Optimizing Sample and Library Prep for Single-Cell RNA-Seq

RNA sequencing (RNA-seq) is a common method of analyzing the expression of genes by groups of cells, such as distinct tissues, or cultured cells. Such bulk RNA-seq methods measure the averaged expression of the group, losing the different gene expression levels of individual cells. Researchers use single-cell RNA sequencing (scRNA-seq) to detect individual expression profiles from isolated cells.

Despite the numerous scRNA-seq methods available today, sample and library preparation remain fairly similar—as is the importance of optimization for better scRNA-seq results. “By optimizing sample preparation, you can have greater confidence that the observed changes in the data are due to the sample itself rather than poor sample quality caused by inadequate preparation,” explains Zuleyma Peralta, Product Manager in Sample Preparation at 10x Genomics. In this article, we provide some expert tips on optimizing the scRNA-seq process by streamlining sample and library preparation.

Sample preparation

Start with healthy cells, and whatever cell isolation process you use, treat the cells as gently as you can, minimizing the duration of the cell isolation procedure. “Be aware that different isolation methods may introduce different levels of bias or variability in the data,” notes Peralta. “Each tissue or cell type is unique, so you may have to optimize your existing sample preparation workflows when preparing new sample types.”

At the start of sample preparation, do everything you can to stop RNA degradation—use cold temperatures and RNase inhibitors as soon as possible, and a streamlined, pre-planned protocol to reduce overall work time. “Cells outside their medium of choice can be very fragile, so lysis without inhibitors can lead to quick degradation of RNA,” says Kit Krishnan, NGS Development Group Leader at NEB. “Quick freezing once cells are isolated, in the presence of RNase inhibitor, will help mitigate this.”

If a delay is necessary after cell isolation (e.g., if isolating multiple cells before progressing to the next stage of the scRNA-seq workflow), freeze or fix cells prior to library preparation to ensure a healthier starting sample. “In single cell workflows, the quality of the starting sample is reflected in the data,” notes Peralta, so efforts to maintain sample quality throughout the workflow are well worth the trouble. “To prevent sample degradation, it’s advisable to stabilize the sample shortly after collection through cryopreservation or fixation.”

While it may sound obvious, it’s a good idea to choose options for your scRNA-seq workflow that support your research aims. “Efficient enrichment of RNA, without bias and degradation, is the most important step, especially when low abundance transcripts are also of relevance,” says Yanxia Bei, Principal Developmental Scientist at NEB. If your research aim includes counting transcripts, then adding barcodes or unique molecular identifiers (UMIs) to transcripts is an essential tool. Also, using UMIs “can help to distinguish true biological variation from technical noise,” adds Peralta.

Choosing your reverse transcriptase (RT) enzyme is another opportunity to optimize your scRNA-seq workflow, and it may depend partly upon your research interests. Some enzymes are better at transcribing full-length cDNAs, while others are more likely to result in partial transcripts. “For full-length cDNA you need a highly efficient RT, which can go through secondary structures and all the way to the 5’ end to capture the entire molecule,” says Krishnan. Full-length cDNAs would be appropriate for detecting RNA variants or alternative splicing, for example.

Analyzing the steps in your workflow

Make sure that the methods used for sample collection, sample storage, and cell isolation are all compatible with each other, and with further downstream analysis. Rehearsing your protocol steps ahead of time, in a dry run without cells, can assist with learning a new workflow so that you can focus more attention on cell health. “A key factor while working with single cells is speed, being fully prepared for all steps from the start to end of a protocol so degradation of material is minimal,” Bei says. “Quick turnaround times after incubation steps are crucial to enable optimal library yield and quality.”

Sample preparation kits can streamline the workflow, reducing hands-on time by eliminating steps such as making buffers. In addition, vendors optimize their kits’ reagents and workflow steps to be as efficient and user-friendly as possible, and usually include extensive instructions. “We optimized every step in the workflow to take place in one tube, sequentially with no cleanups in between, to avoid loss of material,” says Krishnan. Given the range of possible sample types, there may be some optimization required when using a kit. “However, a kit may have known points for optimization, as opposed to a customer-developed protocol,” notes Peralta. “At 10x Genomics, we provide guidelines for optimization and troubleshooting sample preparation in our kit User Guides.”

Automation

An effective method of streamlining your workflow is the use of automation. Whether in the form of an all-in-one workstation, or simply the addition of automatic pipettors, any level of automation can reduce human error and user bias, and may help to increase throughput. Single cell sequencing can be adapted to an automated workflow, depending on whether the sample is sufficiently scalable, says Peralta. For example, cell lines or cells grown in suspension are easier to scale up than cells derived from tissue samples, which require more preparation time during which cell health can decline. However, using fixed samples can prevent this decline. “Fixing samples for use with the Chromium Single Cell Gene Expression Flex assay enables higher throughput assays, as fixation locks the sample’s biological state, easing concerns of sample decay,” Peralta says.

Besides reducing human errors, automation can help to increase the parallelization of the workflow, shorten steps, and enable the use of high-throughput cell counters and multichannel pipettors. “Depending on the scale of experiments, automation solutions will be highly beneficial, and will mitigate user and experiment-to-experiment variability,” says Krishnan. All such endeavors will help to reduce noise and streamline sample preparation steps for scRNA-seq.