Selecting a Gene Expression Profiling Platform

View Podcast Transcript

Allison Wheeler: Hi, everyone. Welcome to Biocompare’s Tech Insights podcast where we speak to scientific experts about new tools and technologies that can help advance your research. I'm Allison Wheeler, Associate Editor here at Biocompare, and I want to welcome today's guests Erin Piazza, Senior Bioinformatics Scientists at NanoString, and Steven Ross, Vice President of IT NanoString. Thank you both so much. for taking the time to talk with us today.

Erin, before I jump into some questions for you, would you like to explain a little more about yourself and your background before we get started?

Erin: Absolutely, Allison, and thanks for the introduction. As you said, I'm a Senior Bioinformatics Scientist at NanoString. I've been at the company about seven years where my primary focus has been on the content behind our panels and making sure that, regardless of the platform we're developing, that we have the best genes in every single product we make. It's one of my great passions in life.

My background is in cancer biology design. I got my Ph.D. from Stanford and have been kind of focused on cell and molecular biology for my whole career. So looking forward to talking with you today.

Allison Wheeler: Awesome. Thank you so much, Erin. We're so excited to have you here with us today.

Steven, before we keep going, would you also like to introduce yourself and explain a little more about your background?

Steven: Yeah, hello, everybody. My name is Steven Ross. I'm the Vice President of IT for NanoString. I've been a CIO in the healthcare space for the last 10 years, and I've been at NanoString now for about a year and was brought on to help build the NanoString IT platform to the next generation. So I'm also looking forward to talking with you all today.

Allison Wheeler: Awesome. Thank you, Steven. We really appreciate having you both here today.

Now, Erin, if you don't mind, we can go ahead and jump right into our questions for today's podcast, the first one being what is gene expression data use for, what information do you glean from it, and how does that help inform research?

Erin: Absolutely. Great question. I think about this a lot because I actually worked primarily with protein actually during my Ph.D., but I've come to love gene expression as an assay since coming to work at NanoString.

I think we can arguably say that protein is the functional unit of the cell, but RNA is a really crucial readout of all of the various mechanisms that control the expression of the genome, and it thus provides a really important view of the state of a cell at any given time. It's used a lot because, honestly, it's far easier to measure than protein, and particularly easier to measure is a multiplex assay. And this makes it a lot more useful as a global view of what's going on in the cell as well as a starting point for developing signatures that you can use for diagnostic or other ways of measuring biology.

Allison Wheeler: Awesome, thank you. Okay, so what are some of the key features of the nCounter gene expression profiling platform, and how is it different from other technologies out there?

Erin: Yeah, great question. So nCounter is just a highly robust, targeted gene expression platform that you can use to measure up to about 855 genes at a time. While we use it primarily for research purposes, we also have two approved diagnostic assays that run on the nCounter platform specifically because of that highly robust readout. And there's a few more coming up the line that are not yet FDA approved but are moving in that direction. It's one of the things we're quite proud of, that the platform is robust enough to support diagnostic assays.

The chemistry itself is all hybridization based and relies on a hybridization of probes directly to RNA targets, which has a few benefits, one of which is that each count you see on an nCounter assay is based on the detection of a single molecule, which means everything from the readout is digital. You actually get integers and not something with a bunch of zeros after it.

And then second is that there's no need for enzymes or CDA conversion like you would do with a sequencing assay, which means that the assay is just a lot more robust sort of buffer additives and other things that can happen that would impact an enzyme’s function. There’s simply just no need for enzymes in the assay whatsoever.

It's also a very quick turnaround time assay. So you can go from a sample to an answer in just about 24 hours, which is very hard to do with most other assays. And finally, the bioinformatics that designed the probes that work on the nCounter assay only target about 100 base pairs of sequence at a time, which means that when you have really degraded samples like the super clinically relevant FFP, you still get really great performance because you're simply not looking for that much sequence to specifically identify a target, and it tends to work much more robustly than other assays, particularly sequencing where you have to identify or at least be able to manipulate more of a sequence than that.

Allison Wheeler: Awesome. Thanks for detailing all that out for us, Erin. So how does the NanoString platform make statistical analysis and bioinformatics easier?

