Hello, everyone. My name is Alex Hastie. I'm the Vice President of Clinical Affairs at BioNanogenomics, and today I'd like to introduce the BioNanogenomics technology for optical genome mapping for the detection of high-resolution structural variation. I want to start off with a timeline to put chromosomal studies into perspective. The first observation of chromosomes and the naming started before 1900. In the 1970s, chromosome banding techniques were developed, which allowed the determination of genetic disorders based on observing chromosomes. More recently, chromosomal microarrays have been invented and used for higher-resolution detection of genetic disorders, genetic abnormalities, and over the last decade or so, various NGS technologies have been used to try to further unravel genetic disorders, however, not able to truly replace karyotyping and microarray and FISH technologies for structural variation and copy number variation so far. In the last few years, BioNanogenomics has been commercialized and applied to a study of chromosomes and structural variation in the context of genetic disorders. I'd like to tell you more about the BioNanogenomics optical genome mapping technology. This video on the left is showing really the foundation of the technology, which is the use of a nanochannel array. The image on the bottom right is an electron micrograph of the nanochannel array. These are silicon wafers that are etched with a series of parallel nanochannels that are about 35 nanometers wide and are extremely uniform. You can see in this electron micrograph how straight these channels are. These channels are used to linearize very long DNA molecules, which is what you're seeing in this video. The video shows first DNA and solution right here, and then it's being pulled into the nanochannel array by unwinding the DNA in these structures of pillars and then feeding the DNA molecules into channels. These are microchannels and then finally nanochannels where once the nanochannels are loaded, the electrophoresis is paused and the DNAs are imaged. The image here is what the data actually looks like. The blue lines represent very long and very uniformly linearized DNA molecules, and they're labeled with a fluorophore, a green fluorophore that allows us to mark and recognize what part of the genome we're looking at. This is all done in the SAFIRE instrument. This is a slide courtesy of Alex Hoyshin from Radboud University Medical Center that compares and puts the bio-nano optical genome mapping technology into perspective. It's very analogous to a karyotype where you are able to look at a chromosomal banding pattern from a cell, and that banding pattern is used to recognize what part of the genome is being observed. Just like that, our bio-nano optical genome mapping technology produces a banding pattern where these fluorophores are added at a specific six-base sequence motif. And then in the nanochannel array, we can image and measure the distance between these sequence motifs and this pattern of sequence motifs can allow us to recognize what chromosome we're looking at. This is very analogous to a karyotype and can do all of the same things that a karyotype can do. However, we have about 1,000 more bands than a karyotype and the linearization allows us to find copy number variations at about 10,000 fold higher sensitivity. So this is the bio-nano workflow, which I'll briefly describe. The workflow starts with isolation of very high molecular weight DNA from samples, then labeling that DNA with an enzyme that recognizes a specific six-base sequence motif, data collection on a SAFIRE chip, which contains the nanochannel array and is done in the SAFIRE instrument. The images of the molecules are processed and the DNA molecules as well as the positions of the fluorophores are extracted. Then those single molecules are used to detect structural variation by comparing them to one another, building either a de novo assembly or just a reference comparison to find all classes of structural variation. In parallel, there's a second pipeline that's a copy number variation pipeline, which is very analogous to a microarray where we can just detect copy increases and decreases. These two pipelines are run in parallel and then a final pipeline for annotating structural variance that allows us to know if those structural variants affect genes and how frequent they are in a control population. The technology can detect all classes of structural variation. Insertions and deletions are found by an increase or decrease in the size of a DNA segment down to 500 base pairs. Copy number repeat array, repeat arrays can be measured accurately because DNA molecules being used are long enough to easily span a repeat array and anchor into the flanking regions of the DNA. Duplications can be detected by the recognition of a pattern in a sample or a case's map occurring twice, where in the reference it occurs only once. In this case, the duplication is a tandem direct duplication. So the first copy here and the second copy right next to it in the same orientation, we can determine the orientation by that mapping of the DNA pattern. If this duplication was inverted, we would call that as well. If it was inserted on another chromosome, we would call that as well. The technology detects balanced structural variants like translocations and inversions. A translocation occurs when a map from the case aligns on one chromosome on the left side and another chromosome on the other side. An inversion is a case where the labeling pattern is oppositely oriented with respect to the reference pattern. So