Genome Biology: Exome Sequencing


Pod Academy are pleased to present the first Podcast from Genome Biology


Genome Biology is an online journal publishing cutting edge research in genomics. This is the study of the genome, which contains entire genetic material within an organism.

This Podcast coincides with a special issue of Genome Biology which includes articles focusing on exome capture sequencing, a technique that has recently expanded in popularity. The Podcast includes interviews with Jim Lupski, Jay Shendure, Elaine Mardis and Joris Veltman about the challenges of identifying disease genes using exome sequencing and whole genome sequencing.

Introduced and Produced for Pod Academy by Jo Barratt, with Genome Biology’s Hannah Stower and Ciaran O’Neill.


Jo Barratt: Genome Biology is an online academic journal publishing cutting edge research in genomics; the study of the entire genetic material within an organism, the genome. In the journal’s first podcast,Genome Biology focuses on the challenge of sequencing the genome in order to discover the genetic basis of disease.

Using genetics to discover the basis of disease is by no means a new concept, but advances in the techniques used to sequence genomes have started to allow researches to look for specific genetic variations or mutations causing disease or playing a role in disease development.

Sequencing genetic material with a view to making predictions about health and disease has obvious wider implications for the future of healthcare, and this podcast looks at several different approaches to finding genetic mutations playing a role in disease.

The first researcher we spoke to, Dr Jim Lupski, talks about his research into Charcot Marie Tooth Syndrome, which has affected him and members of his family, and which he has been investigating for over 20 years. Because a number of different genetic mutations can lead to this disease, their identification has proved particularly difficult. Whole-genome sequencing of Jim’s own genome in 2010 eventually allowed the identification of the mutation causing his own disease. This was an end to a personal quest, but also proof that sequencing the genome of an individual can return meaningful information about a disease; a finding that has significant implications for the future of personalised medicine.

The whole genome approach has been made possible following the generation of the reference human genome, which gives a standard reference for comparing the genome sequences of individuals. Variation in the sequence might give clues as to the genetic basis of disease. Although falling in cost, sequencing a whole genome is nevertheless a costly business, and it was this hurdle that led Dr Jay Shendure and colleagues to try to find a cheaper approach to finding meaningful results.

The approach of Dr Shendure and colleagues was to focus purely on one part of the genome thought to be particularly important to health and disease: the exome. The exome consists of the exons, or regions of DNA that code for the proteins, which are the building blocks of cells and therefore whole organisms. If the DNA in the exome is mutated, abnormal proteins could be produced, and the disruptions this causes could lead to disease.

The exome comprises a comparatively small part of the whole genome, and focussing only on these protein-coding regions reduces the cost of disease gene identification. Dr Shendure, The Guest Editor of September’s special issue on exome sequencing, talks us through the development of exome sequencing as a viable approach to identifying disease mutations. The technique has proven very successful thus far in identifying causes of Mendelian diseases, and Dr Joris Veltman spoke with us about his work on mutations in mental retardation.

As we hear in the podcast, there are caveats to this narrower approach – the exome doesn’t necessarily tell the whole story. Elaine Mardis, a specialist in cancer genomics, talks about the suitability of exome sequencing in the cancer field.

As a young technology in a rapidly changing field, there are many questions currently being asked about how genome sequencing will have a practical application to healthcare. The topic stimulates interesting debate among specialists about what will be needed to bring these sequencing techniques to the clinic, and what ethical issues we need to consider. What, for example, does one do about the discovery of ‘unexpected findings’ when sequencing genomes?

Hannah Stower: Hello and welcome to Genome Biology’s inaugural podcast. I’m Hannah Stower, the Special Issues Editor and I’ll be accompanying you through today’s podcast focusing on the identification of disease genes through whole genome and exome sequencing.

