DNA Today: A Genetics Podcast

6 of 190 episodes indexed
Back to Search - All Episodes

#146 Jo Bhakdi on Whole Exome Sequencing

by Kira Dineen
April 16th 2021

DNA Day and Match Day (for genetic counseling grad programs) are only one week away! To celebrate we have two exciting announcements. 


D. N. A Day and matchday for genetic counseling. Grad programs are only one week away. So to celebrate, we have two exciting announcements are first is that we're going to be live on instagram with Dina D. N. A. On saturday april 24th at 12 PM pacific three p.m. Eastern. We're going to be celebrating both genetic holidays. So come with your genetics questions, including genetic counseling, grad school questions, we'd love to answer them. So head over to our instagram for more information. We're at DNA radio on there and maybe there's going to be a giveaway. I don't know, you're going to have to go to our instagram to find out announcement number two. We have teamed up with Jackson Laboratories and million women mentors Connecticut to create a resource guide to help students and teachers learn about the importance of genetics in our everyday lives. The guide includes genetic lessons with worksheets to help engage students in genetics. It's going to be available this coming monday april 19th at DNA podcast dot com. Also available elsewhere. But that's where I can point you were gonna be celebrating D. N. A day, the whole week leading up to the actual day in april 25th.

You can join us on social media using hashtag C. T. D. N. A day that stands for Connecticut since we're Connecticut based along with Jackson laboratories and million women mentors Connecticut. So looking forward to seeing you all there and I can't mention D. N. A day without mentioning the episode that we did specifically about it. So it's actually 100 episodes go. So it's episode 46 of the podcast that celebrates DNA day including the history of the day activities you can do and careers in genetics. So be sure to check that one out. We find ourselves surrounded by such complexity. Hello you're listening or watching DNA today. We are a genetics podcast and radio show. I'm your host here Dean. I'm also a certified genetic counselor practicing in the prenatal space on the show. We explore genetics impact on our health through conversations with leaders in genetics. My guest today is Zhou Bhakti, the founder and ceo of Quan jean. And in this episode we are going to be exploring whole genome or rather whole X home sequencing.

We might touch base on whole XOM and genome sequencing together. So thank you so much joe for coming on the call here and letting us talk about this specific type of genetic testing. Yeah, you're excited to be here. I think it's a tremendously important topic and of course wanting paves the way in some respects in Germline and also somatic testing but I think today is more germline and yeah I'm very excited to dive in. So when it comes to whole exon sequencing for people that maybe haven't heard of this before. If there are students listening what is whole exon sequencing. So genetic testing started you know very in a very limited way focusing on specific variance of maybe specific genes because there was simply no other option to do things. Because sequencing was very expensive and very complicated to handle. So as you as your listeners probably know, we have 3.3 billion nucleotides on the human genome. Uh these roughly 1% a coding would be called coding uh jeans or quoting um nuclear types specific and they are structured and about.

It's actually an interesting topic how many genes we actually have but we have something around 20-21,000 genes that are being used. Their people stay up to 9000 additional silent genes that are not being used. So that's questionable if they're excellent or not. Um and then you have a vast 99% of the genome is very likely non coding or at least not being used. Not being read. We're not being translated into proteins to be most accurate here. And so when we talk about, you know, we mean the entire thing 23 chromosomes, 2.3 billion nuclear types or a. T. C. Gs. Uh and when we talk about Excel and we talk about this 1% of the 3.3 billion roughly 30 million Nuclear tides across roughly 20,000 genes Um that our body actually uses to produce, you know, these 20,000 different proteins. And so basically the more cost effective way if you would have told everyone 10 years ago that was cost effective. That have been probably $10 million nowadays watching is the first company to bring this price below $1,000 including clinical interpretation.

Uh we just dramatic, it's a dramatic step change. Um and that is very different because if you do a Bracha gene panel for example, we're talking about two genes to BRCA genes, then we do our gene panel. We talked about 20,000. Yes. So it's a very different scope. And Another thing decides the sequencing cost. You need massive amounts of Ai and cloud computing to do the clinical analysis of 20,000 genes. If you want to do it cost effectively not have like a curation scientist worked for two years on examples. Right? So when we're looking at just the exam, as you were saying, we're only looking at the genes that are active and we're only looking at the actual genes were not looking at the entire library of our genetic information. So when it comes to actually doing this in the laboratory, what technology is used for this whole exon sequencing? I mean it's quite an undertaking, as you were saying, there's so much data that comes out of this. But what is the technology that's used to actually get that data? Yeah, there are multiple vendors who specialize in whole exon sequencing kids.

