Next Generation Sequencing as a diagnostic tool in the epilepsy clinic

Remember Guthrie cards and the heel stick for newborn screening? It will be a thing of the past in 10 years replaced by methods performed through Next Generation Sequencing (NGS). NHGRI and NICHD have already committed to a $25M program for Next Generation Sequencing in Newborn Screening and first reports appear describing the value of exome sequencing in solving undiagnosed cases. However, these reports all leave clinicians working in the epilepsy clinic scratching their heads – this all sounds very good, but what can you offer your patients already, not just in 2-3 years?

265 genes at once. A team led by the EuroEPINOMICS researchers Johannes Lemke and Saskia Biskup has now evaluated the feasibility of targeted Next Generation Sequencing of a panel of epilepsy genes and the results published in Epilepsia last week are quite impressive. With their panel of 265 genes, they identified mutations in 16/33 patients with unclassified, presumably genetic epilepsy. While the overall yield of this candidate panel is probably lower than the impressive 50% in their pioneer study, these results clearly show that the general workflow in the epilepsy clinic is ready to shift from candidate gene screening to Next Gen panel analysis.

New and old genes identified. The list of genes identified in their screening is a mixed bag of epilepsy genes, many of which were identified in syndromes with a high degree of clinical suspicion including mutations in SCN1A, SCN2A and KCNQ3. Interestingly, some unlikely candidates also popped up. One patient with a clinical picture of Dravet Syndrome (DS) had a mutation in TPP1, the gene causative for Neuronal Ceroid Lipofuscinosis Type 2. This unexpected finding highlights another important “side-effect” of NGS: we will probably discover many unusual phenotypes for known disorders.

You wouldn’t think so, but panels are sometimes more thorough. Lemke and coworkers identify mutations in SCN1A in three patients with DS. This alone would not be all that remarkable. However, these three patients were previously reported to be negative for SCN1A by Sanger sequencing. This phenomenon is not new. In addition to identifying GABRA1 in SCN1A-negative DS, Mefford and colleagues also identified a mutation in SCN1A by exome in a patient with DS that was missed by conventional sequencing. While it is difficult to compare exome and conventional sequencing, these two anectodes at least suggest that NGS is not fairing any worse than conventional methods.

Study by Lemke et al. demonstrating the usefulness of targeted NGS in patients with epilepsy. Unlike few other genetic technologies, targeted NGS is very likely to alter your work flow in clinic at short term.

Targeted sequencing vs. exome. In the upcoming 12-24 months, we expect an intense debate on whether targeted sequencing is actually necessary or whether you could directly apply diagnostic exome sequencing. Targeted technologies – for now – have the advantage of the higher coverage, i.e. the eventual quality and completeness of candidate gene sequences higher than in exome studies. However, the field is evolving and the next, better technology might already be around the corner.

The surprising truth of your motivation for epilepsy genetics

Why are we doing what we are doing? Academic research appears to be a rat race of high-strung egomaniacs fighting for grant money, impact factors and ultimately their scientific legacy. In this constant struggle you either make it or you don’t, you publish or perish, depending on how good you can elbow your way through. And finally, make no mistake, it’s all about the money. Unfortunately, many young researchers are given this dire, coldhearted perspective by senior scientists and supervisors. Your PhD either results in a Nature or Science paper or you’re gone. Family? Not my problem. Holidays? Why are you even asking. This blog post is about the hidden secrets of human motivation, trying to point out some basic fallacies in these arguments. My brief answer to this is: “This is so 1995…”

Party like it’s 1995. Just imagine we are back in 1995 and we were asked the following question. “I would like you to tell me our opinion about the possible success of two different online encyclopedias. Type A is financed by the world’s largest software company, which has dedicated a generous budget to this project that pays both a highly qualified staff of writers and an experienced management team. Type B is a voluntary encyclopedia with no budget, established through people dedicating their spare time. In 15 years from now, which online encyclopedia will still exist?” In 1995, there was probably not a single person who would have put his or her money on Encyclopedia B based on this description. However, Encyclopedia B has evolved into one of the world’s largest online knowledge repositories, while Encyclopedia A closed its doors for good in 2009.

