Investigating the heterogeneity of biology is difficult. Basic methods, such as the averaging of data, can hide variability. Even fairly advanced methods, such as bulk sequencing, may do the same. As the “bulk” in bulk sequencing implies, it combines cells of varied types into a single sample for analysis, accomplishing another kind of averaging. Fortunately, investigations of biological heterogeneity are starting to take advantage of single-cell sequencing and spatial omics technologies. Indeed, biological heterogeneity has fewer places to hide now that these technologies can penetrate different “omes” such as the genome, the transcriptome, and the proteome. There are even technologies that encompass multiple omes.

Care must be taken, however, to ensure that biological heterogeneity won’t be missed simply because the technologies for probing it are overlooked or underused. Several such technologies are considered in this article. Most are suitable for in-house deployment. Some may be more conveniently accessed via service providers. In either case, they can help investigators overthrow the tyranny of averages that reigns over so many fields in biology.

Separating cells

Before working on single cells, scientists must collect them. Some techniques separate cells with high pressure, and others use tags to enable the identification and extraction of specific cells. Scientists at LevitasBio in Menlo Park, CA, have developed a platform that uses a lighter touch. The platform is called LeviCell. (This name, like the name of the company, takes “levi” from “levitation.”)

“[LeviCell] uses magnetic fields to enrich cells without directly labeling the cells or using any other direct manipulation of the cells,” says Kevin Travers, PhD, senior vice president of R&D at LevitasBio. “This is a very gentle process.”

A cell sample and a paramagnetic solution—basically a weakly magnetic concoction—are introduced to a LeviCell cartridge, directed down a separation channel, and subjected to a “magnetic density field.” In the separation channel, cells of different types levitate at different heights.

“It’s really an indirect effect,” Travers emphasizes. “The cells are different from the solution, and they get pushed up and away from the edges of the magnetic levitation chamber.” A real-time view lets a scientist observe the cells as they levitate within the LeviCell platform and then determine how the cells get separated into wells.

“The flow channels in our system are very large, which allows us to manipulate very large objects,” Travers explains. “So, the same technology can work with everything from bacterial cells up to things as large as organoids.” The system is designed to collect objects that are up to 350 μm in diameter. Moreover, LeviCell works with a wide range of cell numbers. Scientists at LevitasBio have successfully worked with as little as a few thousand and up to millions of cells.

From these samples, scientists can then “analyze the gene expression state of cells in as close to a native state as possible,” Travers asserts. “It ultimately allows researchers to increase their confidence that what they’re looking at is the cleanest data possible by removing the noise that comes along with dead cells and debris that are inherent to a lot of these cellular samples.”

Sequencing single cells

After collecting cells, a scientist might want to sequence them one by one. This can focus on single-cell DNA sequencing or RNA sequencing (scDNA-seq and scRNA-seq, respectively). That’s not easy to do, especially at high throughput. Instead of performing those processes in-house, a company might work with a contract research organization (CRO).

Single Cell Discoveries lab in their Utrecht facility
In July 2023, contract research organization Single Cell Discoveries moved from incubator space at the Hubrecht Institute to its own facility in Utrecht, the Netherlands. The company’s new sequencing laboratory includes a NovaSeq X Plus from Illumina, and a Chromium X from 10x Genomics (shown here). With its collection of platforms, Single Cell Discoveries can sequence RNA from single cells at a range of scales.

For example, Single Cell Discoveries—a CRO located in Utrecht, the Netherlands—operates a laboratory developed specifically for single-cell sequencing. “We are platform agnostic and will offer the highest quality and most requested single-cell technologies,” says Dylan Mooijman, PhD, head of R&D at Single Cell Discoveries. “Our technologies are divided between 384-well-plate-based scRNA-seq methods that we develop ourselves—SORT-seq and VASA-seq—and commercially available high-throughput scRNA-seq methods, such as those from 10x Genomics, Parse Biosciences, and Scale Biosciences.”

Using fluorescence-activated cell sorting (FACS), Single Cell Discoveries puts one cell in each well of a plate. “Afterward, the RNA from these single cells can either be directly reverse transcribed or manipulated to yield either 3¢ (SORT-seq) or full length (VASA-seq) single-cell RNA sequencing,” Mooijman explains. “High-throughput scRNA-seq methods are either droplet based or rely on fixation and combinatorial barcoding.” He adds, “VASA-seq allows for the detection of the full transcriptome,” including small nuclear RNAs (snRNAs) and small nucleolar RNAs (snoRNAs).