So we run on Illumina sequencing machines and oversee 6000 mostly. So the latest iteration of you know, massive parallel sequencing um machines and the actual chemistry kids that are being used. You know, we have tired journey of illumination. You have a bunch of vendors who allow you to take saliva samples extracted in a uh shear the D. N. A. To the right size about peace. And then there is a specific technology that allows you to amplify all the material across the exam and then she pins it on. And so it's interesting with that is you know, as you said, you're like amplifying the D. N. A. That it's not just read once but it's read many many times. So you can see, okay, are we getting that same letter every time at that same location and then compiling all that information, I remember the first time I learned that I was like oh they don't just read it once. They're reading it many times. I mean how many times is it red and compiled in that way? So that's very interesting that deeper you dive into the more interesting it yet.

So what you what everyone has to recognize when it comes to sequencing is this is a game of chemistry and statistics. So you cannot just take a piece of DNA and sequence it right? No one can do that. So what you have to do is you have to massively amplify that piece. Uh Well in that case a lot of pieces uh across the entire, you know and you have to amplify them. So they exist many times hundreds of thousands of times each piece. And then you basically take a pipette and you put it on the flow cell, on the actual sale that reach it out. And that means you you are entering a statistical game realm right? So you don't know, okay, you don't know how how often each fragment of each piece of the D. N. A. H. Uh uh sequence of the DNA is actually exists. You can just do statistics on it. But if you have a trillion pieces and you take the ipad and the trillion pieces uh they contain you know maybe uh in average 1000 times each piece you want to look at. But an average means it's a bell curve. So on the tail end Randomly some of these pieces only exists 100 times and others exist 10,000 times and you have no control over it.

You put it on the floor cell and the flow said kind of randomly picks out statistically how much space you give it right for that specific patients. So there's something called sequencing depth and sequencing depth tells you how often in average you look at each location. And so now it's a little complicated. So the bottom line is in clinical great sequencing, you want to normally you go for 100 to 120 times 100 220 x. It means an average you read 100 220 times each location. But of course why do you do it? You don't need 100 times when it's 100 times you need, you know 10 times To be super safe. You need 20 times. Why do you need more than one? Because if you only do one read and the reader is wrong, which happens in sequencing, it's a laser signal based. Uh then you might read A. T. But it wasn't A. But if you if you look 10 times and you get T T T A D A T. T right at the same location, they say, well it's a team, you know, 70% of that as a T.

So let's just assume it's a T. Because it needs to be one or the other, What needs to be at least 50% 1 of the other depends. And that's how you do it and then you basically reverse engineer and say ok, I need to make sure that all the stuff I'm investigating Is at least read 10 times. How often an average do I have to look at? And then I was like, well under 220 times and then you get to a bell curve there, you know the ones to cover the least. I still cover 10 times. And of course it's all statistics, you always will miss some that's very important to understand like in many cases that always cases there Some locations will always be less than 10 times just statistic. And then you have a quality issue and you actually have to mark that. You have to have very advanced software that tests before you give it into a report to a counselor, you have to understand the exact depth that was actually found. So for example if you found a pathogenic variant but the bam file or whatever file you get out of the sequencing machine after processing it tells you well we only find three copies.

Technically you have to reject that. You have to say well that's just not enough that we can't be confident with that and especially if you were saying like it's a pathogenic variant where we say this could actually be harmful towards someone's health and then you say well we really want to make sure we get that right because that could have you know, massive impact on that person. Yes, exactly. So you have to that is why you need very advanced, you know computing capabilities and cloud capabilities to do so because you can imagine he was 30 million locations and have these little math issues and you find something that's pathogenic. You need to be able to go back to that individual variant and say how many copy instantly actually read and the sequencing machine and that answer needs to be correct and that informs that you can use it or if you have to really understand. And so when it comes to actually ordering whole exon sequencing, I mean when is it ordered? When in a situation with a patient is a healthcare provider saying, okay, this is appropriate. We're going to order whole axum sequencing. Well that is exactly one of the crucial questions that we really find what is normal.

