Interrogation of Prospective and Archived Samples for Population-Scale Biomarker Discovery

本站添加: 2026-05-26

[00:00:11]welcome everyone to our webinar on bridging genomics and proteomics interrogation of prospective and archive samples for population scale biomarker Discovery I'm Lance Ramsey the director of marketing here at formulatrix I'm joined today by two speakers first we will hear from Dr David Corney a principal scientist within the process development team at aenta Life Sciences over the last eight years he has evaluated emerging Technologies for their Gene Wiz portfolio leading the development and commercialization of their proteomics long read sequencing and single cell sequencing services and then we will hear from Dr Usman ashra application scientist at formulatrix responsible for our automated liquid handling Solutions and their applications within the Sciences some brief housekeeping before we begin this webinar is being recorded and will be made available on our website after the sessions today there will be a brief Q&A session at the end of the presentation if you have any questions feel free to add those into the chat box at any time if there are any questions that were not answered during the session we will follow up after the webinar with that I'd like to to hand things over to our first Speaker Dr David Corney Dr Corney the floor is yours thanks Lance for the introduction and welcome everybody to our webinar today as Lance said we'll be talking today about bridging genomics and proteomics and how we can incorporate prospective and archived samples into population scaled biomarker Discovery we'll be talking about the importance of biomarkers and the role all of biomarkers in population skill studies and then talk about a few examples where such Studies have been enriched by incorporating oink promic data I first want to give a quick background to those of you who may have not heard of asenta before maybe you've heard of some of the different companies that together make up asenta for example genis our mission is to enable breakthroughs faster and do so by leveraging the wide range of capabilities within a zenter in fact we've been around for quite a while in fact we're celebrating the 25th year anniversary here at ginwood as I said our mission is to enable breakthroughs faster and we do this in multiple ways by supporting the journey from sample sourcing through storage analytics informatics through to the approval of a drug we will spend much of our time today talking about the house and the why way of deploying various genomic Technologies to obtain data but I think it is important that we start by remembering that samples and especially clinical and clinical trial samples are extremely valuable materials that we can use to answer our scientific questions now while the number and the capability of genomic tools has exploded in recent years we don't always have the perfect technology available to answer every question today in fact we might not even know what that question is today so I think it's critical that samples are maintained appropriately with robust chain of custody such that they can be later retrieved and analyzed thereby maximizing the value of that sample with the consumables instruments software and Sample storage that asenta provides this enables the maximization of many millions of samples maintained within our bio Banks we'll now be talking much about the analytics and the ability to measure samples that we here at a Zena provide we're first going to start talking about the importance of biomarkers and the role of population scale studies now first what is a biomarker for me a biomarker is a measurable indicator of a biological state or condition today we will discuss molecular biomarkers but any measurable feature could be a biomarker biomarkers have applications throughout the drug development pipeline from diagnosing diseases predicting and mon monitoring therapeutic responses to monitoring safety events importantly drug development is extremely expensive rising to over2 billion dollar over the last decade per drug now despite this cost only 14% of compounds used in clinical trials make it through the drug development process to Market and only 20% of patients respond well to the top 10 highest grossing prescription drugs biomarkers have been getting a lot of attention due to the promise they offer to reduce these costs and match patients to a drug that they will responds to and in fact clinical trials that incorporate biomarkers are more than twice is likely to succeed there are many different types of molecular biomarkers some biomarkers are better suited in some applications than others for example genetic markers such as germline variants are static which makes them useful for detecting mandalian disease and DNA is also very accessible and has become very fairly straightforward to analyze at scale by Next Generation sequencing and so it's for this reason that DNA and then later RNA has been a focus of biomicro Discovery and inclusion as readouts in most population scale biomar Discovery initiatives however proteins and especially circulating proteins present within many biological fluids have a unique advantage in that there is incredible complexity in the amount of proteins produced as well as dynamic in their abundance this offers the ability to associate protein abundance with disease progression and monitor the pharmacodynamic responses over time so to summarize then a major goal within drug development and patient management is to use biomarkers to drive success and Adoption of personalized medicines by giving the right patient the right drug and at the right time population scale studies are important tools which provide a strong Foundation to enable this process by analyzing large cohorts of samples rare events occur frequently and now combining this with clinical and omix data associations and Trends can be identified that