So, why have we chosen to make a podcast now? In this month’s edition of Genome Biology, we have a bumper special issue focusing on exome sequencing. This is a recently developed technique that allows the sequencing of only certain targeted parts of the genome – normally all of the exons and hence the term exome sequencing. In the past year or so, this technique has allowed many researchers to identify mutations responsible for a wide range of diseases and so in this podcast we’re going to discuss this technique, and its big brother, whole genome sequencing – a technique that continues to fall in cost as well as being more accessible for researchers. We are going to hear from 4 researchers involved in whole genome or exome sequencing.

Of course, the search for disease gene mutations did not start with whole genome and exome sequencing,researchers have been searching for disease genes for years, and the first person we interviewed, Jim Lupski from the Baylor College of Medicine had been searching for the genetic cause of Charcot Marie Tooth disease, a disease that has affected his family for a long time. This disease is one of the most common inherited nerve disorders, but it is currently incurable. It is genetically heterogeneous, causing Jim a long and protracted journey. Finally in an attempt to identify his causative mutations, he had his genome sequenced. As you’ll hear, Jim’s quest to find the cause of his disease somewhat dictated his academic career and we join him as he remembers his graduate days.

Jim Lupski:
During the time that I was in graduate school I happened to pick my issue of Nature and read Jim Gusella report of linking huntington’s disease with a specific region in the human genome. It seemed pretty obvious to me that this could turn out to be a very powerful way to get at the genetic basis of a number of different genetic diseases. I was fortunate enough to go, in 1986, to the Cold Spring Harbor symposiym that was on the molecular biology of homo sapiens back then. It was there that Kary Mullis got up and started talking about PCR and it became obvious to me that we were going to be able to a lot of the things we doing with bacteria using human DNA. This was also the very symposium where some of the first discussions of the human genome project took place.

I spent 3 years doing paediatrics, and then started my own laboratory. I was collecting families during some of my residence and also collected my own family (back in 1986) and got Electrophisyiological studies done on the three generations that were alive at that time including three of my four grandparents who were still alive in their late 80s. In my family it was recessive so it seemed to be a much more difficult problem. We tried to do a tractable problem— to study the dominant diseases that had not been done for Huntington’s so we collected large dominant families with Charcot Marie Tooth disease. As we honed in on the locus for Charcot Marie Tooth disease, or what we thought was the locus at that point, all we knew, or thought we knew, was that it was seemingly one linked locus on chromosome one and that was done by protein polymorphisms. When most of us checked our families, we found that they were not linked to that locus so it didn’t seem that there was a major locus hit. We really didn’t know if it was going to be one genome or one prominent gene. There was already literature suggesting heterogeneity for this disease because there were exiling forms, dominant forms and reccessive forms, so we knew that there at least used to be one genome x and one gene on the autosomes.

We honed in on this locus and tried to go for the gene. In the process of doing this we kept getting weird findings in our analysis that didn’t quite make sense. We would see an artefact, what we thought was an artefact, of several probes that seemed to be very close. We thought we were going in the right direction but then, when you did linkage analysis, it actually suggested we might be moving away because you saw what was an apparent recombinant. As we honed in further we developed dinucleotide repeat polymorphisms used for linkage — this was about 1990. What we saw in the patients is three alleles rather than the usual two— you would inherit one form each parent. It was a little bit tricky back then because there were a lot of stutterbands you would see with dinucleotide repeat. You really had to carefully look at what you were seeing when we associated that back to the southern blogging data whenever we saw 3 alleles, and yet you were heterozygous, in a patient with a disease heterozygous for RFLP (restriction fragment length polymorphism) alleles suggesting hetrogeneity for this disorder becasuse you could always see dosage was different in the RFLP. It could be either one or the other but you would see two to one dosage or one to two dosage in the RSLT and then it started to click that there was some kind of duplication going on at this locus. We carefully studied all this and then looked at several families and we saw that the majority of patients always seemed to have this duplication. We were working very hard but we still had not found the cause of my disease. At that point we were up to over 30 genes in which point mutations, simple nucleotide variations could cause Charcot Marie Tooth disease. With that in mind and as part of the continuing quest to find out the cause of my own disease we undertook sequencing of my genome as well as trying to perform extensive copy number analysis because it really would have been embarrassing if we had done the whole genome sequencing and we missed copy number. The long and the short of it was that we found one mutation at 6x and by 30x and it was very clear that in this SH3CT2 I had compound heterozygous mutation at that specific locus. furthermore my three affected siblings had the identical pair of mutations and each parent was a heterozygous carrier as would be anticipated from a Mendelian segregation so then we spent a lot of time trying to understand what it meant for the rest of the family.

