This article was originally published Dec. 30.
Fluidigm has released a sample preparation workflow for its C1 Single-Cell Auto Prep system that processes single cells and amplifies their DNA for use in targeted, whole-exome, and whole-genome sequencing.
The new C1 DNA sequencing product consists of chips for the C1 system, a C1 reagent kit, and validated scripts, and uses GE Healthcare Life Sciences' Illustra GenomiPhi V2 DNA amplification kit. It is currently available to early-access customers for targeted resequencing, and Fluidigm plans to release whole-exome and whole-genome sequencing applications in early 2014.
Using the new sample prep product, researchers can analyze somatic mutations, including SNPs, small insertions or deletions, and translocations, in single cells from heterogeneous cell populations.
Fluidigm first released the C1 system about a year ago, initially for RNA and later microRNA preparation from single cells, and has placed about 100 instruments in the market so far. Customers include the Broad Institute, the Wellcome Trust Sanger Institute, and the Genome Institute of Singapore.
Candia Brown, Fluidigm's director of single-cell genomics, said she expects a "substantial proportion" of the installed base to start using the new sample prep workflow for DNA sequencing.
The C1 allows 96 cells to be processed in parallel, automating cell capture, washing, staining and imaging, cell lysis, and amplification of RNA or DNA.
For targeted sequencing, following whole-genome amplification, up to 480 amplicons can be further amplified using Fluidigm's Access Array system and its D3 Assay Design service. The company is currently evaluating several exome amplification kits for the pending exome sequencing application.
The new C1 DNA sequencing workflow requires no hardware changes to the C1, just new protocols and scripts, according to Brown.
Fluidigm sells all required reagents except the GenomiPhi amplification kit, which uses Phi29 DNA polymerase for isothermal whole-genome multiple displacement amplification.
Brown said Fluidigm has improved the uniformity of amplification across the genome to greater than 94 percent, with low allelic bias, by adding components to the GenomiPhi buffer. This, she said, improves amplification of "notoriously challenging" regions, such as areas of high GC content.
The company chose to use Phi29 over PCR-based amplification because it generates long fragments of 10 kilobases without the need for DNA fragmentation and because it is "very compatible" with a microfluidic environment, Brown said.
The new protocol yields sufficient DNA from an individual cell for two sequencing runs plus quality control. The time from loading the cell suspension into the C1 to obtaining sequence-ready libraries for targeted resequencing is less than 24 hours, with less than six hours of hands-on time, she said.
The cost for the DNA sequencing sample prep workflow alone is less than $10 per cell, and under $25 per cell with sequencing library preparation.
In addition, the company is working on a new version of its Singular Analysis Toolset software that will support DNA variant analysis and help filter and visualize the data. The new software, expected to be available in early 2014, will be compatible with many existing DNA analysis pipelines, Brown said.
Fluidigm has validated the new sample prep workflow on different cell types, as well as in collaboration with several external sites.
For example, Paresh Vyas, a professor of hematology at the MCR Molecular Hematology Unit at the University of Oxford, and his team have used it to amplify DNA from primary cells from human acute myeloid leukemia samples and are awaiting their first sequencing results.
Vyas told In Sequence that the C1 has a "good cell capture rate" of about 80 percent. He likes that the system can visualize the captured cells prior to lysing them, and that the protocol is easy to follow.
Drawbacks of the system include that it can only capture one cell population at a time, he said, and that only about half the whole-genome amplifications are successful.
Also, the yield of amplified material is low — about 100 to 150 nanograms — whereas "conventional" whole-genome amplification kits, such as Qiagen's REPLI-g, have yields of 30 to 40 micrograms. That, he said, "may present problems" for whole-exome sequencing or whole-genome sequencing from single cells.
Vyas said the new workflow will be most useful for single-cell genotyping by targeted resequencing. In addition, he plans to use it to investigate the genomic heterogeneity of cancer cells by SNP and CGH array analysis.
Charles Gawad, a pediatric hematologist-oncologist and a researcher at Stanford University School of Medicine, has been using the C1 and the Access Array system for targeted resequencing of single cells from leukemia samples.
He said the automation of the microfluidic system in the C1 "removes the requirement of a skilled single-cell sorter" and reduces the chance of DNA contamination during the amplification reaction. Also, reagent costs are reduced because of the small reaction volumes used.
He and his colleagues have found the performance to be at least as good as other single-cell amplification approaches using multiple displacement amplification, he told In Sequence.
Gawad said allelic dropout and allelic imbalance are the main sources of experimental error in single-cell sequencing, and his team has minimized dropout "to a level that allows us to accurately reconstruct the evolution of our samples based on single-nucleotide variants."
But, he said, "it would be better if the amplification was more uniform to acquire other information," such as copy number variation, and "it would be optimal to perform the sequencing on a larger part of the genome in more cells," especially for tumors with high levels of clonal heterogeneity.
Sequencing costs, he said, are "currently the main limitation of those studies."
Fluidigm's Brown said that the single-cell sequencing field is primarily driven by "the hunt for somatic mutations, and really understanding their variability between tumors."
While bulk sequencing is able to identify somatic mutations that only occur in a subpopulation of tumor cells, it is difficult to distinguish them from sequencing errors. Single-cell sequencing makes it easier to see somatic mutation profiles across tumors, she said.
In addition to cancer research, Fluidigm believes single-cell DNA sequencing will also be of interest to researchers in immunology, neurodegenerative disorders, and aging.
Going forward, the company plans to develop the C1 for other applications, such as single-cell protein and epigenetic analysis. "Because of our partnerships and scientific collaborations with many thought leaders, we hear where they're trying to go," Brown said, "and we are rapidly trying to enable a broader suite of applications."