Ercc spike ins

Cells differentiated at day 13 were harvested with TrypLE Select as described above, resuspended in Stage 5 medium supplemented with 10 μM Y-27632, filtered using a 50 μM sterile Filcon (BD Biosciences) and sorted using a FACSAria Fusion cell sorter (BD) directly into 2 x 384-well plates with ERCC spike-ins (Agilent), reverse transcription primers and dNTPs (both Promega), according to the gating shown in Figure S10B. Repartition was as follow: 216 cells for each NKX6-1-GFP+, NEUROG3-mCherry+ and NKX6-1-GFP+/NEUROG3-mCherry+ population and 104 cells for the negative population. Single cell sequencing was performed according to the Sort-seq method⁶⁸ (link) by Single Cell Discoveries. Briefly, Sequencing libraries were generated with TruSeq small RNA primers (Illumina) and sequenced paired-end at 60 and 26 bp read length, respectively, on the Illumina NextSeq. Reads were mapped to the human GRCh38 genome assembly. Sort-seq read counts were filtered to exclude reads with identical library-, cell- and molecule barcodes. UMI counts were adjusted using Poisson counting statistics.⁶⁸ (link)