One of the first things you’re taught in biology is the basic components of a cell. You learn how cells function and that they can specialize or differentiate to serve specific purposes. You have heart cells, kidney cells, bone cells, etc., and since they serve the same purposes they’re pretty much identical, right? Challenging this is the emerging idea that all cells are different, even within the same populations (i.e., not all of your heart cells are identical). Every organism, organ system, and tissue is actually a mosaic of cells with individual characteristics. How do we measure these individual cellular properties? This installment of “What’s it all about?” aims to introduce the world of single-cell sequencing, Nature Publishing Group’s method of the year in 2013.
What is single-cell sequencing?
Figure 1. Single-cell sequencing workflow. Credit: Amanda Haage
The broad definition of single-cell sequencing is pretty self-explanatory: you take a single cell and sequence its genetic information. The real meat of this field comprises techniques that allow for sequencing with minimal starting material, preservation of the context the cell existed in, and analysis that ties this high-resolution data to relevant questions. Single-cell sequencing experiments have a general workflow as summarized in Figure 1. The sorting or isolation step is often performed in conjunction with several other cutting-edge techniques such as microfluidics or laser-aided microdissection, but can also consist of simpler techniques such as flow cytometry or chemical/physical dissociation. Flow cytometry or fluorescence-activated cell sorting (FACS) is used if you want to isolate a rare subpopulation of cells based on various biomarkers. Once a single cell or rare subpopulation of cells is isolated by a specific technique, the material to be sequenced is extracted. The material extracted from the single cell (or small group of cells) is most commonly DNA or RNA. In addition to DNA and RNA, techniques for single-cell proteomics, DNA methylation, and the parallel measurement of multiple materials have been reported in the past three years. The applications for each type of material extracted can differ, as discussed below.
Next, the small amount of material extracted must be amplified. In the case of single-cell DNA sequencing this is referred to as whole-genome amplification (WGA). There are many methods for WGA and for amplifying RNA (Guin & Oudenaarden provide a good comparison here). The main concern with the method used for amplification is uneven coverage across the material. Structural and molecular variation lends parts of the DNA or RNA for better or worse to amplification, which can introduce possible bias into your sample.
Next-generation sequencing techniques are really what have made single-cell sequencing possible. The term “next-generation” refers to any technique that improves upon traditional Sanger sequencing. These techniques are unified by improved speed and accuracy, the reduction of manual input, and the need for library preparation. Libraries are made from the amplified material by fragmenting it and tagging/ligating the pieces with different adapters/markers. Library prep varies by the method of next-generation sequencing used, but in general, whatever adapters or tags are added to the extracted material allow for simultaneous sequencing of the many fragments. This produces huge datasets of sequenced information in relatively short amounts of time with researchers just having to send samples off or push a button. It’s increasingly clear that the real work of future single-cell sequencing experiments will be in thoughtful design and data analysis, instead of lengthy manual sample prep. Just like with the choice of isolation, amplification, and sequencing technique, a huge body of literature is cropping up about the variations in single-cell sequencing data analysis pipelines (see below in “Using single-cell sequencing in your own research”).
What are the applications?
The applications for single-cell sequencing are nearly endless. It has essentially created a sub-subfield within all subfields of cell biology asking how individual cells within a population differ from each other. It has vastly increased the depth and number of “omic” topics in biology. Extracting whole genomic DNA allows you to examine single-nucleotide polymorphisms or acquired mutations in single cells or small cell populations, like from a tumor biopsy. Extracting DNA based on methylation or chromatin status gives you an overall picture of the epigenome. Extracting and sequencing RNA allows you to examine the cellular transcriptome, including expression profiling and splice variant analysis. Examining all of these sequences together provides a glimpse into how cells integrate multiple levels of information. Comparing this information across different microenvironmental contexts or at different time points allows scientists to ask “How are these cells heterogeneous?” as well as “How does that heterogeneity change?” The seemingly ultimate application of single-cell sequencing is the quantification of all cellular outputs based on various inputs, such as time and environmental cues.
Figure 2 created by Amanda Haage with Pubmed data. Green data from https://www.ncbi.nlm.nih.gov/pubmed/?term=single+cell+sequencing, blue data from https://www.ncbi.nlm.nih.gov/pubmed/?term=single+cell+RNA+sequencing.
What’s happening now and where is this field going?
Though single-cell sequencing approaches now may seem pervasive in cell biology (Figure 2), the field as a whole is just getting started. Consider the mission of the Human Cell Atlas: “To create comprehensive reference maps of all human cells—the fundamental units of life— as a basis for both understanding human health and diagnosing, monitoring, and treating disease.” This ambitious project is grounded in single-cell sequencing of a multitude of materials. As we expand the type of outputs we can measure at the resolution of a single cell, the applications to personalized medicine are an obvious next step.
As single-cell sequencing has revealed, each tissue represents a mosaic of cells designed for a specific purpose, but with undeniable individuality. This heterogeneity has a profound impact on treatment outcomes for different diseases, especially since tumor heterogeneity can contribute to cancer therapy resistance. A future can be envisioned where high-throughput, single-cell sequencing in combination with patient-derived organoids provides clinicians with detailed information describing their patients’ complete “omics”. Doctors could then tailor treatment plans to each person’s known risk factors or drug susceptibilities.
Using single-cell sequencing in your own research
The ability to implement single-cell sequencing experiments largely depends on your access to the necessary technology. These techniques have spurred their own subfield of biotechnology companies offering more and more automated solutions. If you’re just starting to consider single-cell sequencing check out 10X University from 10X Genomics or explore learning opportunities from illumina. In addition, several “practical guides” have been published. See Haque et al. and Lafzi et al. for recent general guides for single-cell RNA sequencing guides. Studies like Ding et al. that compare and contrast the quickly expanding available technologies are also necessary for beginning researchers to get their feet wet in the single-cell sequencing pool.
Once you have next-generation sequencing data, the analysis may seem daunting. Luckily there are many free resources for teaching yourself, including this complete course hosted on github. Several resources are specific for single-cell RNA sequencing, including best practice guides, comparisons of software analysis packages, and workflows in R. Also consider incorporating a trained bioinformatician into your team. Cell biology and academic science are increasingly collaborative, and having specialized team members will only make your studies stronger. Many of the biotech companies mentioned above also provide bioinformatic services for a cost.
About the Author:
Amanda Haage is a newly minted assistant professor at the University of North Dakota. She previously trained as a postdoctoral fellow in Guy Tanentzapf’s Lab at the University of British Columbia and received her PhD in 2014 from Iowa State University in Ian Schneider’s Lab. She is generally interested in how the microenvironment regulates cellular behavior as well as promoting diversity and inclusion in science. Twitter: @mandy_ridd and Email: amanda.haage@und.edu
