Dataset View [GSE65525]

TitleDroplet barcoding for single cell transcriptomics applied to embryonic stem cells
ArticleKirschner MW,Weitz DA,Peshkin L,Li V,Veres A,Tallapragada N,Akartuna I,Mazutis L,Klein AM.Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells.Cell.2015 May 21
Bio ProjectBioProject: http://www.ncbi.nlm.nih.gov/bioproject/PRJNA274274
SraSRA: http://www.ncbi.nlm.nih.gov/sra?term=SRP053052
Overall DesginA total of 8 single cell data sets are submitted: 3 for mouse embryonic stem (ES) cells (1 biological replicate, 2 technical replicates); 3 samples following LIF withdrawal (days 2,4, 7); one pure RNA data set (from human lymphoblast K562 cells); and one sample of single K562 cells.
SummaryRecently, RNA sequencing has achieved single cell resolution, but what is limiting is an effective way to routinely isolate and process large numbers of individual cells for in-depth sequencing, and to do so quantitatively. We have developed a droplet-microfluidic approach for parallel barcoding thousands of individual cells for subsequent RNA profiling by next-generation sequencing. This high-throughput method shows a surprisingly low noise profile and is readily adaptable to other sequencing-based assays. Using this technique, we analyzed mouse embryonic stem cells, revealing in detail the population structure and the heterogeneous onset of differentiation after LIF withdrawal. The reproducibility and low noise of this high-throughput single cell data allowed us to deconstruct cell populations and infer gene expression relationships.
Experimental ProtocolCells were encapsulated into droplets on ice and lysed in the 4nL microfluidic droplets using a final concentration of 0.4% NP-40. Single cell lysates were subject to reverse transcription at 50°C without purification of RNA.; Cells were barcoded using the db-Seq platform, which makes use of the CEL-Seq protocol for library construction (Hashimshony et al., Cell Reports 2011).
Data processingReads were first filtered based on presence in read 1 of two sample barcode components separated by the W1 adaptor sequence (GAGTGATTGCTTGTGACGCCTT); Read 2 was then trimmed using Trimomatic (5) (version 0.30; parameters: LEADING:28 SLIDINGWINDOW:4:20 MINLEN:19).; Barcodes for each read were matched against a list of the 3842 pre-determined barcodes, and errors of up to two nucleotides mismatch were corrected. Reads with a barcode separated by more than two nucleotides from the reference list were discarded. The reads were then split into barcode-specific files for mapping and UMI filtering.; The trimmed reads were aligned using Bowtie (version 0.12.0, parameters: -n 1 -l 15 -e 300 -m 200 -best -strata -a) to the mouse transcriptome. The reference transcriptome was built using all annotated transcripts (extended with a 125bp poly-A tail) from the UCSC mm10 genome assembly.; We used a custom Python and PySAM script to process mapped reads into counts of UMI-filtered transcripts per gene.; Genome_build: mm10 for ES cells; hg19 for K562 cells.; Supplementary_files_format_and_content: Columns = cells; rows = genes. Column 1 contains gene symbols. Values show unique molecular identifier (UMI)-filtered counts per cell detected in the raw data. No normalization is performed.
Public OnPublic on May 20 2015

Cell Groups

 human K562 cells[239]