this table shows structural variant classes that are clinically relevant. And just quickly, we detect all types of aneuploidies. We detect monosomies and trisomies. We don't currently detect triploidy and tetraploidy whole genome, but we will be working on that. We can detect ring chromosomes. We can detect copy number variants such as deletions and duplications, whether they're interstitial or terminal, as well as insertions, balanced or unbalanced translocations, inversions, whether they're pericentric or pericentric. We are working on regions of homozygosity. We will be adding that to a software update in the future. The technology is very good at detection of macrosatellite and microsatellite repeat expansions and contractions such as FSHD and Fragile X. And we do not, BioNano optical genome mapping does not detect single nucleotide variants or indels. That is for sequencing technologies instead. And just one example case study to help understand how the data looks and how it's processed. Any data set from BioNano would be displayed in a whole genome overview on a circus plot. The circus plot shows the cytobands on the outside ring. The next ring shows all of the structural variants, interstitial structural variants that are detected in this case. The next ring shows the copy number profile for the case showing two copies of all of the autosomes, one copy of each X and Y. And then the inner circle would show any translocation fusions if there were any. On an average constitutional case, we would detect about 6,000 structural variants, most of these being polymorphic. So the software has a tool for filtering. And so we can just remove polymorphic variants by setting a filter that asks to see only variants that are absent from a control database. Once you do that and also overlap the structural variants with a list of genes that are clinically relevant, we can just quickly prioritize down in this case to three structural variants remaining. We're able to just hover over these dots and figure out what genes they're affecting and how large they are. And then if we see one of interest, we can just click on that and we can zoom in on another genome browser, which will show exactly what gene we're looking at, in this case DMD. We can see that in this case, the case has a deletion, 164.9 kilobase pairs. In that deletion is exon 45. So this is a deletion of exon 45 DMD. So that's where I'll stop with the introduction. And then we'll have a talk from Nikhil Sahajpal. This is the Georgia Esoteric Lab in Augusta, Georgia. It's led by Ravi Kohle. During 2020, they have implemented SARS-CoV-2 testing, really breakneck speed. They went through the process as soon as emergency use authorization was allowed. And they have been very busy testing the local community for SARS-CoV-2. Nikhil Sahajpal was a key member of implementing that testing. And in the midst of all of that, he was also able to publish 12 papers since the beginning of the pandemic. He also, if that wasn't enough, he also led the implementation of a new technology in the Georgia Esoteric Lab, which is the BioNanoSAPPHIRE system. He has been using the system for many different applications and already has preprints or publications on several applications. He's a part of a study on acute myeloid leukemias using optical genome mapping. Another study, which he's the first author on looking at the host genome structural variation in patients that respond poorly or well to COVID-19 infection. And also has another commentary on using optical genome mapping for prenatal genomic analysis, which he will now be telling us more about work that he's done on prenatal testing analysis. So thank you very much. Thank you so much, Dr. Hastie, for this generous introduction. It is my privilege to share our study at this forum. I will be talking about next-generation cytogenomics. And before I begin, I would like to say that the term next-generation cytogenomics has rather become synonymous with BioNano's optical genome mapping. To put that in context, I will be discussing the utility of optical genome mapping in the characterization of prenatal cases. The current scenario of prenatal testing as recommended by ACMG, ACOG, and ISU-OG is that prenatal testing is offered to every pregnant woman, regardless of age or previous history of pregnancies. In that respect, NIPT screening is the standard of care. And for positive NIPT screens, high-risk pregnancies and abnormal ultrasounds are recommended for invasive diagnostic testing, using either amniocytes or chorionic villi samples. The recommended invasive tests are karyotyping, FISH, and chromosomal microarrays, which are either performed simultaneously or in a tiered fashion, as each technology has its own advantage and respective limitations. Karyotyping is ideal for the identification of whole-genome low-resolution structural variations, but the structural variation location and size remain inaccurate, and some translocations remain cryptic. The FISH panels are targeted, but the common microdeletions and microduplications remain difficult to detect. Chromosomal microarrays are ideal for the detection of copy number variations, but they are limited in their ability to detect balanced translocations, inversions, and low-level mosaicism. With this, I would like to discuss straight the challenges with the current technology in prenatal care, with the help of an example. A classical scenario where a screen positive NIPT or a high-risk pregnancy is recommended for invasive prenatal testing. CVS or amniocentesis is performed and the cells are cultured. A stag fish shows a normal