Hannah Stower: So, we’ve heard from Jim, and his own journey to find the genetic cause for Charcot Marie tooth syndrome. However, sequencing techniques are not always developed with the identification of disease genes in mind. Exome sequencing, for example was developed initially just to reduce the cost of sequencing genomes. In the process of developing this technique, Jay Shendure and his colleagues from the University of Washington also found that by targeting the exomes, the protein coding regions only, that they had reduced the search space in which to look for these disease mutations and could do so with great success. I caught up with him to talk about how he developed exome sequencing and then started using it to ‘solve’ the genetic cause of some diseases.

Jay Shendure: So I guess the origins of my involvement trace back to when I was just wrappin gup my graduate work when we developed one of several mutations of next generation sequencing and the technologies were still at a point where where genome sequencing was cross-prohibitive for most or nearly all groups. As a consequence of that we and others began thinking pretty deeply about methods of targeted capture that were compatible with tehscale and undelying technologies involved in next generation sequencing platforms. I think it became quite clear that the timing was right for a good match between where the sequencing was at and the notion of applying capture technology specifically to comprehensively target the approximately 1% of the human genome that is protein coding which has come to be termed the exome.

In my own lab we began working on array based hybridization capture methods that would target the human exomes to relatively high coverage and quality. The culmination of that line of work was a publication in 2009 in which we performed exome capture and sequencing of 12 humans. Somewhere along the way in the process of developing the technologies and starting to apply it we were thinking about where we could actually demonstrate that it was useful. The simplest place to start is with the simplest diseases which happened to be Mendelian or single gene diseases. There are obviously thousands of Mendelian disorders that have been solved over the past several decades using a combination of linkage mapping and positional clonine or other approaches but there remain thousands of Mendelian diseases for which the causal gene has not been identified. We came up with what we felt was a compelling strategy for applying exome sequencing over a relatively small number of inividuals to try and pinpoint the locus for unsolved Mendelian disorders.

Hannah Stower: Was it obvious that this technology was highly applicable to these diseases from the outset?

Jay Shendure: Not at all. It was neither obvious that the technology would be sensitive enough to pick up causal variants and it was also not entirely obvious what one would need to do to narrow to a single gene. There are some advantages to focussing on the exome in that looking at the history of Mendelian disorders one could have a fairly high belief that causal variant could have an effect and that was the basis for focussing on coding sequencing and adjacent splice sites which is almost always where we find most single gene disorders. What was unclear was that the extent of the background and what it would take to get down to one gene and that was really the exploratory nature of the study. For that purpose we started with Freeman-Sheldon syndrome which was a disorder for which we already knew the right answer and sequenced four individuals with Freeman-Sheldon syndrome used and something which we call discrete filtering which is an intuitive strategy that assuming that variants that are common in the human population are not causal from Mendelian disorders and using that as a filter and in addition to that assuming that all individuals with the same disorder are going to share the same variants in the same gene. In the case of Freeman-Sheldon syndrome this turned out to be the case that narrow it to a single gene dispite only having four effective individuals. There is a few points here that are compelling. One, that the necessity of controls. So had we not included in our study 8 cat map controls and performed exome sequencing on them exactly equivalent to how we performed the exome sequencing on the diseased pro-bands I don’t think there is any way we would have been able to narrow down to a single gene. We also needed extensive catalogues of Variants in the human populato in that are available though DB Snip and the like. At the time this seemed like a lot its only that in that last 2 year and we have thousands of controlled exomes. It retrospect we are very glad we did this so that set of controls. A second important point is that the real breakthrough in terms of applying exome sequencing or genome sequencing to Mendelian disorders is that this is the first time we are able to solve these disorders in a way that is entirely independent of linkage based analysis. for 100 years linkage has been a central aspect of what we do in genetics whereas narrowing to a single gene in the context of Freeman-Sheldon syndrome was something that we did with no reference to linkage. One of the reasons this was an interesting test case was that this is a disiese that could not have been solved with linkage analysis. Freeman-Sheldon syndrome is primarily cause by genovo mutations The individuals who have the syndrome by and large do not reproduce. The only reason we knew the right answer is that it had been identified by a good guess and candidate gene analysis. That aspect, the idea that one can identify a candidate genes by a strategy that does not at all rely on linkage represents a significant shift for the field and one that opens up the possibility of solving other mendelian disorders.