Wikipedia vs. Encarta. If I tell you that Encyclopedia B is Wikipedia and Encyclopedia A is Microsoft’s Encarta, this story makes sense to you. Daniel Pink provides this example in his book “Drive”, which tries to explain the secrets of human motivation. In brief, in contrast to the prevalent belief that strong incentives such as money or titles are the main drivers of human motivation, this “carrot and stick” method only gets you so far and will produce people being productive for the reward, and not for the issue itself. Pink identifies three elements that are the main drivers of motivation, namely Autonomy, Mastery and Purpose. In brief, Wikipedia became what is today by enabling people to work autonomously, to engage their expertise and to feel a sense of purpose through a shared experience and feedback, something that millions of dollars by Microsoft could not buy.

Motivational theory. Pink’s arguments are nothing new. They are based on scientific investigations by Deci and Ryan in the 1970s, who conducted sophisticated psychological experiments to analyse human motivation and who identified these three elements as part of their self-determination theory of motivation.

Application to research. Uri Alon from the Weizman Institute has re-interpreted these results for the field of science in a freely available comment in Molecular Cell, identifying Competence, Autonomy and Social connectedness as the three elements that apply to science. Competence basically relates to working in an environment that is neither too boring nor too challenging. In research, we are mainly faced with leaving people with a task that is too challenging for their current knowledge level. For example, suggesting that a Young Researcher design a sophisticated genome-wide association study on pediatric pharmacology without any prior knowledge of biostatistics is too challenging, eventually decreasing motivation. Altering the project to a candidate gene screening will eventually  increase the researcher’s motivation, despite the possible lack of scientific ingenuity. However, in the end, the second option will be more productive for the team as the young investigator is capable of working at her or his level of competence. Autonomy refers to a related issue. You can only be motivated in science when you perceive a sense of independence and an intermediate level of structure. Not too structured and not too independent. The third strand of motivation in science is Social Connectedness. It’s the proverbial water cooler discussion, the environment that gives you a sense of belonging, the interesting paper that was pointed out by that guy next in the lab next door, the senior postdoc who has nothing to do with your project, but  who is happy to have a look at why your PCR isn’t working. Networks have been the main driver of scientific innovations over the past centuries, which is “Where good ideas come from”, as authors Steven Johnson puts it. Naturally, the science network arising from research consortia such as EPICURE or EuroEPINOMICS is much more than just a collection of scientists. These networks are organic entities and the ideal breeding ground for scientific innovation in the field.

The hidden secrets of motivation. How motivation works in science and how to choose projects that are ideally suited for you in epilepsy genetics.

When scientific projects are really well suited for you. Uri Alon goes on to re-interpret these three elements in the context of scientific projects, suggesting a so-called TOP model (Figure). Projects are particularly well suited for you if they manage to completely engage you, drawing on your talents, your passions and your goals. Epilepsy genetics of the future will be multifaceted with many different niches and subfields that might allow a broad range of scientists with different backgrounds and motivations to contribute. Touching upon diverse fields such as genetics, neuroscience, social sciences, public health, etc., researchers with “cross-over skill sets” will be crucial. The age of the lonely genius researcher hiding out in his secret lab to eventually emerge with a Nobel-prize winning flash of inspiration is over, if it has ever existed. The science of the future will be network science and “chance favors the connected mind”.

The river of frequent rare variants

The flow of exome sequencing papers amounts to a small river.  In the most recent  work of general interest researchers captained by Bamshad and Akey from the University of Washington sifted for rare mutations in 63.4 terabases of exome sequences.  Next to whole genome sequencing the current output will feel like a trickle but their census yields about 300 mutations per genome that current methods for function prediction consider important and a strong bias on ancestry. Nature News has a fairly dry summary of the results.

