The “Why choose spatial” series provides educational resources to guide you in making an informed decision about spatial sequencing technology, including success stories from scientists who are using spatial tools for their research. In Part 1, we share insights from our tissue and spatial biology experts regarding key experimental and technology considerations before choosing a solution.
It’s no secret that the next era of biological discovery will ride spatial waves. Blossoming out of the 2020 receipt of Method of the Year by Nature Methods, spatially resolved transcriptomics methods have aided crucial discoveries across the life sciences (1, 2, 3), establishing them as an essential tool to deeply characterize the cellular and molecular events that drive healthy development, disease pathogenesis, therapy response, and more—where they’re happening.
So, you're intrigued by the promise of spatial, but have some unanswered questions about the technology. Namely, how to ensure it’s an appropriate and accessible solution for your specific research questions and determine which solution in the growing spatial tools landscape is right for you?
In the following sections, we provide guidance for choosing a spatial solution, rooted in the questions you’re already asking as you plan your next experiment. It’s not intended to be comprehensive, but it’s solid advice from the tissue and spatial biology experts that call 10x Genomics home and have developed and supported our Visium and Xenium spatial solutions. We hope the experimental factors, common questions, and customer concerns described here can help inform your decision as you consider: why choose spatial?
Start with a question
It’s the same way you’d plan any experiment. Some aspect of biology fascinates you, and you want to understand it better, so you need to find a way to get access—to get deep enough into the underlying biology to answer your question in a satisfying and comprehensive way.
- How does pancreatic intratumoral heterogeneity affect therapeutic response? (1)
- What are the cellular and molecular drivers of inflammation in acute respiratory distress syndrome? (2)
- How does the architecture of the human endometrium change over the course of menstruation? (3)
There are a number of research use cases that are well suited for spatial transcriptomics technology. Melissa Leone, PhD, Spatial Market Development Manager at 10x Genomics, referenced the well-known Human Cell Atlas consortium to capture an important application of spatial technology: “Now we know what cells are in each organ, but where are they located?”
In addition to studies that elucidate where cells exist in an organ, Dr. Leone pointed to studies looking to define the relationships between cells in a complex tissue network. How does colocalization of certain cell types and the tissue architecture itself affect cellular function and biological outcomes? Take the tumor microenvironment, for example. What causes one cell to invade the tumor, while another remains excluded? “Based on these questions, actually, you’d need to see the differences between one of the cells that’s going to invade and the other that’s not,” which would require a view of the cells within their morphological context.
Does your research question fall into any of these use cases? Alternatively, your research question may be one in which a spatially resolved gene expression readout can be used as a complement to other data types. Making these determinations will help you start on the right footing.
Fill in the methodological gaps
Your interest in running a spatial experiment may arise out of other technology limitations or a need to access new analytes in spatial context for your specific research questions. It will be important to define what technical capabilities you’re looking for from a spatial technology in order to best answer those questions. For example:
- What kind of analyte—RNA or protein—do you need to access?
- What resolution do you need to visualize those analytes in the tissue context effectively?
- Do you need to access multiple analytes to best answer your question?
- Can you leverage a targeted gene panel or are you looking for whole transcriptome coverage?
- Is it easier to find what you’re looking for with a view of the whole tissue section or a confined region of interest?
- Is there a workflow to inform results from other methods, such as other forms of spatial tissue analysis or single cell gene expression, with this spatial data?
Continuing with the example of the tumor microenvironment, Dr. Leone recalled a recent conversation with a pathologist:
“I asked him, why are you going to do spatial, why are you going to get into this now? He said, ‘Well, because I have to. I need to see what’s going on. Bulk RNA-seq isn’t getting me where I want to go, so I’m going to do spatial.’ [...] He had these cells that, at the protein level, look exactly the same. But some are able to invade [the tumor] and some aren’t, so he said, ‘Now I want to look at the RNA level.’”
It may also be essential to run a spatial experiment to round out what you already know about a certain biological system from another experimental angle. Data integration methods can allow researchers to combine single cell and spatial transcriptomics data to build a more complete profile of that tissue. According to Dr. Leone, “This is really about using spatial data with single cell data to get at what’s actually going on biologically.”
It’s all about the tissue
When considering a spatial approach, it’s also essential to give particular attention to your tissue of interest. Which spatial solution fits the tissue samples you have access to? Compatibility will be determined by the species and organ from which the tissue is derived and how the tissue was treated after it was resected. Typically, the tissue will be either fresh frozen or formalin-fixed, paraffin-embedded (FFPE).
These are particularly important considerations since the chemistry of some spatial solutions will be compatible with any species, while others may only be compatible with a few. This also applies to the tissue treatment. How the tissue is prepared can have an impact on the analytes you may be interested in profiling, such as RNA, and the overall data quality. This review paper provides more details on how tissue treatment can affect RNA content.
FAQs about spatial technology
It can be helpful to hear how other scientists have found answers to their questions about spatial technology before taking that first step for yourself. In her many roles at 10x Genomics, Dr. Leone has supported researchers with the full gamut of concerns and “what ifs” related to our Visium Spatial assays. In the following section, we share a few of the most common questions and concerns scientists have as they’re evaluating spatial technologies, as well as expert guidance from Dr. Leone and the team of tissue specialists who support our customers’ spatial experiments.