Hannah Stower: So, exome sequencing has provided new promise for solving the genetic basis of many Mandelian diseases. Simply by reducing the search space in which to look for mutations, researchers are now able to prioritise the variants that their sequencing experiments pick up more easily. Of note, Jay discusses the first disease that he focused on with this approach: Freeman-Sheldon syndrome a rare form of multiple congenital contracture syndrome, which is often caused by de novo mutations. These are mutations that arise in the patient and were therefore present in the parental germ cells meaning that the parents are asymptomatic but are able to have children with the disease. Such mutations are difficult to trace by family trees. With exome sequencing, these de novo mutations are identifiable by sequencing what is known as patient family trios, in which the patient is sequenced along with their two parents. The de novo mutations are present in the patient, but not in the parents. With the advent of these exome sequencing technologies, researchers are beginning to identify more and more of these de novo mutations. For example, cases of Autism, Schizophrenia and mental retardation have all recently been attributed to de novo mutations. I spoke with Joris Veltman from the Radbound University Nijmegen about his work on mental retardation and how he was able to identify de novo mutations using exome sequencing.

Joris Veltman: We have been studying intellectual disability for a long time. We see that for a large majority of patients copy number variations (CNVs) do not explain the cause of the disease so it is logical then to look at point mutations or indels. Exome sequencing is allowing us for the first time to look at the entire coding part of the genome. Also because we knew that the de novo CNVs were frequent we also started thinking aboutde novo point mutations for point mutations but the decided to go for exome sequencing in patient parent trios. The reason being also that if you do exome sequencing in a single patient you will find many variants and of course the prioritisation of the variant is the major issue. How can you find a variant that is causing the disease, you can look for rare variants, but then you still have a few hundred per exome. You can look for variants that influence the protein parts, or amino acids or nonsense mutations, so you have many variants

but then when you look for de novo mutations by actually filtering out all of the variants that are inherited from infected parents then suddenly you just end up with a few variants and that is what we hypothesised though literature so we started doing that last year and ten patients with unexplained sporadic forms of mental retardation and that is how we started with exome sequencing.

Hannah Stower: Could you clarify exactly what de novo mutations are?

Joris Veltman: What we did is we sequenced DNA from blood isolated from patents as well as from parents which you can see then the mutations that are presnt in a hetrozygous state in the patients DNA in blood and you do not see that in the DNA from parents. That means that it is very likley the mutation is present germline in the patient but it is not present germline in the parent. You think that probably something went wrong during the copying of the DNA in either the spermatozoids or the oocytes.

Hannah Stower: Why do you think the importance of this has gone undetected for so long?

Joris Veltman: Of course people have identified de novo mutations before in rare diseases but that was for specific genes where they identified these . For example we identified de novo mutations in the genes CHD7 as a cause of CHARGE syndrome after we identified a region by looking at a deletion so after finding that gene before.So the problem of course is in disorders like intellectual disability in that you expect that genes all over the genome can be mutated you don’t know where to look for. so sequencing is not a tool you can use. Exome sequencing is perfect because in one experiment you can look at all of the variants that are present in the coding part of the gemone of a patient and then by doing a trio approach where you also approach the DNA from the parents you can immediately identify the inherited variants and then filer out those and end up with the few de novo mutations.