Everyone’s genomes are awash with rare variants. The fact alone won’t surprise no one in the field but hopefully drown the claims that these are problems with current sequencing technologies such as statistical artefacts.

This plethora of rare variants will make interpretation of results from exome sequencing studies challenging, but also indicates that large consortiums like EuroEPINOMICS are necessary to navigate this stream of rare variants.

The Marrakesh diaries – The paradox of autosomal recessive common epilepsies

The Djemaa al Fna (Arabic: ساحة جامع الفناء jâmiʻ al-fanâʼ) is a square and market place in Marrakesh’s medina quarter (old city).

When in epilepsy genetics, choose sides. When you (a young, motivated researcher) start working on epilepsy genetics, you have to make a basic decision. Option 1 is the decision to work on the genetics of rare epilepsies. Rare epilepsies, either monogenic families or epileptic encephalopathies are expected to have a strong genetic contribution. In the ideal situation, a gene can be identified if the family is sufficiently large or if you happen to pick the right phenotype. Option 2 involves genetic research in common epilepsies. In this case, you can claim to be working on the seizure disorders which represent more than 90% of patients. Identification of genetic risk factors in this group of patients undoubetedly has strong implications, be it the identification of risk factors for drug response, side effects or novel risk variants, which help you understand the underlying biology. However, few of these risk factors are known and large sample sizes are needed to create a sufficiently powered study. Interestingly, the distinction between these two options pretty much represent the differences between RES and CoGIE.

Why not Option 3? Whichever option you chose, it has apparent downsides. You either work on very rare diseases or exhaust yourselve with more genetic complexity than you can handle. There is, however, Option 3. And Option 3 is what led part of the EuroEPINOMICS blogging team (namely Ingo, Johanna and Sarah) to Marrakesh. Option 3 involves genetic studies in familial forms of common epilepsies, particularly autosomal recessive variants of common seizure disorders.

The paradox. For disorders such as Juvenile Myoclonic Epilepsy (JME), we expect a complex genetic architecture on the population level with a wide range of genetic risk factors, a high degree of locus heterogeneity and phenotypic pleiotropy of risk variants. However, families have been reported, in which such common forms of seizure disorders are inherited in an autosomal recessive manner. For Juvenile Myoclonic Epilepsy, for example, several families have been reported with recessive inheritance. Furthermore, Salzmann and colleagues recently identified a mutation in the gene coding for Carboxypeptidase A6  in a recessive family with Temporal Lobe Epilepsy and Febrile Seizures. These families provide the ideal opportunity to identify genetic risk factors with strong effect, which would help us understand basic mechanism. The downside, however, is that these families are extremely rare. This, again, is where Marrakesh comes into play.

Genetic research in populations with a high degree of consanguinity. In several populations around the world, consanguineous marriages are frequent. Marriage within the family sometimes represents an important social factor. This constellation is seen in many Arab populations. Autosomal recessive disorders are relatively frequent in these populations and represent an important source of morbidity. Interestingly, highly consanguineous population do not have a significant increase in the frequency of common disorders, suggesting that the effect of recessive mutations occurs in a black-and-white fashion, i.e. these mutations cause severe monogenic disorders when present in a recessive state, but do not contribute strongly when heterozygous. Highly consanguineous population, therefore, lend themselves for gene discovery in recessive disorders. The frequency of these disorders is higher and identification of the underlying gene is highly important to the families and population for screening.

