Existing protocols for full-length single-cell RNA sequencing produce libraries of high complexity (thousands of distinct genes) with outstanding sensitivity and specificity of transcript quantification. These full-length libraries have the advantage of allowing probing of transcript isoforms, are informative regarding single-nucleotide polymorphisms and allow assembly of the VDJ region of the T- and B-cell-receptor sequences. Since full-length protocols are mostly plate-based at present, they are also suited to profiling cell types where cell numbers are limiting, such as rare cell types during development. A disadvantage of these methods has been the scalability and cost of the experiments, which has limited their popularity as compared with droplet-based and nanowell approaches. Here, we describe an automated protocol for full-length single-cell RNA sequencing, including both an in-house automated Smart-seq2 protocol and a commercial kit–based workflow. The protocols take 3–5 d to complete, depending on the number of plates processed in a batch. We discuss these two protocols in terms of ease of use, equipment requirements, running time, cost per sample and sequencing quality. By benchmarking the lysis buffers, reverse transcription enzymes and their combinations, we have optimized the in-house automated protocol to dramatically reduce its cost. An automated setup can be adopted easily by a competent researcher with basic laboratory skills and no prior automation experience. These pipelines have been employed successfully for several research projects allied with the Human Cell Atlas initiative (
www.humancellatlas.org
).
In this protocol, the authors describe two automated versions of the Smart-seq2 method for full-length single-cell RNA sequencing: a medium-throughput variant using off-the-shelf reagents and a high-throughput version using a commercially available kit....

High-throughput full-length single-cell RNA-seq automation

10.1038/s41596-021-00523-3