Category: Cytogenomics

Phase Genomics’ recent release of the RUO cytogenomics platform, CytoTerra™, was accompanied by a webinar which covered an in-depth analysis of current technologies and emerging opportunities in reproductive genetics and oncology.

“The genome is the blueprint of life,” beginning the webinar, Ivan Liachko describes Phase Genomics’ history of discoveries and contributions to genomic research. Through the development of various genomic, metagenomic, and epigenomic platforms, Phase Genomics has risen as a leader in next generation sequencing (NGS) solutions. Now, the company’s latest platform leverages their ultra-long-range sequencing technology to be used for cytogenomic applications. 

Chromosome rearrangement is a known driver of many diseases, including cancer, infertility, developmental delay, and immunologic complications. Thus, the detection and treatment of these rearrangements is essential in the advancement of modern medicine and therapeutics. However, current methods are limited in scale, throughput, and resolution. Additionally, challenges in sample types and analysis constraints present a cascade of costly tests to run in order to assemble a complete view of the genome. Some of these challenges include culturing dividing cells for cytogenomics, obtaining advanced knowledge of the targeted abnormality for fluorescence in situ hybridization, and working within the limited scope of rearrangements detectable by chromosomal microarray analysis. Further complicating the process of genetic analysis, most cancer biopsies are stored as formalin-fixed paraffin-embedded (FFPE) samples—a wax-like encasing which kills the cells and traps the DNA. Historically, there has been no way to access the DNA to perform NGS testing in these sample types. However, recent cytogenomics platforms created by Phase Genomics do not require a priori knowledge, improve chromosomal abnormality detection, and unlock information in FFPE samples, offering a promising solution to many challenges in the oncology and reproductive genomics spaces. 

Phase Genomics’ new cytogenomics platforms, CytoTerra and OncoTerra, are powered by ultra-long-range sequencing—using proximity ligation data and artificial intelligence  to analyze the breadth of chromosome arrangements in a single assay—which eliminates the need for sequential testing, includes a scalable approach to genomic detection, and unlocks information stored in difficult sample types, including FFPE and frozen samples. 

Watch the webinar for more information on the expanding possibilities of chromosomal aberration detection and contact Phase Genomics to start a project.

Transcription

00:00:01:18 – 00:00:15:08

Speaker 1

Hi, everyone. My name is Ivan Liachko and I’m one of the founders and chief scientist at Phase Genomics. I’m joined today by Jill Tapper, our cytogenetics product manager. Today, we’re going to tell you about our new next generation cytogenomics platform.

00:00:15:19 – 00:00:36:23

Speaker 1

This new platform is powered by a unique next generation sequencing technology and has the power to transform how clinicians and researchers approach oncology and reproductive genetics. All right. Let’s get started. So, for those of you who are not familiar with Phase Genomics and what we do, essentially our thing is building genomes.

00:00:37:09 – 00:00:54:16

Speaker 1

What we do is we capture unique genomic information to reconstruct genomes and genome structure in order to transform research and clinical applications. We got our start by building cutting edge genomic tools to assemble genomes for non-model organisms.

00:00:55:00 – 00:01:13:09

Speaker 1

So, we came out a few years ago with the first chromosome scale non model genome scaffolds as a way of basically putting together an end-to-end chromosome scale genome for anything—for plants, animals, fungi. We’ve also developed tools to haplotype phase a genome of any size.

00:01:14:13 – 00:01:37:00

Speaker 1

This is something that at this point is fairly well accepted by the field. We’ve published this over 100 times. It’s even made it into the popular press. And what this sort of technology is based on is a method that has many names, the most descriptive of these is ultra-long-range sequencing.

00:01:37:10 – 00:01:52:11

Speaker 1

What it does is it allows us to sequence DNA molecules that are really far, far away from each other. And the way it works is you will take a cell that is intact and within the cell. The genome is condensed into this three-dimensional structure, right?

00:01:52:11 – 00:02:11:01

Speaker 1

Remember, a genome is just linear molecules being squished into a ball and they condense into these three-dimensional shapes. The way the technology works is it captures physical junctions between DNA molecules that are close to each other in three-dimensional proximity.

00:02:11:02 – 00:02:26:22

Speaker 1

So, in three-dimensional proximity with each other, we can capture these junctions and sequence them. And what that does is it tells us it gives us a way to count how often every part of the genome is close to every other part of the genome.

