Comparative analyses of single-cell RNA sequencing protocols used in the Human Cell Atlas and similar projects reveal performance differences between methods, according to two studies appearing in this week's Nature Biotechnology. In the first report, a group led by scientists from the Barcelona Institute of Science and Technology generated benchmark datasets that they used to compare 13 commonly used scRNA-seq and single-nucleus RNA-seq protocols. Their analysis shows marked differences in protocol performance, with protocols varying in library complexity and their ability to detect cell-type markers, "impacting their predictive value and suitability for integration into reference cell atlases," the scientists write. In the second study, a Broad Institute team compared seven methods for single-cell and/or single-nucleus profiling, including two low-throughput and five high-throughput methods. The methods were tested on cell lines, peripheral blood mononuclear cells, and brain tissue, generating 36 libraries in six separate experiments in a single center. Among the findings were differences in consistent performance among the high-throughput methods and superior performance by the low-throughput methods in studies that require the highest sensitivity.
Investigators from the University of Chicago report in Nature Methods this week a new approach to labeling single-stranded DNA for sequencing. Called kethoxal-assisted single-stranded DNA sequencing, or KAS-seq, the technique takes advantage of the fast and specific reaction between N3-kethoxal and guanines in ssDNA. According to its developers, it enables the sensitive and genome-wide capture and mapping of ssDNA produced by transcriptionally active RNA polymerases or other processes in situ using as few as 1,000 cells, all within five minutes. The team also shows that KAS-seq detects transcription dynamics during transient physiological environment changes, such as inhibition of protein condensation.