allows for biomarker development to support the diagnosis prognosis and so on indeed the development of biomarkers is uh even making news in in the popular press as well as the literature we see several examples here of how molecular biomarkers can now enable more effective prognosis and diagnosis we're particularly interested in identifying such effective prognostic and diagnostic tests for oncology and neurodegenerative diseases and others for the reason that early detection is critical for the best Pat patient outcomes here I show an overview of the many population scale genomic projects that have been funded and launched in recent years a major enabler of the scale of these studies is next Generation sequencing which has really propelled the quick update of population scale genomics across the globe we've seen that such uh NGS has been instrumental in improving our understanding of human health and disease and allowing such uh advances at scale you'll also see there are Geographic disparities with few studies taking place in Asia Africa and South America but I think this is going to change as we further realize the importance of diversity in the genomic tools and databases that we frequently use and indeed projects such as the human Pang Genome Project which aims to increase the number of sequenced individuals from underrepresented populations has a great deal of promise and I think there's a lot more potential that we will be seeing as these population studies come to fruition however I think one thing to to mention here is a large proportion of such population scale Studies have focused on the genome and to an to an extent the transcriptome and I think adding proteomics has a great deal of potential to enrich these data sets now I say this for a couple of reasons I think we've obviously started out looking at genomes due to the accessibility not only of of obtaining a sample that allows us to look at the DNA but also just that the DNA itself is is a stable molecule there's a reason that we can look at the woly mammoth DNA right that is a stable molecule that lends itself to interrogation now at the same time the complexity of the DNA is fairly low this lens is self too looking for variants identified that that may uh predispose individuals to disease but is less informative to monitor diseases over time as we Branch out through epigenomics and transcriptomics the complexity of these molecules and the number of molecules present really explodes to the extent that we expect to see over 200,000 RNA transcripts with just distinct expression patterns produced just from these 20,000 genes and in parallel with this increase in complexity we see the accessibility of these samples decreasing this is really taken to the extremes when we look at the proteum where we expect over a million different proteins and after considering post transational modifications over 10 million different modified proteins now this a great deal of complexity has been a challenge for the field um the the ability to look at these different proteins at scale has been very difficult to do and at the same time obtaining samples that allow us to look at these uh different proteins has also been a deal of of challenge so it's for this reason that we've been really happy with the solutions provided by oink to enable the high thrit and the highx analysis of the protein there are a number of different solutions from oink whether you are interested in looking at thousands of proteins or tens of proteins there is a solution here typically we see the the incorporation of the Oink Explorer workflows for Discovery applications such as what we will talk about predominantly for the population scale studies where up to about 5,000 or so different protein biomarkers can be interrogated in just two microliters of sample but on the right hand side we also see the more targeted and focused assays which allow for the interrogation of a handful of proteins again using a small amount of uh input material but allows for um more more targeted interrogation all of those oink products r on the proximity extension assay and so we see a little schematic of that here for each protein Target Two antibodies must bind and these antibody pairs are conjugated with DNA tags that are complementary and a neol when in proximity subsequently the double stranded oligo is extended and then detected either by qpcr in the case of midlex assays such as Target 48 and Target 96 or Next Generation sequencing as in the case of the Explorer assays now importantly the requirement that the target protein is bound by two different antibodies allows for a sensitive assay and reduces the number of false positive hits so that you spend less time following up on dead ends now to enable the Oink workflow at scale is critical to deploy Automation and here at aenta we've worked close closely with oink and formulatrix to implement the fast liquid Handler high accuracy and precision is always important in any molecular biology workflow but that is especially the case here when when deploying the Oink workflow this is for several reasons such as working with very low petting volumes wanting to obtain a quantitive readout with a low coefficient of variation and peping different liquid classes with various viscosities which also presents a challenge the liquid volumes we handling require special attention in the case of the target workflow we are preting as little as one microl ler which could feasibly be done by hand but the Explorer workflow requires dispensing of as little as 0.3 microliters of sample which would be impossible to accomplish manually or using traditional liquid handlers and for that reason we are using the fast we Ed the instrument both for the low P petting steps as well as some of the sample pooling operations and PCR setup throughout the protocol even when larger volumes are used due to the accuracy associated with positive displacement by pading I also