It was not not in our opinion, it should always be ordered. And so here we are very controversial, right? A lot of people say no, don't do it, don't do the whole xo for random people proactively only do it in rare disease cases where we have a genotype that no one understands and we can't find the solution but we are kind of have a certain confidence. It is you know, it is um it is genetic and then people don't know what to do and then they order collects them. We want to change the paradigm so it becomes the default because the only reason not to do it for everyone was actually financial reasons that it was too expensive. What we believe is in the interest of medicine and medical advancement and in the interest of patients, there is no reason not to do it for everyone all the time. As long as you have informed consent and the patient understands the implication of genetic testing and to collect them. So what we do at quantum serenity is actually changing that paradigm and try to get more people into a whole exon sequencing because there's a tremendous amount of data and information that might be valuable for rotation.

And it's a paradigm just away from reactive medicine towards preventive and proactive medicine. So in reactive medicine you say why would I test that? There's no genotype, there's no issue. So don't test it in proactive and preventive medicine at all. That's the whole point. We don't want to wait until there's a major problem. We want to see everything before we know what's going on. So we know what's going on on the genetic level. So that's why we believe that there should be a shift to hold exon sequencing and that's why we made it available um at a reasonable price and in a preventive proactive setting. And so thinking about this is a preventive setting with with whole exon sequencing. This is similar to newborn screening where you know, I can only speak to the U. S. But all babies unless their parents opt out are screened at birth for conditions that are very serious. Most of them have treatment that can be life changing. So early diagnosis is very key. Do you see whole exon sequencing replacing what we're currently doing with newborn screening or you were talking about consent to?

We can't get a baby's consent for this. So I don't know do you see that being replaced in terms of newborn screening? Or am I just too many decades in the future for that? Well there are two topics here that are very interesting number one um miners right, how do you deal with the ethics question of generating deep genetic data for minors before they can before the age of consent. Because you're dealing with a human being and you're dealing with deep insights very, very deep insights. And is it ethically, you know, defensible to make that decision for another human being without their consent? That is something that is something we have to discuss. It's a complicated question. I don't have a view on that. We listen to many people. Um it's a difficult question and you know, because once you do this for for a baby it's it's there. The baby can't say later. Okay, let's I mean they could say to need it but can they really need it in the health care system. I'm not sure. Um so that's a different question. Um the other question is, you know, do you want to have comprehensive inside and who makes this decision If you want it?

If you decide medically it's medically actually appropriate. I would recommend it for everyone. But I wouldn't recommend it for people who can't make decisions and I don't want to force them to do it. So I would recommend for everyone. But it needs to be the decision of that specification. And the next question is okay who pays for it and who makes these decisions outside yourself? Let's say you say, okay I want it. Who is the next decision maker here? That is also why we believe it. Quantum only launched this thing this year. We need to start with a sense player market. We need to say, okay, that's probably a very aggressive on price. But we are very cost conscious and try to make it very affordable. So everyone can pay out of pocket for it. For a simple reason. If you go to insurance or Medicare, same thing, you're suddenly dealing with a decision making apparatus that is completely disconnected. And I mean that completely disconnected from your patient interest, your basically outsourcing the decision either to the government or to privately for profit companies.

I don't know which one is worse. By the way, it's equally bad. You're basically giving it either to bureaucrats or bureaucrats who work for profit. And so bureaucrats who work for politicians versus bureaucrats who work for profit. So who knows? And they make the decision right here. Uh you know what? No, we're not going to pay for you. And that is not rooted in a deep evaluation if that's good for you or not, it's a deep evaluation that's good for them or not. And thats why a safety system is much more powerful also to advance medicine because you say, well, as long as we can find enough customers and make our case with them, they simply pay, it's between us and you can quantum deliver the quality and provide protection for you. And can you pick wanting the price that we are asking that hopefully is reasonable and then we are done. There's no fair party and they're like oh we should also ask the U. S. Government what they think or like some random insurance companies and so Because if you wait for these guys to make up their minds you basically have to force them on an economic argument level to do it and that might take 1020 years of clinical trials and evidence and all kinds of stuff that they want.