What is recessive JME? You might argue that recessive Juvenile Myoclonic Epilepsy does not exist and probably represents a form of Progressive Myoclonus Epilepsy (PME) which either present in a mild form or is not properly diagnosed. While it is difficult to argue against the presence of an atypical phenotype in the absence of genetic data, a long non-progressive course of a mild myoclonic epilepsy without ataxia or additional features is difficult to reconsile with PME. And PME genes are notoriously absent in cohorts of patients with JME. These disorders are simply distinct. Therefore, recessive JME holds the great promise to identify novel recessive genes for IGE/GGE that will undoubtely unravel novel mechanisms. From the top of my head, I would not know a single gene that, when absent in a homozygous state, might cause JME and the mechanisms to be identified hold great promise. The Morocco trip (supported by the International Bureau of the  BMBF) has allowed us to already include some families into the EuroEPINOMICS project with more families to come. 

Balanced translocations in neurodevelopmental disorders

Major genomic rearrangements without loss of genetic material — balanced translocations — are infrequently found in patients with neurodevelopmental disorders but also in unaffected individuals. Genome sequencing and break point analysis of 38 patients now show a fine-grained view of the implication of such chromosomal abnormalities in neurodevelopmental disorders.

Most balanced translocations in patients with autism or ID are non-recurrent and leave the geneticists who want to understand cause and effect scratching their heads as the overall genomic content appears normal as all genes are present at first glance.

Study design of the recent study by Talkowski et al. on break point sequencing in balanced translocations in neurodevelopmental disorders

Continue reading

The 30% boundary of exome sequencing in seizure disorders

Let face it: Current exome sequencing technologies will illuminate only a small fraction of the genetic load in seizure disorders. This post might not  motivate you starting another large-scale gene hunt using exome sequencing. Also, it won’t cheer you up if you have promised your funding agency that new techniques will discover “a large fraction of the genes implicated in seizure disorders” or “explain a large proportions of the missing heritability” – phrases frequently used in modern grant proposals?

Continue reading

Mutations in the same residue and phenotypic consequences in SCN1A

The link between a mutation and the corresponding phenotype in genetic epilepsies is sometimes not trivial. Mutations in the same gene can lead to different phenotypes (phenotypic heterogeneity) and different genes can lead to the same phenotype (genetic heterogeneity). These issues appear to be particularly prominent in some forms of seizure disorders. One of the many active research fields in genetics is studying whether environmental factors or other mutations lead to the development of a given syndrome.

Generalized epilepsy with febrile seizures plus (GEFS+) and severe myoclonic epilepsy of infancy (SMEI) can be due to the mutations in the same residue of the alpha-subunit of a voltage gated sodium channel encoded by SCN1A. Japanese researches now report in Epilepsia that the interaction with the beta-subunit of the channel rescues the GEFS+ associated mutant A1685V but not A1685D considered to be responsible for SMEI.

It’s a small step up on our understanding of such mutants. Computational analysis suggests that both variants have strong effects. E.g. Polyphen-2 predicts both to be probably damaging. It will still require further research on interactions to assess the differences. Part of this research is carried out in EuroEPINOMICS projects.

Probing autism for hidden autosomal recessive mutations using exome sequencing

Study design applied by Chahrour et al., PLOS Genetics 2012 to identify autosomal recessive genes in non-syndromic autism.

Autosomal recessive neurological disorders are usually distinct and severe diseases that result from the combination of two recessive alleles transmitted by parents. Autosomal recessive disorders are rare, but collectively account for a significant fraction of the genetic morbidity.  With respect to neurodevelopmental disorders including epilepsy, neurometabolic disorders and storage disorders frequently result in complex phenotypes that also comprise intellectual disability, behavioural issues and seizures. Particularly in populations with a high degree of consanguinity such as certain Arab populations, recessive disorders represent a major challenge.

Is autism a recessive disorder? Some recessive disorders might present with atypical phenotypes and are “hypomorphic“. Given that recessive disorders may appear sporadic, i.e. only a single child is affected, it is virtually impossible to distinguish the inheritance pattern in a single individual, particularly in small families. Accordingly, the question frequently arises, if and to what extent neurodevelopmental disorders may either be atypical presentations of known recessive disorders or may be due to novel, as yet unknown recessive mutations. Continue reading