Single Cell Discoveries can sequence various quantities of RNA in as many as a million cells in just one run. “High-throughput methods generally have a slightly lower sensitivity but make up for it in sheer numbers,” Mooijman notes. “These high-throughput methods ensure coverage sufficient to sequence multiple guide RNAs targeting about 16,000 genes in a single-cell experiment.”

In many situations, scientists want to explore the immune system in basic and applied research. To help with this, Single Cell Discoveries is developing a method that utilizes 384-well plates to profile a sample’s immune cells. With this method, scientists will be able to sequence full-length T-cell and B-cell receptors from a small number of cells.

“Many immune cell–selection methods end up with only 100 to 300 cells, which is not sufficient for any current immune profiling method,” Mooijman points out. “By offering a method tailored to this need, we hope to further research on antibody discovery and antigen-specific T-cell receptor research.”

Adding imaging to omics

It’s one thing to analyze a single cell’s omics, and another to simultaneously determine that cell’s location. To do both, scientists must add some form of imaging.

As one example, Singular Genomics, based in San Diego, CA, recently announced its G4X Spatial Sequencing Platform. To offer this technology, the company combined its sequencing-by-synthesis (SBS) chemistry and high-speed imaging platform with novel methods to enable spatially resolved sequencing inside of cells and tissue. “We’ve figured out how to apply SBS sequencing to in situ read out of RNA transcripts and protein in a FFPE tissue section at high throughput with resolution of half a micron.” says Drew Spaventa, CEO of Singular Genomics. Because the G4X can collect more than two billion pixels per second while covering an imaging area of 40 cm2, it can process dozens of samples in a single day.

“By sequencing RNA in situ, researchers can analyze gene fusions, single nucleotide polymorphisms, and insertions/deletions in spatial context with other multiomic data,” Spaventa remarks. “This application will be particularly powerful for cancer and immunology research, providing a more comprehensive picture of the tumor microenvironment and cell interactions.” He also asserts that the unprecedented throughput of the system is uniquely suited for large-scale retrospective spatial characterization of tissue banks for new biomarker identification and therapeutic response stratification.

Locating protein biomarkers

In addition to technologies for determining DNA and RNA locations, there are technologies for determining protein locations. For example, a protein-localization technology called CellScape has been developed by Canopy Biosciences, a Bruker company in Saint Louis, MO. “CellScape is an imaging platform for precise spatial multiplexing,” says Thomas Campbell, PhD, Canopy’s associate director of product management.

“Historically, if people wanted to look at many protein biomarkers on a tissue sample, you had two options,” Campbell continues. “You could take multiple sections of your tissue and look at two or three markers per section, or you could grind that tissue up and then do a higher-plex assay, such as flow cytometry, but then you would lose spatial context—how different cells are interacting with each other, and how they’re spatially distributed within the sample.”

Canopy's Cell Scape Whole Slide Imaging Chamber
In April, Canopy Biosciences launched its CellScape Whole-Slide Imaging Chamber, which works with the company’s CellScape Precise Spatial Multiplexing platform. In the example shown here, scientists used CellScape to provide high-resolution imaging of a collection of targets in human tissue microarrays.

With CellScape, a scientist can analyze dozens and dozens of protein biomarkers in one tissue section, all without disrupting any spatial information. “And we can do that with single-cell resolution,” Campbell asserts. “So, you can phenotype every individual cell really deeply and start to understand all the different cell types and subtypes that are present within a sample and how they’re interacting with each other—all in a really quantitative fashion.”

To accomplish that, CellScape uses the company’s proprietary high dynamic range (HDR) image acquisition, which provides an 8-log detection range and is combined with a digital resolution of 182 nm/pixel. “We take the same image multiple times using different exposure times,” Campbell explains. “That allows us to quantify really bright signals using short exposure times and to quantify really dim signals using long exposure times. We can create one composite image that has a broader dynamic range and enables you to truly quantify both high-expressing and low-expressing cells within the same sample, and to really differentiate cells in situ across the entire expression range.” Plus, because the workflow is nondestructive to the sample, the same tissue section can be reexamined again the next week, month, or even year to study additional protein biomarkers.

From this combination of capabilites, says Campbell, “We have data that is more representative of the biology that’s happening, which will help fuel discoveries across a broad range of applications.”

Previous articleMultiomics Couples Data Generation and Clinical Science
Next articleCervix on a Chip Latest Advance for Women’s Health