Hannah Stower: Of course, exome sequencing may not be applicable to all diseases. So far we have heard how exome sequencing is particularly applicable to identifying disease genes involved in Mendelian disease. However, what about other more complex disease such as cancer? I caught up with the cancer sequencing expert Elaine Mardis who researches at the Genome Centre, Washington University. I asked here about the application of sequencing technologies and we began by discussing the application of whole genome sequencing vs exome sequencing to cancer.

Elaine Mardis: Exome sequencing certainly has some value. Our laboratory has long been committed to cancer genomes just because there is so much structural variation in the cancer genomes to not use the exome by rather using whole genome based sequencing methods. We have begin to document the number of tumrs continues to get higher and higher in terms of specific tumor types. its quite clear that genes can be impacted by a multitude of different types of variation in the somatic genome. So yes it is true that point mutations are maybe there, and that could be picked up by exome sequencing, but of course we also see lots of evidence of portions of entire genes being deleted, being altered by chromosomal translocation and variation within that subset of structural variation is quite wide even if you look at some sort of sub-class of deseise that’s characterised by translocation. That translocation looks different in every patient we examine, so the notion of putting together a tiling part or something to capture that in a directed fashion is essentially pretty fraught with peril. l I think in general. It’s likely that it wouldn’t work if you were going in after one specific thing with a targeted capture method. This is the essence of the choice: whole genome vs exome. I think that the papers that have used exome do show a reasonable approach— you do find mutation for sure. Its just that if you get a negative result or if you only capture one copy, theres not an easy way to ascertain that using the targeted capture methods. Its much more straightforward with whole genome.

Hannah Stower: So there we heard how for cancer, exome sequencing may not be the most appropriate technique because of the certain types of genetic variation that occurs in cancer. That is, copy number variations of genes are frequent in cancer. These cause there to be multiple copies of genes and so it is difficult to discern exactly which one is captured by exome sequencing, and also difficult to ascertain the level of copy number variation. I also discussed general limitations of exome capture with Elaine.

Elaine Mardis: the other thing that we are really convinced of is that as more and more projects begin to tease out the regulatory elements of the genone that are often not in genes — they can be in regions in and around genes— this is a agin an areas where an exome capture does not necessarily go after that particular region of the genome.

Hannah Stower: To sum up, Elaine is saying is that by choosing to sequence the coding regions of the genome only, you may miss a lot of the additional information in the genome that codes regulatory sequences for genes. That is, there are sequences that may be important in switching genes on and of and when they are mutated they may cause disease. So, exome sequencing is missing this information. In light of this, I asked Jay Shendure if he thought that there was an expiry date to exome sequencing.

Jay Shendure: I would argue that we’ve dichotomised this idea of exome vs genome and I question if this is appropriate in the sense that the way I tend to look at it is more of a continuum. So On one extreme you have have exome as a comprehensive approach for doing protein coding sequencing only, then you have genome. But this notion that whole genome sequencing as whole genome sequencing is wrong. We have not even finished the first human genome so the idea that next generation sequencing leads to a truly comprehensive picture of all genetic variation that is present in an individual is not true. Even whole genomes fall short of comprehensiveness. I think it is really a question. Of trade offs and your aprory belief in what the right answer is, and relative costs. I think we are approaching a point where the difference in costs between exomes and genomes is diminishing and as that gap drops it will make more sense to shift towards genome sequencing but I think that will not change much in terms of how the initial analysis is done. Much as the first whole genome papers focussed almost exclusively on the exome in terms of their analysis of potential causing variants. I think this will continue to be the case until we approve our ability to understand the functional consequences of non-coding variation. The problem is our inability to interpret hole genomes.