That is not the evidence the patient needs or the doctrines because its economic evidence and that's our stands on it. So to your question really depends who you're dealing with and what the business model is and hopefully it's aligned to the patient in order to some other entity and then it's a question that each individual can decide, you know, do we make a compelling case? Right. This is important for you. If we are not compelling, don't do it. If it's compelling you can just do it. And I would hope eventually that you know with all genetic testing, eventually that insurance companies do catch up and say, okay, this testing actually is financially worth it. Unfortunately as you said, that's kind of how the system works but financially worth it because preventing disease and treating it before um is most of the time less expensive um than treating it afterwards. Um So it's kind of you know, I would hope that, you know, even if we're starting out with this model where people are paying out of pocket for testing that, you know, hopefully insurance companies would start coming around to that um when it comes to uh yes, but it's such an important point because it's our core philosophy here.

I think absolutely they should pay for it and we should make them pay for it. But the way to make them pay for it is not to bake them, but to create a product in the market that creates enough political pressure to force them to do it. Because that's the only way to really get it done, I think. And when it comes to the importance of doing whole exon sequencing, I mean, what information our patients getting out of this that can change their health. What are a couple examples of conditions or results where patients can make health choices that are different to help their health overall? Oh, that's a very far reaching questions we could list all day. Right? Yeah. I mean, there are like dozens and dozens and dozens of dangerous medical conditions or even hundreds if you include rare diseases uh, that can be found and they can be found to be uh, you know, dominant or present, like in the genotype that maybe it's not expressed yet. So, meaning, you know, there are diseases that if you know about it, you have certain metabolic shortcomings, um, you know the way you met metabolize iron for example, that only manifest As as clinical symptoms when you're 30 or 40.

Uh but you can actually take action once you know what it is uh far earlier. And then you never get the symptoms because you know, your body cannot metabolize iron to the extent necessary. And there are many there like hundreds of these examples. Then there's, you know, pharmacogenomics that you have drug interactions, right? You know, your genes are incompatible with certain drugs. So prescription drugs um, you have higher or lower metabolism for these drugs, which means the dosage you get is the wrong dosage to follow FDA requirements or recommendations For approval for the drug. You have the wrong dosage because you are half or 200% metabolism of the drug. So you get an overdose or under dose, which can lead to extremely bad things if this is about cancer for example. Um so you need to know that too. Um, of course you have carrier screening aspects, right? If you want to have kids and you have yourself tested similar things as discussed before. The insurance normally forces one partner to get the test and not both.

And then sees that there's a problem. And if there's a problem in the second partner get tested and only for specific conditions, it's much more powerful to say skip it. Test both whole exam and you have everything and that can be tremendously life changing because you can find things that you will definitely not find in conventional carrier screening and that can be reproductive choices. So that's the interesting thing about genetics, if you ask. Okay, what does it do? You can go on and on and on and on and they didn't for And also the reports were generating. There's probably generate regenerate reports around the medically actionable targets. But it's clear what to do. But they are also thousands, tens of thousands of additional targeted as unknown. Or like I'm not medically actionable but they're informed nutrition. So there's it's basically an entire treasure through of the deepest possible insights about yourself that you get from whole exon sequencing and that treasure trove was not dead with with a single report or a single session.

It's a lifelong thing. So you're generating this thing where you have all your coding genome, your exam sequence and available. And then it's an ongoing service to actually connect the latest science. Is it forever? Like it's an ongoing thing that never ends. Uh And I think for most people it's hard to even wrap their heads around genetics. And when you hear the first time. But I think once you get it is I think in No 10 years we have a large share we will have a large share of the population. We're not even consider not having the whole genome probably tested at this point and not use it for one of report or something. But as an ongoing super data warehouse where you constantly have an ai watching over your genome and connecting it to global science and every time someone at random you know research proven Southern Sweden or something find something out. You you have it the next minute on your phone and say like there is a variant in your genome where a new study has found that you might have problems metabolizing X.