Can I handle the workflow?
In other words, is the workflow accessible given my research experience and available infrastructure? Working with tissue is no small feat, and there are different aspects of a tissue-centric spatial sequencing workflow, such as histology staining and microscopy, that can be new for many scientists. Beyond upfront tissue and slide preparation, you’ll need to become comfortable with library preparation and sequencing as well. According to Dr. Leone, the key to successfully performing this kind of spatial experiment is collaboration. Every scientist has their strengths, and every scientist has things that they don’t know how to do yet. In light of that reality, success will look like building a collaborative team in which you can leverage each other’s strengths.
Dr. Leone explained,
“It’s trying to figure out, ‘Where am I going to get this stuff done?’ [...] [Some researchers] don’t know how to cut tissue, but they know how to do microscopy. Or they have expertise in cutting tissue, but they don’t have any idea about the molecular biology. So I often explain, ‘We’re in such a cool phase of science that we are now combining three very different modalities to create this amazing dataset.’ [...] We’re combining histology and imaging, which are their own fields, and now we’re adding in molecular biology too.”
The payoff of a cutting edge dataset makes the effort of creating connections with other scientists to establish a workflow all the more worth it from Dr. Leone’s perspective: “It’s pretty amazing that we can take advantage of all these modalities and the expertise of others to come up with this data. It really is worth doing all that work.”
Dr. Leone also warned against two potential responses to the challenges of a new workflow. “Some scientists might say, ‘I don’t know what this is, I’m not going to do it.’” Alternatively, some scientists fall into the overconfident but underprepared category: “So they’re like, ‘Ok—I don’t know about tissue,’ and they just go for it, and it’s a disaster.”
This is not meant to discourage trying, of course. Rather, with that fair warning in mind, it will be important for you to figure out what’s possible in your unique situation, with your lab, institution, or university, in order to move forward with spatial experiments in a smart way. Dr. Leone emphasized, “By 2030, this is probably going to be mainstream, but, today, it’s pretty advanced. You’re getting in on an advanced technique, but think of everything going into producing this data. It’s worth doing. It’s worth putting some effort into, and I’ll be there to help you.”
Will I be able to analyze the data?
This is an incredibly important question, and will require pre-work to ensure you have the proper data analysis tools and workflow in place before starting a spatial experiment. Finding the story in your spatial sequencing data could require collaboration with experienced bioinformaticians. But that’s not your only option. There are a number of third-party tools, as well as data analysis platforms freely available alongside commercial spatial solutions, that you can use to perform analysis. In fact, the availability of data analysis tools and support could be an essential front by which to evaluate a spatial technology.
Dr. Leone recommends doing your homework:
“When we’re talking about starting spatial projects, I show everybody our datasets. We look at Space Ranger data together so they understand what they’re getting. Now let’s imagine—what’s your cell type, what’s your tissue type? I give homework because I want people to play around with sample data so they're ready to go when it's time to analyze their real data."
Take the time to explore some spatial gene expression datasets. How is the data visualized? Are the visualization tools intuitive, easy to use and navigate? Is there a clear channel for technical training and support? You should become convinced that you’ll get the help that you need before making a decision on a platform.
Our analysis guides, a collection of introductions, tutorials, and blogs for data analysis, may be helpful as you continue to explore options for analyzing your data. This review paper also provides an overview of what’s possible with spatial data analysis.
Overcoming research barriers with spatial
One of the most important considerations you’ll need to make before embarking into the world of spatial is whether or not using this technology will significantly impact the research barriers you’re currently facing. How would a readout of gene expression in the tissue context help to answer your questions? Understanding this will likely come as you see how other scientists have leveraged spatial gene expression technology to make new discoveries and advance their research projects.
Alternatively, how would missing the spatial context impact the progression of your research? Dr. Leone reframed this same consideration: “If we don’t look at where your T cell is located in relation to your cancer cell, we’re not going to know what’s happening in that interaction. If we see it’s really far away—or we see all this stuff on the single cell map, but if we can’t see where the cells are relative to each other, we’re missing a whole bunch of information.”
This impact may not be confined to a single experiment either, but to your broader research program and the progression of the questions you want to ask. Dr. Leone continued, “I think we all just know that if you understand where things are next to each other, you’re going to design much better follow-up questions. Because doing a spatial experiment isn’t the end. That’s just the beginning. And then how does that inform the next questions you ask?”
Stay tuned for Part 2 of our “Why choose spatial” series as we hear directly from researchers who are using spatial gene expression technology for their experiments. We’ll learn why spatial was a good fit for their research questions, what experimental hurdles it has helped them overcome, and how they’ve used it to make new discoveries in their field of biology.
And find more information about Visium Spatial assays from 10x Genomics here →
- Berglund E, et al. Spatial maps of prostate cancer transcriptomes reveal an unexplored landscape of heterogeneity. Nat Commun 9: 2419 (2018). doi: 10.1038/s41467-018-04724-5
- Boyd D, et al. Exuberant fibroblast activity compromises lung function via ADAMTS4. Nature 587: 466–471 (2020). doi: 10.1038/s41586-020-2877-5
- Garcia-Alonso L, et al. Mapping the temporal and spatial dynamics of the human endometrium in vivo and in vitro. Nat Genetics 53: 1698–1711 (2021). doi: 10.1038/s41588-021-00972-2