Hannah Stower:It therefore seems that we have a long way to go before whole genome sequencing will be the norm. Obviously, if we are able to use exome sequencing to pinpoint genetic variation that causes disease, there is potential to take this to the clinic for diagnosis. Often it is difficult to diagnose people that present with rare genetic disorders and we were interested in whether researchers that use exome sequencing now believe that this technique will be easily applicable in the clinic.

Jay Shendure: There have been two parallel stories over the last 2 years in this field. One has been the application of exome sequencing to solve new mendelian diseases. The second thing that has been occurring has been a large number of instances in which people are attempting to solve new mendelian disorders by identifying new diseased genes simply end up diagnosing a new mendelian disorder in a patient where a diseased gene is already known it was simply that that particular gene was not necessarily suspected for one reason or the other. This essentially points to the fact that we are not very good at phenotyping in the clinic. We are good in certain contexts. It is possible that exome sequencing may represent a much more cost effective way of immediately going to a diagnosis. Even for mendelian disorders where the degree of suspicion is high in a particular gene or set of genes. I think for a lot of patients this may avoid diagnostic journeys that individuals go though from one clinician to another trying to get the correct diagnosis. Second represents a unbiased way of doing a comprehensive survey of all genes and all mendelian disorders that is picked up in one shot that is becoming increasingly cost effective to do. Its becoming cost effectve so quickly that I think that in the long run it might end up saving the health care system dollars to do this earlier in the diagnostic process.

Hannah Stower: In summary, Jay Shendure thinks that in principle, taking exome sequencing to the clinic is very possible and could reduce what he termed the ‘diagnostic journey’ in which patients go from clinic to clinic to find the cause of their disease. However, what needs to happen in practical terms to bring exome sequencing to the clinic? Elaine Mardis

Elaine Mardis:I think a lot of things have to occur to make that a reality. Much of this is ongoing at present in terms of work that multiple groups are involved in. In the US we have been having lots of conversations with regulatory groups form the government. Trying to understand what the essence of the challenges are to reproduce ability for the methodology including the bioinformatic analysis and its healthy to have these kinds of dialogue because it gets all of the approaches out in the open and we understand where the errors come from and how to make more systematic in terms of the approach and the results that come from this. So I think that is definitely a portion of what needs to take place. We also need to do the kind of ground work to show that next generation methods are comparable or better than the current state of the art clinical laboratory setting which is typically using standard PCR and capillary sequencing approaches to get an answer. These cross comparisons have to be done to the point where those clinical labs and the regulatory agencies feel that the same answer or a better set of answers are obtained using the next generation methods. The other component that is going to become really critical is the transition to education. Not only about which tests a physician should order when. Once we have the physician education up to the point where they know to order the test then the y have to be able to interpret the results of the test in terms of what that means for the therapeutic options available. There is a lot still to work out but I think we have to get started on it and then all of these necessities will fall into place as we go along rather than being frightened by all the things that need to be done and not doing anything. We just need to get these things underway to the point that we feel comfortable introducing it into the clinical space.

Hannah Stower: We do therefore have quite a way to go before we can use these sequencing techniques as diagnostic tests. Clearly, reproducibility is a key problem when it comes to anything with a medical application. However, there are other ethical issues when it comes to sequencing personal genomes or, indeed exomes. When sequencing a whole genome or exome, it is likely that other disease causing mutations will also be found in that individual as Elaine Mardis found in one of the two papers that we discussed. Unfortunately, we only have time for this one story;

