Experimental Protocols

Understanding scRNA-seq protocols helps you interpret published datasets and choose the right platform for your experiment. Each protocol makes different tradeoffs between throughput, cost, and transcript coverage.

Droplet-Based: 10x Genomics Chromium

The 10x Genomics Chromium is the most widely used scRNA-seq platform. It uses microfluidics to encapsulate single cells inside tiny droplets.

How It Works

1. Cell suspension is loaded onto a Chromium chip alongside gel beads and oil.
2. The chip creates thousands of nanoliter droplets called GEMs (Gel Bead-in-Emulsion). Each GEM contains one cell and one gel bead.
3. Each gel bead carries millions of copies of a barcoded oligonucleotide. This oligo contains three parts: a cell barcode that identifies which cell, a UMI (Unique Molecular Identifier) that identifies which molecule, and a poly-dT tail that captures mRNA.
4. Inside the GEM, the cell lyses. The poly-dT tail on the bead captures mRNA molecules. Reverse transcription creates barcoded cDNA.
5. The emulsion is broken. All barcoded cDNA from all GEMs is pooled into a single tube. Standard library preparation and sequencing follow.

After sequencing, the cell barcodes tell you which reads came from which cell. The UMIs let you count the original number of mRNA molecules, removing PCR amplification bias.

This video from 10x Genomics shows the full Chromium workflow from cell loading to sequencing-ready libraries:

Key Characteristics

• Throughput: 500 to 10,000 cells per run. The newer Chromium X can capture up to 20,000 cells.
• Transcript coverage: Captures the 3’ end of transcripts only. The 5’ chemistry is available for immune profiling where you need TCR and BCR sequences.
• Cost: Roughly $1 to $2 per cell, including library prep and sequencing.
• Instrument: Requires a Chromium Controller or Chromium X.

Note

The 3’ capture means you do not get full-length transcript information. You cannot study alternative splicing or isoform usage with standard 10x data.

When to Use It

10x Chromium is the default choice for most scRNA-seq experiments. It provides good throughput at reasonable cost. If you are profiling a tissue and want to capture the major cell types, this is likely the right platform.

Combinatorial Indexing: Parse Biosciences

Parse Biosciences uses a fundamentally different approach called split-pool barcoding. No microfluidics instrument is needed.

How It Works

1. Fixation: Cells are fixed, which preserves RNA inside the cell and allows cells to survive the barcoding process.
2. Round 1: Cells are randomly distributed into a plate of wells. Each well adds a unique barcode through in-cell reverse transcription. Cells are then pooled back together.
3. Round 2: Cells are randomly redistributed into new wells. Each well adds a second unique barcode. Cells are pooled again.
4. Round 3: A third round of splitting, barcoding, and pooling adds a third barcode.
5. After three rounds, each cell has a unique combination of three barcodes. The probability that two cells received the same three barcodes is extremely low.

The barcodes are added while the molecules are still inside the cell. This means you do not need to isolate single cells into individual compartments. The combinatorial math does the work.

This video from Parse Biosciences walks through the Evercode split-pool barcoding workflow:

Key Characteristics

• Throughput: Up to 1 million cells per experiment. This is 10 to 100 times more than droplet methods.
• Transcript coverage: 3’ end capture, similar to 10x.
• Cost: Lower per-cell cost at high cell numbers. The kit cost is higher, but splitting across many cells brings the per-cell cost down.
• Instrument: No dedicated instrument required. Uses standard lab equipment like pipettes and centrifuges.

When to Use It

Parse is a strong choice when you need very high cell numbers. Large-scale cell atlases, perturbation screens, or experiments with many samples benefit from the high throughput. It is also useful for labs that do not want to purchase a Chromium instrument.

Plate-Based: Smart-seq2 and Smart-seq3

Smart-seq protocols take a different approach. Instead of capturing thousands of cells, they focus on getting the most information from each individual cell.

How It Works

1. Single cells are sorted into individual wells of a 96-well or 384-well plate, usually by FACS.
2. In each well, the cell is lysed and mRNA is reverse transcribed into cDNA.
3. A template-switching oligo captures the 5’ end of the transcript. This enables full-length cDNA synthesis.
4. The cDNA is amplified by PCR.
5. Each well is processed as an individual library. After sequencing, reads cover the entire transcript from 5’ to 3’.

Smart-seq3 is an updated version that adds UMI counting to the full-length protocol, combining the benefits of both approaches.

Key Characteristics

• Throughput: Hundreds of cells per experiment. Limited by the plate format and FACS sorting speed.
• Transcript coverage: Full-length transcripts. This is the main advantage over droplet methods.
• Cost: $5 to $10 per cell or more. The high per-cell cost comes from individual library preparation for each well.
• Instrument: Requires a cell sorter for single-cell isolation and standard molecular biology equipment.

When to Use It

Choose Smart-seq when transcript coverage matters more than cell numbers. Isoform analysis, splice variant detection, and SNP detection along transcripts all benefit from full-length data. It is also commonly used for rare, pre-sorted cell populations where you have already identified the cells of interest by FACS.

Tip

Smart-seq data works well with standard bulk RNA-seq tools because you get full-length coverage. Many researchers analyze Smart-seq data with bulk tools like DESeq2 after accounting for the single-cell nature of the data.

Platform Comparison

Feature	10x Chromium	Parse Biosciences	Smart-seq2/3
Throughput	500 to 20,000 cells	Up to 1,000,000 cells	Hundreds of cells
Transcript coverage	3’ end only	3’ end only	Full-length
Cost per cell	$1 to $2	< $1 at scale	$5 to $10+
Instrument required	Chromium Controller	None (standard lab)	Cell sorter (FACS)
UMI counting	Yes	Yes	Smart-seq3 only
Best for	General profiling	Large-scale atlases	Isoform analysis
Cell input	Live suspension	Fixed cells	FACS-sorted cells
Sensitivity	Moderate	Moderate	High, more genes per cell

Choosing a Protocol

The right protocol depends on your biological question.

“I want to profile cell types in my tissue.” Use 10x Chromium. It provides the best balance of throughput and cost for standard experiments.

“I need to profile hundreds of thousands of cells across many conditions.” Use Parse Biosciences. The split-pool approach scales to very high cell numbers without proportionally increasing cost.

“I need full-length transcript information for splice variant analysis.” Use Smart-seq2 or Smart-seq3. Droplet methods only capture one end of the transcript.

“I want to study a rare cell population that I can sort by FACS.” Use Smart-seq. If you already know which cells you want and can sort them, Smart-seq gives you the deepest per-cell data.

Sample Preparation Matters

Regardless of which protocol you choose, sample preparation is often the most critical step. All scRNA-seq platforms require a high-quality single-cell suspension.

Dissociation: Tissues must be dissociated into single cells. Enzymatic digestion with collagenase, trypsin, or tissue-specific cocktails is standard. Over-digestion damages cells. Under-digestion leaves clumps that clog microfluidics or create doublets.

Viability: Dead cells release ambient RNA that contaminates the data. Most protocols require greater than 80% viability. Filter dead cells with FACS or MACS before loading.

Speed: Cells change their gene expression in response to dissociation. Work quickly and keep cells cold to minimize transcriptional artifacts. Some tissues benefit from fixation before dissociation.

Caution

Poor sample preparation is the most common cause of failed scRNA-seq experiments. No amount of computational analysis can rescue data from dead or damaged cells. Invest time in optimizing your dissociation protocol before running an expensive scRNA-seq experiment.

What Comes Next

The next page walks through the standard computational analysis workflow. You will learn what happens after sequencing: from raw FASTQ files to cell type annotations.