题目：Ultra-high-throughput single-cell RNA sequencing and perturbation screening with combinatorial fluidic indexing
期刊：Nature Methods
通讯作者：Christoph Bock
奥地利科学院分子医学研究中心，维也纳，奥地利
人工智能研究所，医学统计、信息学和智能系统中心，维也纳医科大学，奥地利维也纳
我们开发实验和计算方法，以获得独特的视角。例如，我们将下一代测序与生化技巧相结合，帮助我们绘制各种类型的表观遗传修饰图，并开发生物信息学算法，从此类高通量数据中推断相关的癌症生物学。他的研究将实验生物学（高通量测序、表观遗传学、CRISPR筛查、合成生物学）与癌症、免疫学和精密医学的计算方法（生物信息学、机器学习、人工智能）相结合。 他是10x Genomics公司的联合创始人，该公司开发和制造单细胞基因组学产品。他也是单细胞基因组学计划的联合创始人，这是一个非营利性组织，促进单细胞基因组学技术的开发和使用。
1 背景
微流液滴发生器（microfluidic droplet generators）是目前最流行的单细胞测序技术平台。它们具有高通量、直接处理和一致的数据质量，有助于在基础生物学和生物医学研究的许多领域广泛采用单细胞RNA-seq (scRNA-seq)。在基于液滴的scRNA-seq中，两个细胞被装载到同一个液滴中，它们的转录组将被标记为相同的细胞条形码。因此，在数据分析过程中，来自这两个细胞的转录本是不可区分的，从而产生了可能导致结果偏倚的cell doublet。为了最大限度地减少cell doublets的数量，单细胞悬浮液以非常低的浓度加载到微流体装置中，使得两个细胞不太可能进入同一液滴。这种解决方法对于许多应用来说已经足够了，但它导致了一个主要的概念限制:大多数液滴是完全功能的(即它们包含条形码微珠和逆转录试剂)，但从未接收细胞，因此不能产生任何单细胞转录组。所以，基于液滴的scRNA-seq试剂的使用率非常低，这导致了该方法的高成本，并使大型研究的成本高昂。
Cell hashing：DNA-labeled antibodies；lipid-tagged indices or expressed genetic barcodes。另外，genotype information derived from the transcriptome sequences can be used to identify and remove cell doublets when samples from genetically different individuals are pooled. 然而，由于大多数转录本不携带任何细胞特异性index，因此不可能解析doublet细胞的转录谱。因此，细胞哈希或细胞基因型在提高scRNA-seq通量方面的效用是有限的，需要更强大的方法。
2.问题及目的
问题：loading细胞数量少，为避免doublet；其他方法，不能很好地区别转录本
目的：对所有转录本进行细胞特异性条形码标记，然后对微流体液滴进行大规模过载。
3.工作流程
4.方案
4.1 上样量
10x Genomics Chromium system：maximum recommended loading concentration (15,300 nuclei per microfluidic channel).只有16.4%的液滴含有一个或多个核(平均每个液滴含有0.2个核)。值得注意的是，即使100倍的过载(每个通道153万个核)也会产生稳定的液滴乳液，并且不会堵塞微流体系统。当我们增加加载浓度时，液滴填充率和每滴核数以可控的方式增加，液滴填充率高达95.5%，平均每滴9.6个核(每个通道153万个核)，液滴直径高度一致。
Chromium ATAC-seq reagents
4.2 细胞/细胞核固定
4.2.1 Sample preparation and permeabilization
A total of 5 million cells were washed with 10 ml of ice-cold 1× PBS (Gibco catalog no. 14190-094) with centrifugation (300g,5 min, 4 °C) and fixed in 5 ml of ice-cold methanol (Fisher Scientific catalog no.M/4000/17) at −20 °C for 10 min. After two further washes (centrifugation: 300g,5 min, 4 °C) with 5 ml of ice-cold PBS-BSA-SUPERase (1× PBS supplemented with 1% w/v BSA (Sigma catalog no. A8806-5) and 1% v/v SUPERase-In RNase Inhibitor (Thermo Fisher Scientific catalog no. AM2696)), permeabilized cells were resuspended in 200 μl of ice-cold PBS-BSA -SUPERase (1× PBS supplemented with 1% w/v BSA and 1% v/v SUPERase-In RNase Inhibitor), and filtered through a cell strainer (40 μM or 70 μM depending on the cell size). We then used 5 μl of the sample for cell counting in duplicates on a CASY device (Schärfe System) and diluted to 5,000 cells μl−1 with ice-cold PBS-BSA-SUPERase. We immediately proceeded with the reverse transcription step of scifi-RNA-seq.