report. Ketotyping is performed and again in this case the ketotyping shows a normal profile. However, with chromosomal microarray we observe a deletion on chromosome 7. So a genetic abnormality is detected with the help of CMA and a diagnosis is reached. If the CMA was performed simultaneously to ketotyping it adds to the cost of the test. Whereas if it was ordered as a follow-up after a normal ketotype it adds to the time to reach a diagnosis which is very critical in prenatal care. So this raises a question that could we reach a definitive diagnosis in one step with one technology, assay or analysis? And these are the results of the same sample with the BioNano's optical genome mapping. And on the circus plot itself we can see the deletion on chromosome 7 and a duplication on chromosome 2 which can also be viewed at the CNV track and the SV view. And also the chromosome 2 duplication can be viewed at the CNV track and the SV view. So to answer my question, yes we can reach a definitive diagnosis in one step and with one technology, assay or analysis. But to probe further into this case the duplication on chromosome 2 was not called out by CMA. Whereas it was called by the optical genome mapping and can be viewed at the level of CNV track. But in addition what we can see is that this copy number gain is rather a tandem duplication. So I want to make two points here. The first is that you can reach a definitive diagnosis in one step and you can identify the clinically relevant structural variations that are detected by the current methods. The second point is that with this technology you are bound to identify additional structural variations which are either not identified or called by the current cytogenetic technologies. With this in purview I would like to move now to the SAFIRE workflow and the validation study. So the SAFIRE workflow is absolutely similar to what Dr. Hasty presented in the introduction. The only modification that appears in the prenatal setting is that the starting material, the cells are amniocytes or chorionic villi cells which is actually the starting material for any conventional cytogenetic method at present. In our prenatal study we have tested 19 samples so far of which one is a trio. All cultured amnio samples were tested with cytogenomic methods which is kerotyping with fish or chromosomal microarray. All samples tested on OGM were blinded to the cytogenetics data. Further the accuracy and precision was measured at the level of sites, operators, and instruments. Now as optical genome mapping identifies a plethora of structural variations I wanted to discuss the representative cases. I'll start with common trisomies, trisomy 21 and trisomy 13. An example of sex chromosome aneuploidy, a case of mosaic Turner syndrome and then move to complex structural variants. Other SV classes such as trisomy 8 and a case a normal profile where no clinically relevant structural variation or copy number variation was detected with optical genome mapping or traditional cytogenetic genetic methods. Case one is an example of trisomy 21 which is the most common live-born trisomy and has an instance of one in 600 to one in 800. I want to highlight two points in this slide. One is that when you open the bio nano access software the first data that you see is actually a circus plot. So on the circus plot itself you can see the trisomy 21 and again on chromosome 15. The second point I want to make is about the filtration criteria. So if you filter out against the bio nano control population you land up with the rare SVs in this genome which are enumerated here. For clinicians that are used to looking at the CMA data would appreciate this CNB track and can see that the trisomy 21 is called out here and in the zoomed in view of the CNB track looking at chromosome 21 which is very similar to how the CMA report is. Case two is an example of trisomy 13 which is again a live-born trisomy and has an instance of one in 15,000. Here again on the circus plot we can see trisomy 13 and a copy number loss on chromosome 15. On the CNB track we can see trisomy 13 and a copy number loss on chromosome 15. Interestingly if you see the SV track the homozygous deletion is called out here both at the level of such a variation and the CNB track. This homozygous deletion was not called out by karyotyping. Case three is an example of mosaic Turner syndrome. On the circus plot itself we can see the loss of chromosome X and again we can see in the CNB track a mosaic loss of chromosome X which is mosaic Turner syndrome. With these examples I would now like to move to cases with complex structure variations. In this example we see a translocation 15 to 21 and a rearrangement on chromosome 15. This is a zoomed in view of chromosome 15 and chromosome 21 showing the translocation and copy number gain on chromosome 15. The CNB track also shows the copy number state of chromosome 15. If we dive deep into chromosome 15 and look at more closely we see that there are two segments with a copy number state of three. If we look at at the SV track what we see is that these two regions are actually fused and if we zoom into this map we see it is in the inverted orientation. So what is represents is that if you consider this lower arrow as the chromosome segment with a copy number state of three is actually fused with this segment again with a copy number state of three as it is inverted and then continues in this orientation of the chromosome. For the sake of simplicity you can derive your own ideograms based on optical genome mapping data. Here we have represented chromosome 15 as segments