want to give an acknowledgement and uh you know very proud of our team here that's deploying this workflow of course we have the uh great technology from from oink and we have the by petting accuracy and precision from the fast instrument but I also want to recognize the attention to detail and the expertise from our team here that's really enabled us to deploy this workflow at scale for our customers in addition to the validation work performed by oink we also performed our own qualification studies in house to support the use of the assay within our lab we took a couple of different approaches here first we took a handful of samples and processed them multiple times both in the same and in different processing batches and then looked at the difference in the expression or the abundance of proteins across the different replicates and impressively saw around 12% coefficient of variation across those replicates we also took 48 sample that had been previously analyzed by Oink in their laboratory and reanalyzed them here at aanta and saw a 95% correlation in the abundance between these two measurements this is very impressive uh it talks not only to the assay itself but also the hands that we have uh running this assay especially considering the Benchmark expectation for the correlation was uh 0.9 R2 so I'm going to dive into a little bit more detail and and show a little um show show a couple of case studies on how promic has enriched population studies now the first one I want to talk about is the UK biobank which was really a leader in uh not only uh deploying genomics for population scale studies but incorporating proteomics as well well in the first phase of the study uh around 500,000 individuals had their genomes tested by whole genome sequencing as well as whole exom sequencing and combining these datas with health registry data and metabolomics in this phase one study uh or the phase one um uh segment of this project o link promic was incorporated as well by studying over 50,000 of those individuals and taken together enabled the generation of 162 million protein biomarker data points very importantly and very interestingly I think to show the potential of this approach was the fact that over 10,000 primary genetic associations were identified between the genome and the prodium of which 85% were previously unreported just showing the potential that this approach offers a study that we here at aenta have been very proud to be involved with is the finen study which is looking at a very large proportion of individuals in Finland by deploying genotyping as well as wgs and taking that data again with health registry data together now with oink prics this study aims to look at 10,000 individuals um by by interrogating the prodium with oink proteomics workflow and over the last just couple of months we've already gone through about a quarter or so of those samples generating 12 million protein biomarker data points impressively the data passing rate is extremely high in these individuals we've seen a 98.7% of samples passing the QC metrics and 99.8% of the data points successfully Meeting those QC uh criteria so we're really excited to see what uh comes next from these prod from these population scale studies and I think it kind of leads to the question what do you do next so you you go through your discovery workflows um using the Oink Explorer uh hlex workflows but then what do you do after you've gotten that data and I think naturally you would want to expand the number of samples that you would look at and really focus on perhaps a handful of different protein biomarkers and and look at them in more depth and so for the last couple of slides I'm going to focus on uh the ability to use the the more midlex and the lower Plex assays that the target 48 and and the flex assays to to do just this we were approached by a p biop pharmaceutical company who was entering phase one 2A trial for their therapy which was targeting a rare blood disorder needed for an analytical readout of inflammation in patients so after some discussion with the client we recommended a longitudal monitoring solution for quantifying protein measurements uh within the samples collected from those individuals and in particular I decided to use the Oink Target 48 cyto account panel to look at the abundance of cyto and chemokines Within These individuals the study was designed such that we started with a baseline collection followed by collection every two um two weeks or so as you can see in the schematic samples were also collected following any indication of a hypers sensitivity reaction all of these samples were processed analyzed and stored here at aena and then analyzed by the Oink Target 48 workflow and those samples that did have the indication of hypers sensitivity were analyzed and reported within less than one week to allow for Meaningful and appro management of those individuals so to conclude I hope that I've really shown and and Justified that proteins are important biomarkers that kind of enable um a transformation of the drug development process both supporting early efforts in research and and R&D all the way through development and into postmarket surveillance population scale genomic Studies have broadened our understanding of human health and then moved us closer to the personalized Health remembering that uh the traditional studies looking at genomes are a little bit more uh disconnected from the proteum we think more about the proteum being very close to the phenotype now now the proteome is now more accessible than ever before due in in large part to the different technologies that we've spoken about today and now has enabled the deployment of these Technologies for population scale studies to enrich existing data sets with prodeum data so I want to conclude um and really mention something that's kind of hopefully comes across well in in what I've spoken about today is that although we've