Y. Z. And that implies stop eating these things for example. Yeah I think that's such a great explanation of just how many Specific genetic tests that doing a whole genome or whole axum sequencing could replace that. As you were talking about like carrier screening that something as a prenatal genetic counselor. I'm offering to patients all day every day and you know they may have done carrier screening earlier but it was only for 22 conditions and I'm offering a panel of hundreds of conditions and all of this. It's like well we could replace all that and you could just have it this whole exon sequencing and then you're pretty much done for genetics. I mean looking at if someone has their whole xOM sequenced and has that data would that need to be redone later in life. I mean obviously someone's genetics isn't changing but the technology does get better. I mean we were talking we started our conversation talking about like the depth so how how accurate are those letters that we think that are there in terms of like what comprises of the genes. I mean do we think that's going to really improve if someone would need to repeat this testing or do you think?

No once you've had it done that's good for your whole life. I see it more like this. So our technology the technology that are now going standard in the market since like the last two years. They are very very good. So to your question I think of course no one knows the future. Maybe some find something out that we don't even know. Maybe there's a new aspect to the genome. Maybe there's like uh mp genetics or something that is also important in the hereditary sense. And we are also electric biopsy companies. So then that's the deep end when you go into actually cancer mutation detection in D. N. A. In the blood and things like that. But that's a whole that's uh a rabbit hole by then. Comes to Germany in my opinion. You know it is likely that we have a very very good solid foundation and if you do it now you're done with the eggs. Oh and I think what comes next is the genome. So you in my opinion if you do a serenity whole exon sequencing test you will be fined for 34 years. And what we will do then is repeat the test for whole gino to just catch up with the remaining 99% because we know there is stuff going on it's much more opaque, it's much more difficult.

And right now everything clinic, nearly everything clinically actionable is in the X. O. But there's for sure some information in the genome beyond the exam. That has some meaning in the entrance right in the non coding regions like who the hell knows that might be important. And but it's like the 80 20 rule only. That's probably the 95 5 rule. The exam gives you 95. So 1% of the genome gives you 95% of all information is necessary but the remaining 5% is spread out over the last 99% and then we want that information at some point. So do your question. I think short answers I would predict you are very well suited with Wrexham but in four or five years we would repeat it with whole genome for everyone. Just kept up with the rest but it doesn't replace. I don't think there will be any new insights on your eggs. Oh it's just adding more stuff. Yeah. It's interesting to see that once you know someone has all that data as you said they could you know go to their psychologist and they're looking to get on a certain medication and then they're looking at that to see okay what medications are going to work best with your genetics.

What dosage. You know you're then having a baby later in your life and you're looking at that for your carrier screening results, you start having symptoms for disorder and you're looking at that and saying, oh, we can actually see that you have a change in this gene that's giving us this diagnosis. So really just like following through someone's life of just all the information that it could provide. Um, you know, in terms because it's such an important point. What Quantum is doing with the gene cloud where we put your X on into our cloud of course, super privacy protected. Not even insurance or other physicians get to that, but you have the key. So you have that in the cloud from there on you have this giant treasure trove of data. The number of reports you can generate from that is basically limitless. You're not throwing limit this report that you have a specific initial report, but here's what it does down the road for you in your life. Any time you need any kind of clinical grade panels for any gene, You can pull it from that cloud within 48 hours. And that is as you know, as a genetic counselor, that's a huge revolution because if you have a cancer patient, pancreatic cancer and they want Bracha for example, investigators that has no impact on pancreatic too.

Um that might take six weeks or four weeks to order because you need to collect saliva or blood, you need to extract it. You need to run it in the lab. You need to sign it off. All that is already done before you even start because now it's in the cloud. So all you need is pull the the director of the clinical report from the lab director and so that is a big game changer and so basically have them, the strength to talk can constantly dig through it for any reason that comes up. Yeah, it's amazing to have just as a resource as you said, I mean when I'm doing genetic testing for patients, most of the waiting for results is getting their sample and then having to ship it off. So usually it's going somewhere in the US and usually it's not right. You know, I'm new york based, usually it's not right in new york. It's setting out to Utah or California usually and then you're waiting on the lab to process all that and but it only takes a couple days to actually look at that genetic information and then have it analyzed. So most of the waiting is like in the shipping and the processing and everything. So when it does come to whole exon sequencing, if someone is, you know, doing either saliva or spit, how long does it take for results to come back?