00:02:27:10 – 00:02:47:10

Speaker 1

And so, if you know how often two sequences are physically touching each other, you can figure out how close they are because the sequences that are closer touch more and sequences that are further touch less. And if you know this, if you know this sort of three-dimensional distance between all the sequences in the genome, you can

00:02:47:10 – 00:03:08:09

Speaker 1

reconstruct that into a genetic map, right? Sequences that are closer touch more sequences that are further touch less, and that enables you to do to basically use computational tools to reconstruct that information into a karyotype. And so, if you have a genome that you don’t know how it’s supposed to go together, you can use

00:03:08:09 – 00:03:23:05

Speaker 1

this information to scaffold it by arranging all the pieces. But if you have a genome like the human genome where you know what it’s supposed to look like, this is a really robust way of detecting rearrangements big chromosome scale karyotype, style rearrangements.

00:03:24:18 – 00:03:41:10

Speaker 1

There’s a lot of other things you can do with this technology. I’ll mention them briefly, just for reference. So, the first thing that I’ve mentioned is it allows us if you have this data, you can reconstruct essentially high-resolution genetic map for whichever organism it is you’re working with.

00:03:42:07 – 00:03:52:14

Speaker 1

But it also allows us to assemble and phase genomes de novo. So, when you don’t have a sort of a scaffold, a genome with some new organism you’ve never seen before, it allows us to assemble the genome from scratch.

00:03:53:10 – 00:04:10:22

Speaker 1

This technology allows us to understand the three-dimensional architecture of the genome. So basically, it allows us to study the 3D structure of a genome, which is a very, sort of very interesting biological property that every genome sort of lives in.

00:04:11:13 – 00:04:25:16

Speaker 1

We also have a number of cool tools in the microbiome space. So, this technology I won’t go into this technology has lots of really neat properties that allow us to discover new bacteria, new viruses, new mobile elements.

00:04:26:02 – 00:04:44:04

Speaker 1

It allows us to track the movements of mobile elements such as antibiotic resistance genes in infectious disease microbial environments, in addition to building a suite of wet lab molecular tools, which we, we sell all sorts of kits and services in the space.

00:04:44:16 – 00:04:58:23

Speaker 1

We also are a very informatically focused company. And so, we develop the tools that are needed to take this unique information type and actually turn it into actionable insights. So, we’ve developed everything so Proximo, our genome scaffolding platform.

00:04:59:04 – 00:05:10:12

Speaker 1

We’ve developed tools such as Falcon Phase, which allow us to phase genomes, haplotype-phase genomes. We have a number of tools for doing karyotype instead of genetic type studies, and that’s what we’re going to talk about today.

00:05:10:24 – 00:05:31:23

Speaker 1

And then we have a suite of methods that leverages this technology for microbiome discovery. ProxiMeta is for discovering new bacterial genomes, ProxiPhage for discovering new phages and then ProxiLink is for looking at the transmission of antibiotic resistance in complex microbial communities.

00:05:33:11 – 00:05:49:12

Speaker 1

And so, the focus of today’s talk is really going to be on one of the properties of this technology. It allows us, you know, we really want to understand the structure of chromosomes, the structure of genomes. This is extremely important in the medical space, right?

00:05:49:12 – 00:06:07:18

Speaker 1

There is a whole army of diagnostics that have been designed specifically to look at the structure of chromosomes. But these diagnostics of the day of today, the sort of the most well adapted ones, you know, they’re limited, they’re limited in scale, they’re limited in throughput, they’re limited in their resolution.

00:06:08:01 – 00:06:27:07

Speaker 1

And our technology can solve these problems to a large degree. And that’s what we’re going to be displaying today. So, we recently launched a method called CytoTerra It’s a new platform that we’ve developed that enables us to leverage this technology to really benefit folks who are trying to do cytogenomic testing.

00:06:27:18 – 00:06:30:19

Speaker 1

And that’s what Jill is going to talk to you next.

00:06:32:20 – 00:06:58:09

Speaker 2

Thanks, Ivan. I want to start with some basic background context as to how our platform plays a role in advancing precision medicine focused research and diagnostics. And it’s really built on this fact that we know very well, which is genomic and chromosomal rearrangements are drivers of every aspect of disease, from etiology to prognosis to therapy selection.