A, B, C, and D and chromosome 21 as segment E. On chromosome 15 segment A and C are the segments with CN state of three. So translocation 15, 21 is actually simple where this segment D of chromosome 15 translocates to 21 and results in derivative 21. Because we saw the inversion which results in a foldback fusion with this segment of chromosome A a derivative 15 results. Now when this blinded data was compared to karyotype and FISH it was found 100% concordant with karyotyping and FISH. On the karyotype you see two arrows that show derivative 15 and derivative 21. The derivative 15 results from a diacentric 15 with a duplication of chromosome 15q and the other derivative 21 results from a translocation of 15, 21. The FISH studies performed on the metaphase confirm the duplication of chromosome 15 using the Snerpin probe tabled in red. Another example of a complex structural variation is the case one where we had observed trisomy 21 but we did not discuss the copy number gain on chromosome 15. Again the CNB tracks showing the zoomed in view of trisomy 21 and gain on chromosome 15. Looking more closely at chromosome 15 we see that there are actually two segments one with a copy number state of four and one with a copy number state of three. If you look closely that the region with copy number state of three is actually inverted and if we zoom into this region we see that there is an overlap of the copy number region of three with copy number region of four. Again the lower arrow represents this part of the chromosome with CN state of four and this region of CN state of three is actually inverted and fused and continues with this region of chromosome 15. Again an ideogram can be derived based upon the OGM data and we see a two normal copies of 15 and an additional derivative 15 which results from the foldback fusion of this segment of 15B and aligns and fuses with the segment of 15A resulting in derivative 15. The trisomy 21 was clear from the circus plot and the CNB track. What I want to highlight is that the OGM data is 100% concordant with karyotype and FISH but OGM in one SA detected both the trisomy 21 and the identity of the marker chromosome. As you can see in the karyotype data it identified three copies of 21 but the marker chromosome is not identified. The metaphase and interface FISH identified the marker chromosome to be chromosome 15 material. Case 5 is an example of trisomy 8 and we can see in the circus plot the copy number gain on the entire chromosome 8 and a gain on chromosome 13. Again the CNB track shows the gain of entire chromosome 8 which is trisomy 8. However on chromosome 13 what we observe is that this gain is not a simple copy number state of three but rather a complex structural variation. As this region of the genome duplicates in tandem but it becomes complicated as this segment of the genome is inserted between the tandem duplication in inverted orientation which becomes evident from this inverted orientation of the labels. I wanted to say here that as we look on these maps these complex structural variants start becoming obvious and they do not remain complex at least to your eye. Again this is the karyotype for the same case and we see a trisomy 8 however the duplication on chromosome 13 was missed by karyotyping. Case 6 is an example where no clinically relevant such a variation or copy number variations were detected with optical genome mapping or even not with karyotyping and represents a true negative case. In this validation study five samples were run at two sites and there was 100% concordance between sites, between operators, between instruments and also at the level of data. When compared to karyotyping FISH and CMA optical genome mapping shows a 100% concordance with respect to identifying clinically relevant structure variations and copy number variations and it identified in four cases that I have discussed additional findings which were missed by karyotyping or CMA. So to conclude this presentation I can begin by saying that invasive prenatal testing is recommended after a positive NIPT screen and for high-risk pregnancies. Karyotyping FISH and CMA are the recommended tests employed by routine cytogenetics labs. Optical genome mapping has the potential to identify all the structural variants that are currently identified by the traditional cytogenetic methods but also better characterize these structural variations. The OGM data in this study is 100% concordant with karyotyping FISH and CMA. This data is a path to an LDT on OGM for prenatal applications and it has several advantages that it has a reduced hands-on time, a faster turnaround time with sample to reporting in four days and it is cost effective as compared to the combined cost of karyotyping FISH and CMA. We have also published recently a commentary on the application of optical genome mapping in the prenatal setting which discusses several examples of different structure variation classes including fragile eggs. With this I would like to acknowledge Dr. Rabindra Kohli who is the Director of Cytogenetics and Molecular Lab, Dr. Ashish Mondal and our collaborators at University of Mississippi, Dr. Ferribose Colvier, Dr. Suzanne Hurley and obviously the Bionanogenomics team. I would also take this opportunity to invite you to our posters which are on the utility of optical genome mapping in prenatal setting which I have just discussed and also the application in brain tumors. Thank you so much for your attention.

BioNanogenomics

optical genome mapping

chromosomal studies

genetic disorders

DNA molecules

structural variations

copy number variations