spoken about yix in quite some detail we really don't want to look just at the prodium and really the value in in omix in general is deploying a wide range of different Technologies to look at different modalities whether that's the genome or and the epig genome or the transcriptome um microbiome and so forth and so I'll just leave with the the mention that we do offer all of these different Services um within our regulatory Labs as well as our research labs and we be happy to have a discussion with you to see how these different modalities can help uh your research and really maximize the value that we can get from the valuable samples that are in storage and and available to you so with that thank you for your time and I look forward to the questions in the question and answer session thanks for that Dr corny up next from formulatrix we have Dr Usman Ashraf I'll hand things over here Dr Ashraf the floor is yours thank you Lance hello everyone now we will have a brief overview of our fast instrument a Next Generation positive displacement liquid Handler which is specifically designed for Speed and accuracy at low volumes let me highlight some of the key features that set it apart from other liquid handlers fast uses disposable positive displacement tips that ensures accurate piping of liquids with different physical properties these features make the system liquid class agnostic fast is highly precise for pipeing low volumes ranging from 0.1 to 13 microl with a CV below 5% at 0.1 micr equipped with the 96 Channel head fast and quickly performs tasks such as Spate stamping between 96 and 384 well plates fast uses a built-in touch probe to automatically Define lever geometry and calibrate the deck that significantly reduce manual setup errors F features Advanced Computer VN to handle automatic tip management without the need for manual intervention the system has a removable deck that accommodates SPS footprint labware and holds up to two tip carries each containing 384 tips let's dive into the key component of the fast its unique disposable tips think of these tips as Tiny singes each with a disposable piston inside the plastic tip as you can see the piston is in direct contact with the liquid being transferred this eliminates the air gap and make the system agnostic to the variation in physical properties of liquid such as viscosity and surface tension the 96 channnel head offers exceptional flexibility in tip selection and pickup it can pick any number of tips up to 96 any configuration of rows and columns suitable for variety of pting tasks such as plate stamping broadcasting or even a single tip for Cherry pickup this versatility is powered by computer V technology here's how it works the system scans the tip CAD to identify both the number and position of tips before starting a workflow fast automatically arranges the tips according to the protocol eliminating the need for manual tip management additionally if the system detects insufficient tips for a protocol it alert the user to replace the caddy for a seamless workflow the same vision technology checks for RN tips underneath the dispens set preventing the system from potential damage and maximizing up time defining different play Types on conventional liquid handlers can be a complex process it often requires the assistance of an automation engineer however fast significantly simplifies this process you can use the plate definitions available from the labw list or create a new ones through the labw manager feature let's take the example of defining an IFC plate in the fast system simply select the appropriate IFC plate type from the drop- down menu the fast software is equipped with predefined layouts for various IFC plates facilitating an easier setup next place the IFC plate onto the system pray and position the onboard touch probe inside the well A1 this probe determines positional coordinate by touching the sides of the vals then it touches the plate surface to measure the overall plate height and finally it moves down to meor the well depth once the process is completed the IFC plate is is precisely defined and save to your Lael list making it easy to work with different plate types without the need for specialize expertise let's move on to setting up an experiment with fast it is simple and quick and does not require any programming expertise it operates similarly to a multi- channel pipit for simple transfer and you just need to input the number of tips and volumes for your transfer to set up an experiment simply drag and drop the source and Target blades from the lab Wireless to the desired locations for example in a 96 SP plate stamping we have selected 96 chips from the cad and configured the transfer volumes for both the source and Target plates after completing the transfer simply click the W spin icon to discard the used tips and that's it quick and easy as shown by this example of 96f plate stamping Beyond Simple transfers fast offers a range of predefined options for more complex protocols such as cedal dilution normalization plate broadcasting and pooling providing flexibility for various experimental setups now let's talk about our collaboration with oing to advance the automation of their workflows When selecting a liquid handling solution for o link essays several key factors are considered first one is the sample viscosity oink technology is compatible with various type of samples including serum plasma and cell culture media so whatever the sample viscosity is using fast there is no need to adjust liquid classes its positive displacement pipeting system makes it liquid class agnostic oink s is particularly High Plex SES demand rapid and high throughput liquid handling the 96 Channel head on Fast ensures quick sample transfer and is compatible with 384 well plates for higher throughput oink s typically require low sample volume often in the range of 1 to 2 microl fast can pipe it down to 0.1 microl with