4.2.2 细胞核准备
4.2.2.1细胞提核方案
Preparation of fresh nuclei （方法类似10x genomics）
5million cells in 500 µl of ice-cold Nuclei Preparation Buffer (10mM Tris-HCl pH 7.5, 10mM NaCl, 3mM MgCl2, 1% w/v BSA, 1% v/v SUPERase-In RNase Inhibitor, 0.1% v/v Tween-20, 0.1% v/v IGEPAL CA-630 and 0.01% v/v Digitonin, followed by 5min of incubation on ice. Lysis of the plasma membrane was stopped by adding 5ml of ice-cold Nuclei Wash Buffer (10mM Tris-HCl pH 7.5, 10mM NaCl, 3mM MgCl2, 1% w/v BSA, 1% v/v SUPERase-In RNase Inhibitor and 0.1% v/v Tween-20). Nuclei were collected by centrifugation (500g, 5min, 4 °C), resuspended in 200µl of ice-cold PBS-BSA -SUPERase, 之后同上。
Preparation of nuclei——to fix
5million cells in 500 µl of ice-cold Nuclei Preparation Buffer without Digitonin and without Tween-20 (10mM Tris-HCl pH 7.5, 10mM NaCl, 3mM MgCl2, 1% w/v BSA, 1% v/v SUPERase-In RNase Inhibitor and 0.1% v/v IGEPAL CA-630, followed by 5min of incubation on ice. Lysis of the plasma membrane was stopped by addition of 5ml of Nuclei Wash Buffer without Tween-20 (10mM Tris-HCl pH 7.5, 10mM NaCl, 3mM MgCl2, 1% w/v BSA, 1% v/v SUPERase-In RNase Inhibitor). Nuclei were collected by centrifugation (500g, 5min, 4 °C),
4.2.2.1细胞核固定
nuclei were fixed in 5ml of ice-cold 1× PBS containing 1–4% formaldehyde (Thermo Fisher Scientific catalog no. 28908) for 15min on ice. Fixed nuclei were collected (500g, 5min, 4 °C), the pellet was resuspended in 1.5ml of ice-cold Nuclei Wash Buffer without Tween-20 and transferred to a 1.5ml tube. After one more wash with 1.5ml of ice-cold Nuclei Wash Buffer without Tween-20 (500g, 5min, 4 °C), fixed nuclei were resuspended in 200 µl of Nuclei Wash Buffer without Tween-20, snap-frozen in liquid nitrogen and stored at −80 °C. （猜测：修改方案，应该是为了适用于细胞核保存）
对于scifi-RNA-seq，冷冻样品在37°C水浴中解冻1分钟后，立即放在冰上。离心(500g, 5min，4°C)，固定核重悬于250 μ l ice-cold Permeabilization Buffer (10mM Tris-HCl, 10mM NaCl, 3mM MgCl2, 1% w/v BSA, 1% v/v SUPERase-In RNase Inhibitor, 0.01% v/v Digitonin , 0.1% v/v Tween-20 for 5min of incubation on ice。250 µl of Nuclei Wash Buffer without Tween-20 was added per sample, and nuclei were collected (500g, 5min, 4 °C). After one more wash with 250 µl of Nuclei Wash Buffer without Tween-20, nuclei were taken up in 100 µl of PBS-BSA-SUPERase. 之后同4.2.1。
4.3 Detailed scifi-RNA-seq protocol description
4.3.1 Reverse transcription
10,000 permeabilized cells or nuclei: 25 µM; 5min at 55 °C
Maxima H Minus Reverse Transcriptase
(with heated lid set to 60 °C): 50 °C for 10min, 3 cycles of [8 °C for 12s, 15 °C for 45 s, 20 °C for 45 s, 30 °C for 30 s, 42 °C for 2min, 50 °C for 3min], 50 °C for 5min, store at 4 °C. Processed cells or nuclei were washed with ice-cold 1× PBS containing 1% BSA.
酶的对比：Maxima H Minus versus Superscript IV。Maxima H Minus会更好一些。
4.3.2 Microfluidic thermoligation barcoding on the Chromium scATAC v1.0 platform/ scATAC v1.1 (Next GEM) platform
准备sample：以scATAC v1.1 (Next GEM) platform为例，The resulting pellet was resuspended in a mix of 41.9 µl of nuclease-free water, 12 µl of 10x Ampligase Reaction Buffer (Lucigen catalog no. A0102K), 2.3 µl of 100Uμl−1 Ampligase , 1.5 µl of Reducing Agent B and 2.3 µl of 100 µM Bridge Oligo。(直接把连接RT引物和ATAC beads 的oligo 以及DNA连接酶加入10x ATAC beads反应体系，加第二轮10x cell barcodes)
纯化cDNA：第一步Dynabeads MyOne Silane纯化类似10x ；第二步改成 1.0× cleanup with SPRIselect beads，eluting in 22 µl of EB Buffer.