Erin: Yeah, one of those things that I get asked a lot as a bioinformatician for sure. We tried to develop a platform that puts people like me out of a job for the most part. We try to put the bioinformatics upfront.

So because we digitally read out each of these targets, the roughly 800 answers or so you get from a single sample can be easily read and understood on a normal laptop and without a lot of interpretation. We offer some free software products we call nSolver and nSolver Advanced Analysis that can perform all recommended analysis steps that we tend to have with our assays like normalization to certain reference genes that can control for different amounts of sample one might have loaded. And then our panels and custom products are all compatible with our advanced analysis suite, which I mentioned before, and that offers a lot more in-depth tools like pathway analysis, cell type profiling and differential expression.

We’ll get to panels next, but those are actually designed with analysis in mind. And so we -- it's one of our things that we try to make sure analysis and design and product kind of go hand in hand.

Allison Wheeler: Perfect. Thank you, Erin.

So NanoString offers many curated panels across various applications. So what goes into designing a gene expression panel at NanoString?

Erin: Awesome. Well, you hit on pretty much what I do for a living, so slightly longer answer. But our content design comes from our bioinformatics group because we believe that the best decisions are made by people who both understand the biology and can make data driven choices that lead to better data for each experiment.

Good content design really starts with a deep understanding of biology, which we get from papers, talking to experts and going to conferences. Once we get that sort of baseline in the biology, we develop what we call a framework, which is a sort of conceptual structure for the biology of a subject area. Those frameworks have both themes, which can be something like the Hallmarks of Cancer by Hanahan and Weinberg or something we develop ourselves. And then they also contain biological pathways, and those biological pathways drive the genes that we consider in the data driven fashion for their informative value.

We'll use a really large whole transcriptome dataset, something like Cancer Genome Atlas, although cancer’s very lucky to have a dataset that nice; not all of them do. But certainly, we try to get as good of a dataset as we can. And then we use certain methods that have been developed internally by our biostatistics team to look at each gene of about 4 to 6,000 per panel, and we determine which of those have just the most informational value compared to the rest.

We also would add genes that have really clear literature relevance based on both literature review and publication during metrics. And then finally, we always talk with experts in the field to make sure that we have covered everything that's important and also things that are leading edge that might not have been caught by some of our other development work.

And then once that's all done, we wrap it up in a way that's on full display in our analysis methods where we use pathways to analyze the data. And we also have a methodology called cell type profiling that you can use to look at the relative abundance of cell types within the sample, typically immune cell types, but we have a cardiovascular disease panel that now looks at cell types within the heart, as well. So we try to tailor all of those analysis methods, depending on the application we're building for.

Allison Wheeler: Excellent. Well, thank you so much for those insightful answers, Erin.

Steven, I have one question I'd like to direct your way if you don't mind. So NanoString recently launched an upgraded nCounter analysis system called the nCounter Pro, which offers advanced cybersecurity features. Can you elaborate on what those features are and why cybersecurity is so important in the modern research environment?

Steven: Yeah, you bet. Thanks for the question. And as you said, the new nCounter Pro does include cybersecurity features that are necessary to protect the integrity of your data and build a holistic data security environment. And the industry best practices enhancements that we implemented begin with the new Windows 10 IoT operating system, fully patchable. There is a hard drive encryption using AES 256, and there’s data encryption using SHA 256.

What this means is that your data is protected at rest and in motion, which is important. There's also a whitelisting application in there that prevents unauthorized, executable access, and the system is also fully CFR 21 Part 11 compliant, too, so you can retrieve audit reports and tracking and reporting these to help validate the system.

These kind of enhancements are really important today's cybersecurity world in healthcare with sensitive patient information and just the overall threat landscape that’s out there. So we felt it was very important to do this from a strategic perspective.

Allison Wheeler: Awesome. Thank you so much, Steven and Erin, for taking the time to share your expertise with us today.

For more information on gene expression analysis, check out Biocompare’s recent Tech Insights article, Gene Expression Data You Can Count On.

For more information on products, technologies and the latest scientific advancements, Be sure to check out biocompare.com.

Thanks for tuning in, everyone, and have a great day.