What is the turnaround time? Right now? The turnaround time for the first time, like when you actually do the entire thing is roughly 5-6 weeks including report and counseling session. Um I think we will bring this down in this year to probably four weeks or three weeks or something. But realistically right now it's like more 60s. That's way less than I thought you were going to say. I feel it's been a while since I was in pediatrics but I feel like it was much longer um when we were ordering that for patients that had like an undiagnosed condition that I don't want to misspeak. I feel like it was a couple of months or something for turnaround time. Used to be still for many. I mean what quantity did we have? A six year Andy tracker again now. Right. So we built very, very massive cloud and ai systems behind it. So this is not like some random person sitting there and which it is the most laps. There literally there is a person with google going through your variants and there's no fun. So we have a very advanced cloud portal, the irish pre structures everything connects dozens and dozens of databases to it.

You know, cross checks all the variants and basic pre digest the entire report for the actual creation scientists. And then the curation scientist goes through all critical variants and does the manual check to be hmD compliant? Uh double check quality and then we have some algorithms in place to avoid any kind of false positives and problems but there's a huge, huge amount of work that the computer or the cloud does for the eye Because you have to I mean these are non trivial little problems that you have, you get a giant file out that kind of millions of variants uh at least hundreds of thousands. And you technically have to revisit like every single variant it goes through, you know, five Areas of dimensions that all count up to like up to 28 different dimensions you have to take for each single variant you know, population frequency, protein impact and change The publications and reference materials. You have to read through for every single variant if you might be brothers with 100,000, you're not getting anywhere.

So what the iron clark can do is highly automate this process and then as tremendous efficiencies but also tremendous quality improvements because what you don't want us to do this manually because there's too many human errors that can go in. Yeah, certainly. And that's just a lot of work because you think of all the data that's coming out of this. Um and then if you're going on the whole genome side, I mean even more data going through that. So certainly a lot to sort through. But I just want to thank you so much for coming on the show and just sharing all your insight with whole exon sequencing. I think it's really cool to see. I mean just where we are today. Just blows my mind. But then thinking about where we could be in the near future. So thank you so much for sharing all that today. And where can people find out more about yourself in quarantine? Yeah, of course. On quantum dot com. That's always good. You can go to my personal twitter. I'm trying to tweet a little bit more uh not succeeding yet, but uh dot com. So let's go back to uh, and then we have, you know, our linkedin page, if you're interested in jobs, uh general news, uh, you know, when linked in if you look for quarantine um, yeah, it's, that's the main source is fantastic.

And we're going to include all of that in the show notes for today's episode which is available if you're listening to this as a podcast. Um if you just swipe right or upper depending on the different podcast app you're using. All the information is also available at DNA podcast dot com. You can connect with us on twitter, instagram, youtube facebook by searching DNA today and any questions for myself or joe feel free to send into info at DNA podcast dot com. We are happy to answer your questions and one favor before we let you go. Um please rate review on Apple. We'd really appreciate that. That's how more people can find out about the show coming up next on DNA today, A brand new series. May a cystic fibrosis month. So we are honoring the disorder in a three part series in these interviews were diving into the history of the disorder, current upcoming medications in clinical trials, patient advocacy and so much more look out for the first installment, which drops on May 7th. And we're also gonna have a giveaway go with that episode so you don't want to miss it. Also want to remind you to join us live on instagram saturday april 24th at 12 PM pacific three p.m. Eastern.

We're gonna be celebrating both D. N. A. Day and matchday for genetic counseling program. So come with your questions are really excited to engage with you. Also be sure to check out our D. N. A Day resource kit. We've teamed up with Jackson laboratories and million women mentors Connecticut so that we can help students and teachers learn about the importance of genetics and our everyday lives with lesson plans. So be sure to check that out. It's going to be available again. That's D. N. A podcast dot com. Thanks for listening or watching. Please join us next time to learn, discover new advances in the world of genetics and DNA.

#146 Jo Bhakdi on Whole Exome Sequencing
#146 Jo Bhakdi on Whole Exome Sequencing
replay_10 forward_10