have been a combination of recessive disease with very bad luck, or something else. The authors performed exome sequencing on two affected daughters and prioritized shared variants that were very rare (MAF<1% and never seen in dbSNP) and predicted to be damaging. Because the inheritance pattern was unclear, they considered all possible inheritance patterns.

The number of variants uncovered was not provided, but the authors They identified a heterozygous frameshift insertion in IQSEC2, an X-linked gene already associated with intellectual disability and seizures. All affected daughters had the insertion, but neither of the parents did. This appears to be a case of gonadal mosaicism, since the odds of a recurrent de novo mutation at the same place are vanishingly small. The parents declined further testing that might determine whose germline had the mutation (I don’t blame them).

Study 2: A Founder Mutation In Distal Myopathy And You Won’t Believe the Gene

Peric et al describe a study that was part of the MYO-SEQ project, in which exome sequencing was applied to 91 unrelated Serbian patients with unexplained limb-girdle muscle weakness and elevated serum CK levels. Of those, 19 had a predominantly distal phenotype, and 14 of them carried a specific nonsense variant believed to be a founder mutation. The gene is titin (TTN). Now you probably understand the click-baity section header.

At over 30,000 amino acids, titin is the largest known protein-coding gene in humans, and as a result, comes up in genetic analyses all the time. In cancer genomics studies, we often see it when searching for recurrently mutated genes. However, it’s considered an effect of the gene’s massive size, and usually written off. The gene is so large that trying to view it in the gnomAD database gave me a network error.

The 14 patients that carried the nonsense variant (p.Gln35879*) that, I kid you not, is in exon #362. This truncates the very end of the longest TTN transcript, so I’d be dubious of its relevance except for two facts:

1. TTN is already implicated in cardiomyoapthy, with dozens of pathogenic mutations reported in HGMD, and
2. All 14 Serbian patients were either homozygous for this variant, or compound-heterozygous for it and another truncating variant.

The variant appears to segregate on a ~1 megabase haplotype in individuals of Serbian ancestry, suggesting that it’s a founder mutation (i.e. can be traced back to a common ancestor).

Study 3: Siblings Discordant for Mitochondrial Myopathy 

I can’t even count the number of times I’ve been in a meeting where an under-powered genetic study is being discussed, and someone says, “Maybe we’ll get lucky.” However, I *can* count the number of times we actually did. Zero. But it does happen, as illustrated by Nazli et al. They sequenced an individual with a complex encephalomyopathy / suspected mitochondrial disease and her healthy sibling.

Prioritizing variants with MAF<3% that were homozygous in the proband but not the control yielded 13 candidate variants (I have no idea why this arbitrary MAF threshold was used, or why compound hets weren’t considered). Only one variant affected a gene, TMEM65, that localizes to mitochondria. The variant itself disrupted a splice site and was heterozygous in both parents. So far, so good. To their credit, the authors carried out a series of functional assays to demonstrate the gene’s key role in mitochondrial respiratory chain function.

Study 4: Another TMEM Gene in A Rare Syndrome

Ta-Shma et al have a study that takes advantage of clever ascertainment: two unrelated families in which patients suffered a remarkably similar severe phenotype (CNS, cardiac, renal, and digital anomalies) and were the result of a consanguineous union. In family 1, an Ashkenazi Jewish pedigree, the child had unfortunately passed away, so the authors sequenced the parents to search for shared rare heterozygous variants. In family 2, an Arabic pedigree, only the mother and affected child were available.

It turns out that the unrelated patients from both families had another thing in common: homozygous truncating mutations in TMEM260. An extensive (and impressive) set of experimental follow-up work –including CRISPR/Cas9 zebrafish experiments and immunocytochemical/biochemical studies — demonstrated that the disorder arose from nonsense-mediated decay of a specific isoform of the gene.

Study 5: Consanguineous Cohort Uncovers New Gene for ID/Microcephaly

Tawamie et al had access to an entire cohort of consanguineous families affected by presumably-autosomal-recessive intellectual disability (ID). A data sharing consortium allowed them to identify homozygous missense variants that segregate with ID in two unrelated families. One was a family of Kurdish descent in which only the proband was sequenced, and regions of homozygosity were used to prioritize candidate variants. The other was a family of southern Italian origin with two affected siblings. Both were sequenced, and shared the same homozygous variant.

Given the genetic evidence that TAF13 played a role in ID, the authors performed an interesting experiment to assess its role during development. They used siRNA to suppress the gene in human neuroblastoma cell lines, and then applied RNA sequencing. Over 1,200 genes were deregulated in the knock-down cell lines compared to controls, suggesting an important role for TAF13 in development.

Study 6: Variable Phenotypes from Similar Genetic Lesions

Mingchu Xu et al conducted a very nice study of 10 affected individuals from 7 families of various ancestries. All had a similar phenotype of retinal degeneration with some additional developmental anomalies. Although unrelated, all of the affected individuals had recessive-acting truncating mutations in CWC27, a component of the spliceosome.

Two allelic mouse models (created with CRISPR/Cas9) showed that biallelic loss caused lethality and/or a severe phenotype, whereas partial inactivation caused only a retinal degeneration phenotype. This mimics the range of clinical presentations remarkably well, and highlights the complexity of the spliceosome network in human disease.

Rare Disease Genetics in 2017:

Here’s a brief table summarizing the studies discussed above:

Phenotype	Study Design	Analysis	Gene	VarType	Inheritance	Supporting Evidence
Intellectual disability and epilepsy	Affected sib pair from a family with 4 probands	Shared variants, MAF<1%, present in dbSNP, impact	IQSEC2	Truncating	X-linked dominant	Segregation testing, which showed gonadal mosaicism
Adult onset distal myopathy	Unrelated patients of Serbian ancestry	Relevant gene list, MAF<1%, predicted impact	TTN	Truncating	Recessive	Haplotype analysis
Complex encephalomyopathy (mitochondrial)	Discordant siblings	Segregation, MAF<3%, Control Exomes	TMEM65	Splice Site	Recessive	Immunofluorescence/immunoblotting, siRNA knockdown, subcellular fractionation, MT function (enzymology, protein abundance, O2 consumption).
CNS, cardiac, renal, and digit abnormalities	2 consang. kindreds; Mom-Dad and Mom-Proband	Shared hets, q>30, depth>10x, MAF<1%, in-house controls / Homozygosity mapping, IGV, MAF<1%, Q>30	TMEM260	Truncating	Recessive	RT-PCR, CRISPR/Cas9 in vivo testing (zebrafish), Immunocytochemical and biochemical studies
Intellectual disability and microcephaly	Homozygosity mapping + proband; affected sib pair	Segregation, MAF<0.5%, recurrence	TAF13	Missense	Recessive	RNA sequencing after knockdown, cell proliferation/differentiation assays
Retinal degeneration with/without developmental anomalies	10 affecteds from 7 unrelated families of diverse ancestries	Segregation, MAF<0.5%, recurrence	CWC27	Truncating	Recessive	CRISPR/Cas9 disease modeling (mouse)

I’m struck by the diversity of study designs — from individuals to small families to cohorts of unrelated patients — that enabled these discoveries. I’m also impressed at the careful experimental work (especially in the last four studies) that was done to functionally validate the findings. Studies like these set a high bar for publication-worthy manuscripts, but also provide useful models for unraveling the genetic basis of severe inherited conditions.