00:06:59:00 – 00:07:22:03

Speaker 2

And we see proof of this in long standing examples like the 9:22 translocation and CML shown on the left and those patients’ response to a very specific therapy Gleevec. We also see in an example like the spectral cure type tumor on the right, where there are likely many abnormalities as opposed to one specific abnormality contributing to this

00:07:22:03 – 00:07:46:16

Speaker 2

tumor’s development. And those are examples in cancer, but we know these rearrangements play an equally important role in many other diseases and conditions like infertility, recurrent pregnancy loss, developmental delay, and those are just a few. So, we have a collection of or genomic methods we use to try and help uncover these genomic disease drivers.

00:07:47:01 – 00:08:07:19

Speaker 2

These are karyotyping or chromosome analysis, FISH, and microarray. So, among these current solutions, we have a combination of high and low throughput approaches, high- and low-resolution approaches. But even with this range of resolution and throughput, each solution still has its drawbacks.

00:08:08:15 – 00:08:34:00

Speaker 2

For cytogenetics, we need live cells to grow in culture. We also need highly skilled personnel to do the analysis and interpretation of the results. For FISH, we need advanced knowledge or advanced suspicion of the abnormality. With Array, it’s difficult to detect things like balanced rearrangements inversions, low-level mosaicism

00:08:35:01 – 00:08:55:05

Speaker 2

So collectively, we’re trying to balance these limitations such that we get as comprehensive of a view as possible with respect to the size and type of abnormalities that may be present. And typically, to get that comprehensive view, it’s necessary to use these methods in a sequential format.

00:08:55:13 – 00:09:15:09

Speaker 2

So, when we look at that in terms of workflow and timeline, we end up with a resource intensive, very long, very expensive cascade testing approach. And this is where our Phase Genomics platform has a major impact in that it offers an efficient, streamlined workflow.

00:09:15:21 – 00:09:37:19

Speaker 2

And that’s because ultra-long-range sequencing can provide the large structural and copy number variation detection capabilities that we find with cytogenetics, along with the molecular precision of FISH and Array, all in a single assay. So, we’re eliminating the cost and time associated with the current cascade approach.

00:09:39:23 – 00:10:01:05

Speaker 2

Ivan touched on the technical aspects of ultra-long-range sequencing earlier on. But at a high level, we’re able to leverage that method’s unique capability to capture the physical proximity of DNA sequences in the genome. We then use our proprietary analytic software to convert the proximity counts to genomic distances.

00:10:01:12 – 00:10:31:17

Speaker 2

And as a result, we’re able to detect a wealth of abnormalities like balanced and unbalanced translocations, inversions, insertions, aneuploidy, and a lot more. And these capabilities are part of a comprehensive sample-to-report service workflow, with results ultimately being returned using standard ISCN in sequencing nomenclature and returned in a familiar clinical style report format.

00:10:32:16 – 00:10:53:02

Speaker 2

Well beyond the report, we’re creating a valuable data resource for novel variant and biomarker discovery. So, does it really work? Yes. And here is some data from proof of concept work we conducted with an academic health system clinical genetics lab.

00:10:53:21 – 00:11:14:07

Speaker 2

And in these 100 plus samples, we see our Phase Genomics platform not just meeting, but exceeding the detection capabilities of the current set of genomic approaches, with some low-level translocations not previously identified being detected. And we’ve had similar success with other sample types.

00:11:14:13 – 00:11:31:16

Speaker 2

And you can see a list of some of those here. Everything from whole blood and cheek swabs to POC tissues. But it’s not just compatibility with numerous sample types that makes this platform so flexible. It’s flexibility in sample condition as well.

00:11:32:05 – 00:11:50:22

Speaker 2

The platform works with fresh samples, frozen samples and very notably, it works with FFPE samples. And here’s just a small representation of some of the FFPE tissue types we’ve worked with in the past. So why is sample type flexibility compatibility…

00:11:51:02 – 00:12:16:14

Speaker 2

Is detection capabilities particularly meaningful to reproductive health in oncology? Well, to start, there are some sample related challenges in these areas. Obtaining fresh sample material for cell culture is difficult in POC samples, for example, this tissue is often non-viable, with very high failure rates for tumor and bone marrow samples.

00:12:16:20 – 00:12:36:03

Speaker 2

There’s often a limited amount of sample to work with, and the cells are not necessarily unviable, but they’re disease cells that are often very challenging to work with. Formalin fixation and paraffin embedding are also prevailing collection methods for these sample types, so cell culture is immediately out of the question.