Elaine Mardis: The second paper is an interesting one in that one of the things that people often say about whole genome sequencing as a negative, is that what is typically referred to as uninternded consequences or results— where you are sequencing a persons genome for one purpose and in the process you identify something else. Something you were not looking for but information that needs to be returned to that patient or their family for their own healthcare purposes. In this case the patient was actually deceased. She was being studied as part of a large project sequencing the genomes of therapy related AML or PAML. These are patients that are typically treated for a solid tumour or two (this patient has breast then ovarian cancer). Because of the aggressiveness of the chemotherapy they later go on to develop acute myeloid leukaemia – often this is fatal. This was the case with this patient. What we found in sequencing her genome is that she had a de novo mutation in a gene called TP53 that would not have been picked up by conventional methods of PCR sequencing. The underlying germline mutation in TP53 is essentially indicative of a syndrome called Li-Fraumeni syndrome. Patients with this have mutations that they develop in TP53 which gives them a lifelong tendency to develop cancer. Certainly her clinical history was indicative of this. Again this patient was deceased but had three living children, so this information needed to be transitioned back to the family members. These children needed to understand that they should be tested to see if they carried the mothers mutation in their genomes and if they were positive for this test then they need to begin a lifelong vigilance about the development of cancer from a very early age. Early 20s is when you start giving patients full body CT scans to pick this up. It is important for people to know that you will find things that you often don’t anticipate but this can and should be dealt with, and should be. So not not doing whole genome sequencing because of wht you might find.

Hannah Stower: There are a lot of important ethical factors to consider when sequencing genomes of individuals – not least that they might not understand the implications of the identification of additional disease mutations. In the example just discussed, Elaine used the resources at her medical school, the cancer genetic councillor. So, when these techniques do eventually go into clinical use, it will be important to make sure that these resources are available. In addition, informed consent procedures before sequencing will have to be altered to be appropriate. I spoke about this with Joris Veltman too.

Joris Veltman: Its geared at both when you start doing genome sequencing as well as exome sequencing. It becomes and issue. How do you handle the data, how do you reprt back to the patients. How do you handle clinically relevant findings which are not related to the dignostic requests. Clearly when we started thinking about diagnostic applications we of course had to completely set up and revise our informed consent procedures. This is a major issue. We are moving towards diagnostics, and a lot of people now see that there are major challenges in terms of how to organise and inform patients. What to feedback to them. Its also something where we have to learn a lot and it s very important to do this in a good way. We of course want the patients as well as the patient organisations along with us. We need to be very clear on what we do.

Hannah Stower: Someone that would understand the implications of identifying additional mutations is Jim Lupski, who, we discussed has previously had his genome sequenced. I caught up with him to discuss any unexpected findings when he had his genome sequenced.

Jim Lupski: There is always a bit of trepidation when you are dealing with the unknown and trying to move forward. Certainly we have all heard the naysayers who said how we shouldn’t do this and why and how we might find something that could really upset people. This is something that we have to deal with in clinical medicine all of the time but what is most important most of the time is to have the correct information. When we did do the sequence one of the variants we initially found was remarkably a hit that said that that vary variant was found in 12 year old boy in a persistent vegetative state. This was adrenoleukodystrophy locus. Then you ask yourself is Jim Lupski in a persistent vegetative state, or isn’t he? Well I didn’t think I was and it was a little scary, but we went to the original paper that reported that and we saw that they actually found more than one variant in that patient and this is the one that they picked that they thought was the most likely to be pathogenic. Maybe that patient needed all three variants for the pathology or maybe one of the other ones was the important one but obviously I didn’t have that neurological disease. The other funny thing was that we did find variants that could turn out to be very useful for me. We found variants in different pharmacogenetic traits. Although it didn’t help me at that very minute it turned out that three years before we did my genome sequence, here I was worrying about my Charcot Marie Tooth disease in some of my future health issues regarding that and I woke up one morning shaving and there was a lump in mu neck and and I ended up getting months of radiation. Had I needed chemotherapy. We saw from my profile that there were certain drugs I would want to stay away from.

Hannah Stower: So, the unexpected findings in a genome are not always negative, but can provide some useful information too. making sure that people receive the appropriate advice on their results will be a very important consideration if these studies are taken to the clinic.

So, this is all we have time for today. We’ve discussed the applications and development of modern sequencing and capture sequencing technologies. If you are interested in finding a bit more about exome sequencing and its applications, please do check out the September issue of Genome Biology. In our special issue we have articles on applying exome capture in mouse and wheat, in addition to an article investigating the genetics of hereditary deafness. We’re very excited about it and we hope you are too.




Tags: , , , , , , , ,