Single-cell RNA sequencing (scRNA-seq) reveals the transcriptional heterogeneity of cells, but the static snapshots fail to reveal the time-resolved dynamics of transcription. Herein, we develop Well-TEMP-seq, a high-throughput, cost-effective, accurate, and efficient method for massively parallel profiling the temporal dynamics of single-cell gene expression. Well-TEMP-seq combines metabolic RNA labeling with scRNA-seq method Well-paired-seq to distinguish newly transcribed RNAs marked by T-to-C substitutions from pre-existing RNAs in each of thousands of single cells. The Well-paired-seq chip ensures a high single cell/barcoded bead pairing rate (~80%) and the improved alkylation chemistry on beads greatly alleviates chemical conversion-induced cell loss (~67.5% recovery). We further apply Well-TEMP-seq to profile the transcriptional dynamics of colorectal cancer cells exposed to 5-AZA-CdR, a DNA-demethylating drug. Well-TEMP-seq unbiasedly captures the RNA dynamics and outperforms the splicing-based RNA velocity method. We anticipate that Well-TEMP-seq will be broadly applicable to unveil the dynamics of single-cell gene expression in diverse biological processes.

单细胞 RNA 测序 (scRNA-seq) 揭示了细胞的转录异质性，但静态快照无法揭示转录的时间分辨动态。在此，我们开发了 Well-TEMP-seq，这是一种高通量、具有成本效益、准确且高效的方法，用于大规模并行分析单细胞基因表达的时间动态。Well-TEMP-seq 将代谢 RNA 标记与 scRNA-seq 方法 Well-paired-seq 相结合，以区分由 T-to-C 替换标记的新转录 RNA 和数千个单细胞中每个细胞中已有的 RNA。Well-paired-seq 芯片可确保高单细胞/条形码珠子配对率 (~80%)，并且珠子上改进的烷基化化学极大地减轻了化学转化引起的细胞损失（~67.5% 回收率）。我们进一步应用 Well-TEMP-seq 来分析暴露于 5-AZA-CdR（一种 DNA 去甲基化药物）的结直肠癌细胞的转录动力学。Well-TEMP-seq 无偏地捕获 RNA 动力学，优于基于剪接的 RNA 速度方法。我们预计 Well-TEMP-seq 将广泛应用于揭示不同生物过程中单细胞基因表达的动态。