00:12:36:20 – 00:13:03:08

Speaker 2

And then there are concerns about obtaining sufficient quantity and quality of the high molecular weight DNA that’s needed for things like Array and many NGS assays. These are also areas where balanced rearrangements play a significant role. There are cryptic translocations or seemingly balanced rearrangements in areas of visual homology that can be causative, so knowing if something is

00:13:03:08 – 00:13:31:19

Speaker 2

truly balanced is critical. We also know that gene fusions resulting from balanced rearrangements drive cancer and tumor development. So, in the end, these challenges present as missed opportunities, opportunities to uncover diagnostic and prognostic information, to discover biomarkers for therapy and treatment development, and to make genotype phenotype disease associations to further disease understanding.

00:13:33:24 – 00:13:54:10

Speaker 2

So, in comparison to the current cytogenetics methods, our Phase Genomic platform can avoid these and many other missed opportunities by offering genome wide simultaneous detection of the multiple types of genomic rearrangements that cause and characterize disease, and it can do so in a single assay.

00:13:59:20 – 00:14:11:04

Speaker 2

The platform has capabilities well beyond what we expect of our current cytogenetics methods. I’m going to hand things back to Ivan so he can talk about what some of those expanded possibilities are.

00:14:14:18 – 00:14:29:20

Speaker 1

Thank you, Jill. So, as you have just seen, this platform is very useful in the field of cytogenetics. But there’s more there’s a lot of things you can do with it beyond just sort of an improved way of conventional testing.

00:14:30:14 – 00:14:57:20

Speaker 1

one of the main challenges in oncology is that while you know there are obviously so many different cancer types, only a small subset of available cancer samples get processed, cytogenomically and get analyzed by cytogenomic assays. And the reason is that the vast majority of cancer biopsies in cancer samples are stored as FFPEs.

00:14:58:03 – 00:15:20:07

Speaker 1

They’re stored in formalin fixed paraffin embedded format. And what that does is that kills all the cells and also ruins the DNA for long read sequencing for optical genome mapping. And so, it makes most karyotypic assays and large-scale structural analysis virtually impossible.

00:15:20:21 – 00:15:31:04

Speaker 1

And this is one of the things that our technology can overcome. So first, let’s take a look at what this data looks like we’ve been talking a lot about. So, the technology and what it can do? But here’s what.

00:15:31:16 – 00:15:54:05

Speaker 1

At its core, here’s what the data looks like. When you plot this type of ultra-long-range sequencing information, you can you generate these sorts of maps these heatmaps. Imagine if you’re not familiar with this, imagine a just a matrix where an x axis and the y axis you just lay out the chromosomes like left to right

00:15:54:19 – 00:16:18:03

Speaker 1

and you’re seeing this coordinate system. This is chromosome, you know, 1, 2, 3, 4, 5 and chromosomes along this line. And these boxes are showing you how much interaction there is within that combination of coordinates. So, this heat inside of this box tells you that there’s a lot of interaction between chromosome two and chromosome two other parts of

00:16:18:03 – 00:16:33:07

Speaker 1

the same chromosome. They’re touching each other because they’re close. But there’s not a lot, for instance, between chromosome two and chromosome four. Right? But then if you look at a cancer sample like this one, you will see that there is this hotspot in this box.

00:16:33:07 – 00:16:48:05

Speaker 1

And what that means is that this area of chromosome two and this area of chromosome four are touching each other way more than they’re supposed to. They’re closer together. So, this was caused by a translocation and there are different types of these events.

00:16:48:06 – 00:17:07:20

Speaker 1

Sometimes they look like squares, sometimes they look like bow ties, et cetera. We have essentially trained the analytics, right, we’ve built this A.I. that recognizes these things, and that’s how we can generate these. These karyotypic reports, karyotype maps, but we can do this in FFPEs.

00:17:07:24 – 00:17:26:18

Speaker 1

And the reason why we can do this in the FFPE is because the first step of our method involves fixation with format with formaldehyde or formalin, which is what the f is in FFP. And so, we’re able to not only generate these kinds of cytogenetic profiles on fresh frozen tissues and cells and these sorts of things

00:17:26:23 – 00:17:44:22

Speaker 1

, but also, FFPE slices and even a single FFPE slice can generate a really cool complex karyotype. So, this is an example from one of our collaborators. This is a solid cancer. FFPE slice from a solid cancer is just a slice.

00:17:44:22 – 00:17:59:23

Speaker 1

You don’t need to consume the entire FFPE block. Everything works with sort of how people are used to looking at it, and you can see again, in this case, the data is shown in a different color. It’s now orange instead of blue, like in the previous slide.

