bigSCale: An Analytical Framework for Big-Scale Single Cell Data

Single-cell RNA sequencing significantly deepened our insights into complex tissues and latest techniques are capable to analyze ten-thousands of cells simultaneously. With bigSCale, we provide an analytical framework being scalable to analyze millions of cells, addressing challenges of future large data sets. Unlike other methods, bigSCale does not constrain data to fit an a priori-defined distribution and instead uses an accurate numerical model of noise. We evaluate the performance of bigSCale using a biological model of aberrant gene expression and simulated data sets, which underlined its speed and accuracy. We further apply bigSCale to analyze 1.3 million cells from the mouse developing forebrain. The directed down-sampling strategy identified rare populations, such as Reelin positive Cajal-Retzius neurons, for which we determined a previously not recognized heterogeneity associated to distinct differentiation stages, spatial organization and cellular function. Together, bigSCale presents a perfect solution to address future challenges of large single-cell data sets. Single cell transcriptomes from 1847 neuronal progenitors differentiated from iPSC of one healthy donor, two patients with Williams-Beuren syndrome and two patients with 7q11.23 microduplication syndrome.单细胞RNA测序（single-cell RNA sequencing）极大地深化了我们对复杂组织的认知，而最新技术已可同时对数以万计的细胞开展分析。依托bigSCale，我们开发了可拓展至百万级细胞分析的分析框架，以应对未来大规模数据集的分析挑战。与其他方法不同，bigSCale不会强制数据拟合先验定义的分布，而是采用精准的噪声数值模型。我们通过异常基因表达的生物学模型与模拟数据集对bigSCale的性能进行评估，结果证实了其优异的速度与精度。我们进一步使用bigSCale分析了小鼠发育性前脑中的130万个细胞。通过定向下采样策略，我们成功识别出包括Reelin阳性Cajal-Retzius神经元在内的稀有细胞群，并发现了此前未被报道的异质性，该异质性与不同分化阶段、空间排布及细胞功能密切相关。综上，bigSCale为应对未来大规模单细胞数据集的分析挑战提供了理想解决方案。本数据集包含1847个神经元祖细胞的单细胞转录组数据，这些细胞由1名健康供体、2名威廉姆斯-博伦综合征（Williams-Beuren syndrome）患者以及2名7q11.23微重复综合征（7q11.23 microduplication syndrome）患者的诱导多能干细胞（induced pluripotent stem cell, iPSC）分化而来。