high Precision maintaining a CV below 5% at such low volumes with its disposable positive displacement tips fast eliminates the risk of cross contamination ensuring the Integrity of both samples and SS additionally reserves for all link Target and flex series are red on q00 system which uses specialized IFC plates these plates have an unconventional well layout incompatible with most automated liquid Handler limiting many researchers to manual pipeting however IFC plates definitions are preloaded into fast lab list and users can easily add their own IFC plate formats using the system's automatic play definition feature fast is also easy to use and requires no programming skills or specialized EXP properties to create protocols its Advanced features offer oneclick serial delution normalization and pooling features that are very helpful to streamline sample preparation and o link explore workflows let's move on to the o link workflows for which fast has been validated and the specific steps where it is used in hlex workflows such as explore 3072 384 and explore HD Protocols are executed in two distinct environments prep PCR and post PCR fast has been validated for all the transfer steps indicated in Black from sample dilution to index transfer and PCR product pooling the steps highlighted in red are not currently validated with fast but we are actively working towards consolidating all steps to be performed using fars to provide a fully automated solution for for o link platform for o link midlex workflows fast currently supports automated protocols for Target 48 and Target 96 these protocols were previously performed manually but thanks to fast compatibility with IFC plate formats this ensures the seamless transfer of samples and primers from standard 96 well plates to IFC plates significantly reducing the risk of primer and Sample mixing furthermore lowplex workflow such as Flex has also been validated and automated using fast given its validation any new products or update to existing workflows will be immediately compatible with the system ensuring there are no delays in the implementation since integrating fast with oink platform we have seen its widespread adoption across numerous Labs researchers have shared positive feedback highlighting how automating these workflows has streamed their processes and significantly enhanced their lab efficiency as we wrap up a bit about formul Trix founded in 2002 and headquartered in Dubai we have grown into a global organization our journey begin with protein crystallization and we have since expanded our Innovative Technologies in liquid handling bench stop filtration General lab Automation and cell culture areas we are a growing family of around 1,300 individuals our systems are trusted worldwide with over 5,000 products deployed across 2,000 customer sites globally so with that I want to thank you all for joining and attending our webinar thank you very much thanks Dr Ashraf now let's proceed to the Q&A portion of the webinar all right this first question is for Dr Corney after I've identified biomarkers using oink Explorer how do I move forward testing these biomarkers in a more cost-effective manner thanks L yeah that that's a great question that we we do get from our customers as well and and so what we have done in the past was really recommends to evaluate the content of some of those midlex panels available from oink such as Target 48 Target 96 which tends to be built around specific Pathways and around specific disease areas and look to see if the proteins that you're interested in are available in any of those off-the-shelf panels if that's not the case then we would recommend to consider building a custom panel using the Oink Flex workflow uh this allows you to select from around 200 or so proteins um and to pick 30 of those 200 for your custom panel but again uh this will be a discussion that we can have uh together with oink on on selecting that panel all right uh thanks Dr Corney we have another question here for you what sample types are compatible with the Oink workflow I think the most common sample types that we've handled and those reported in the literature are around plasma and serum collected from a wide range of blood tubes including ET CA and and others um those are both validated samples by oink but the beauty of the Oink system is that it is adaptable to other sample types too and other sample types have included CSF cell culture superintendance tissues or or cell lysates these have all been successfully analyzed as well so I think there's some additional considerations to make uh for those different sample types we want to have that discussion to make sure the controls are being included and so forth but very wide range of sample types can be analyzed I know we are over time here so this will be our last question for the session again for Dr Corney can Bank speci be used or must samples be freshly collected so it's certainly possible to use banked archival samples and that's kind of what I I spoke about to to a degree as well in my presentation that there's a really large cohort of samples that are sitting in biab banks that are great candidates to use for oink and and for determining the prodium in those samples so it's not necessary to collect samples freshly there's definitely some consideration of the impact of the the sample Age and and how that can impact the quality of the data so again I think this is where we should have a discussion and think about how to analyze the data um correctly um so typically we would want to include sample age as a covariant within uh that analysis but it's certainly possible that is all the time that we have for today again any questions not answered during the session will be followed up after I want to say thank you to our speakers today and to everyone who joined us from around the world with that we'll close out the session thanks again everybody enjoy the rest of your day