00:18:00:09 – 00:18:20:21

Speaker 1

But basically, again, these boxes are the chromosomes and these little shapes. These events out here that are marked by black arrows represent the structural chromosomal aberrations within this FFPE slice. And so, you know, there’s there are these little bow ties and squares like before, and you can sort of divide them.

00:18:21:13 – 00:18:30:15

Speaker 1

Here’s what they look like when you when you zoom in. If you were to kind of visually analyze it, this is what they would look like. Of course, we use software, but you can actually look at them and see them with your eyes.

00:18:31:19 – 00:18:46:23

Speaker 1

And so, this is what a balanced translocation looks like an unbalanced an inversion because this technology is sequencing based, Illumina sequencing based. It allows you to do all the other things that you do with Illumina so you can detect deletions and copy number changes.

00:18:46:23 – 00:19:10:18

Speaker 1

You can detect amplifications; you can detect aneuploidy and other similar things. And so, you can generate these really complex karyotypes right off of FFPE without sort of, you know, in the very in a very manageable way. And so, what we’re going to show you in this video here is a comparison of a data set from a fresh

00:19:10:18 – 00:19:28:11

Speaker 1

frozen lung cancer sample and an FFPE matched sample from the same biopsy. And so, what you’re looking at again is just like before the boxes in the middle of the chromosomes and the events out in sort of in this yellow space.

00:19:28:16 – 00:19:43:15

Speaker 1

These are your translocations and other karyotype aberrations. So, we’re going to zoom in. This is being visualized in a tool called high glass. And so, what we’re doing is we’re sort of zooming in in sync with a fresh frozen in an FFPE sample.

00:19:44:03 – 00:20:06:07

Speaker 1

And when I hope you can see is that we’re able to even in a FFPE sample, detect really, really sort of crisp, large scale structural rearrangements. And this is going to zoom in on its kind of just so you can see the sort of the beginnings and the ends of these things, and we can then map

00:20:06:07 – 00:20:17:01

Speaker 1

them to, of course, their genomic coordinates and what you’re looking at here is this is the entire genome. And so, you can see there’s lots of things happening here. There’s some of them are small, some of them are big.

00:20:17:11 – 00:20:39:20

Speaker 1

So, this allows you. This method allows you to reconstruct high quality karyotype maps specifically for FFPE samples, as well as for any kind of fresh frozen sample. And finally, what I’m going to end with is this technology because it is based on Illumina sequencing.

00:20:40:01 – 00:20:55:14

Speaker 1

It is compatible with target capture methods. So, if you are looking for if you’re looking for specific snips, if you’re looking for mutations, if you’re looking for a loss of heterozygosity in specific regions, you can actually pair the two methods.

00:20:55:20 – 00:21:15:17

Speaker 1

You can use pretty much any capture panel out there. They all work extremely well with this technology, and that allows you to do simultaneous snip profiling looking for specific mutations in specific genes while also generating a structural map of karyotype for your given sample.

00:21:15:17 – 00:21:39:02

Speaker 1

Be it fresh frozen, be at cells, be it FFPE directly out of one sample, so you can combine a target capture panel and a cytogenetic work, up from the same sample with this technology. So just to summarize and finish off this talk, what we’ve shown you is that applying ultra-long-range sequencing to various biological samples

00:21:39:09 – 00:22:00:14

Speaker 1

is basically a really useful way of enhancing your cytogenomics game. It’s a scalable platform for cytogenomics. It’s simple. It can be performed in your lab. You don’t need a special machine for this. Aside from Illumina sequencers, which can which are sort of, you know, pretty common these days, we provide both services kits as

00:22:00:14 – 00:22:20:07

Speaker 1

well as companion cloud-based analytics, so you don’t have to sort of figure out your own computational pipeline. We integrate both the analytics as well as the molecular methods. It works on all sorts of difficult sample types. It works, and if it works on cheek swabs, it works on all sorts of samples that are usually very difficult

00:22:20:10 – 00:22:36:07

Speaker 1

to process by traditional cytogenetic methods. As I mentioned, you don’t need to make sort of buy a special machine for this. You don’t need high molecular weight DNA, so it’s a much easier way of generating large scale cytogenomic data.

00:22:37:08 – 00:22:52:16

Speaker 1

If this is something that you’re interested in, please let us know. And the reason why we’re giving this webinar is to announce sort of our early access program. And so, we are recruiting folks to do all sorts of cool research with us.

00:22:53:00 – 00:22:58:19

Speaker 1

So, this is of interest, please. You can scan that barcode. You can go visit us on the website and shoot us an email.