NEW YORK – Single-cell technology companies are using this week's virtual meeting of the American Society of Human Genetics to showcase new products that greatly increase the scale of experiments.
On Wednesday, Parse Biosciences, a University of Washington spinout commercializing split-pool ligation-based transcriptome sequencing technology, announced a new kit that can prepare transcriptomes from up to 1 million cells at once. The Evercode Mega kit can process up to 96 samples at once, using combinatorial indexing technology. Like its previous products, the kit works with fixed samples that can be stored for up to six months before processing.
And in a talk sponsored by 10x Genomics, Sisi Chen, director of the Beckman Institute Pilot Center for Single-cell Profiling and Cell Engineering at Caltech, provided an analysis of an 80,000-cell experiment run with new high-throughput assays on the Chromium X, an instrument released this summer that can likewise process 1 million cells at a time. In another talk, Sarah Parsons, 10x's lead field applications scientist, presented a publicly available data set of about 24,000 cells, also obtained from a high-throughput assay on the Chromium X.
The increased scale will help lower per-cell prep costs, the firms say. 10x claims per-cell costs on the Chromium X can be as low as $.02, while Parse estimated per-cell costs at less than $.02, compared to $.09 per cell with its 100,000-cell kit.
"It changes the types of experiment we can do," said Rahul Satija, a single-cell genomics expert at the New York Genome Center and New York University, who has used both 10x and Parse products. "Even if we don't need a million cells for one experiment, it's likely we'll need a million in aggregate."
As single-cell genomics continues to grow, increasing scale and lowering cost are becoming ever more important. "Cost per-cell is an essential metric," Satija said. "Even a year ago we could profile a million cells by having sufficient resources. The limitation wasn't that you couldn't do it, but that it was cost-prohibitive."
Parse cofounder and CEO Alex Rosenberg said that the company hadn't told customers the Evercode Mega kit was coming out until this week, but the customers had told the company they wanted its capabilities.
"Customers have been telling us, 'We have plans to do these larger products,'" he said. "It's a consistent conversation we've been having with them."
Some researchers have even turned to "hacking" existing products to boost the number of cells possible per run. In June, researchers from the Austrian Academy of Sciences published a method that used combinatorial indexing and droplet overloading of the 10x Chromium to obtain 1 million single-cell transcriptomes in a run.
In an email, a 10x spokesperson said the company has performed Chromium X runs with more than 500,000 cells per run. The instrument and high-throughput assays have been shipping since August and customers are in the process of generating data with them.
Larger experiments mean new types of experiments. A million cells could allow detection of very rare cell types, Rosenberg said. Moreover, researchers "are realizing the need to do a lot more replicates. Scaling up samples and cells can produce more robust results," he said.
Satija stressed that what's useful is not just running more cells per sample. "It's not only that we want to have 1 million cells from one sample; we may also want to profile samples from 100 different donors, or from many time points."
For the Parse kit in particular, the ability to work on fixed cells could provide an edge. "When working with dozens of samples, they may not all be ready at exactly the same time," Satija said. With the Parse kit, his lab could wait until it has a sufficient number to justify using the larger kit. "That gives us a lot of flexibility," he said.
Satija noted that he was "very impressed" with his initial run of Parse's 100,000-cell kit, "both in terms of the number of cells we were able to profile and in data quality per cell that we received." He was also satisfied with Parse's low doublet rate. The new Evercode Mega kit offers a doublet rate of 3.2 percent for 1 million cells. Satija is also using the Chromium X and high-throughput assays.
But bigger isn't the only trend in single-cell studies. Parse is also going to offer a "mini" kit that preps about 10,000 to 20,000 cells. The intended customer here is someone new to single-cell genomics, who may not want to commit to a larger experiment, or experienced researchers who just want to pilot an experiment before moving forward.