PS：两个平台Chromium scATAC v1.0 platform/ scATAC v1.1 (Next GEM) platform，数据差异不大。
4.4 Template switching
20 ul of sample from the previous step was mixed with 10 µl of 5× Reverse Transcription Buffer, 10 µl of Ficoll PM-400 (20%, Sigma catalog no. F5415-50ML；maybe起到类似PEG8000左右，可以分割分子), 5 µl of 10mM dNTPs, 1.25 µl of Recombinant Ribonuclease Inhibitor, 1.25µl of 100µM Template Switching Oligo and 2.5 µl of Maxima H Minus Reverse Transcriptase. The template switching reaction was incubated for 30min at 25 °C, 90min at 42 °C, storage at 4 °C and cleaned with a 1.0× SPRI cleanup, eluting in 17 µl of EB buffer.
4.5 cDNA enrichment.
15 ul of sample from the step above was mixed with 33µl of nuclease-free water, 50µl of NEBNext High-Fidelity 2× PCR Master Mix (NEB catalog no. M0541S), 0.5µl of 100µM Partial-P5 primer, 0.5µl of 100µM TSO Enrichment Primer, and 1µl of 100×SYBR Green in DMSO (Life Technologies catalog no. S7563). cDNA was amplified in a thermocycler as follows: 98°C for 30s, cycle until fluorescent signal >1,000 relative fluorescence units (RFU; 98°C for 20s, 65°C for 30s, 72°C for 3min), 72°C for 5min in another thermocycler, storage at 4°C. cDNA was cleaned by one 0.8× SPRI cleanup followed by a 0.6× SPRI cleanup, quantified with a Qubit HS assay (Thermo Fisher Scientific catalog no. Q32854), and 1.5ng were checked on a Bioanalyzer High-Sensitivity DNA chip (Agilent catalog no. 5067-4626 and 5067-4627).
4.6 Tagmentation
EZ-Tn5 Transposase (Lucigen catalog no. TNP92110)
Tn5 reaction buffer was prepared: 50mM TAPS (Sigma catalog no. T9659-100G), 25mM MgCl2 (Ambion catalog no. AM9530G); the pH was adjusted to 8.5 and the solution was sterile filtered. Tn5 dilution buffer was prepared: 50mM Tris-HCl pH 7.5, 100mM NaCl , 0.1mM EDTA (Invitrogen catalog no. AM9260G), 50% glycerol (Sigma catalog no. G5516-100ML), 0.1% Triton-X100 (Sigma catalog no. X100-100ML), and supplemented with fresh 1mM DTT (Sigma catalog no. 646563-10x.5ML) before use. To achieve maximum library complexity, the entire sample was processed in several tagmentation reactions with 1ng input each. cDNA was diluted to 0.2ngµl−1with nuclease-free water and 5 µl (1ng) per reaction was distributed into a 96-well plate on ice. A mix of 11.25µl nuclease-free water, 5 µl of 5× Tn5 reaction buffer, 2.5 µl of dimethylformamide (Sigma catalog no. D4551-250ML), and 1.25 µl of freshly diluted i7-only transposome (prepared as described below and diluted 1:4.5 in Tn5 dilution buffer) was added. Reactions were incubated for 10min at 55 °C, then cooled for 1min on ice. The enzyme was inactivated by addition of 2.5 µl of 1% SDS (Sigma catalog no. 71736-100ML) for 5min at room temperature. Next, the volume was brought to 50 µl and the fragmented cDNA was purified with a 1.0× SPRI cleanup, eluting in 17µl of EB buffer.
4.7 Library enrichment.
NEBNext High-Fidelity 2× PCRMaster mix
Partial-P5 primer
barcoded P7 primer
Tm= 65°C
Libraries were cleaned with a 0.7× SPRI cleanup with AMPure XP beads. All wells with the same P7 barcode were pooled and subjected to a 0.8× SPRI cleanup, eluting in 0.2 bead volumes of EB buffer.
5.应用（CRISPR screen - scifi data）
6. 附录(需关注的)
6.1 doublet rates
总体来说：相比于其它技术（SPLIT-seq；sci-RNA seq；sci-Plex细胞哈希和多样本sci-rna-seq结合检测）或intact 细胞/nuclei; MeOH-fixed cells 展现最低的doublet rates。
6.2 不同固定条件，数据质量
Performance metrics for scifi-RNA-seq experiments using a mixture of human Jurkat cells and mouse 3T3 cells, starting from whole cells permeabilized by methanol, freshly isolated nuclei, and nuclei fixed with 1% or 4% formaldehyde (cryopreserved, re-hydrated, and permeabilized)
FA固定的细胞核并没有展现出很好的数据，除了细胞核本身很大程度是因为冻存的影响，但FA固定影响也存在（从4%和1%数据来看，貌似差别不大）。此外，freshly isolated nuclei 数据是最好的。