GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM5343035

Query DataSets for GSM5343035

Status

Public on Jun 03, 2021

Title

E2CD1col3

Sample type

SRA

Source name

differentiating iPSC-derived cardiomyocytes

Organism

Homo sapiens

Characteristics

cell line: multiplexed sample (pool of 18907, 19159, 18489)
differentiation day: multiplexed sample (pool of 7, 3, 1 respectively)

Treatment protocol

To begin cardiomyocyte differentiation, E8 media was replaced with "heart media"--RPMI with B27 supplement minus insulin, 2mM GlutaMAX, and 100mg/ml Penicllin/Streptomycin--along with 1:100 Matrigel hESC-qualifed Matrix and 12uM of CHIR99021 trihydrochloride on Day 0. On Day 1, this media was removed and fresh "heart media" was added. On Day 3, new "heart media" was added along with 2uM of Wnt-C59. On Days 5, 7, 10, 12, and 14, old media was removed and replaced with fresh "heart media."

Growth protocol

Feeder-free iPSC cultures were maintained on Matrigel Growth Factor Reduced Matrix with Essential 8 Medium and Penicillin/Streptomycin.

Extracted molecule

polyA RNA

Extraction protocol

Samples were collected using a 125 um Drop-seq microfluidic device. Single cells were captured in droplets along with a cell-specific barcode bead. After collection, RNA molecules were reverse transcribed and cDNA amplification performed according to the Drop-seq protocol (Macosko et al 2015).
Library preparation was performed using the Illumina Nextera XT DNA Library Preparation Kit, for paired-end sequencing using the Illumina NextSeq 500.

Library strategy

RNA-Seq

Library source

transcriptomic

Library selection

cDNA

Instrument model

Illumina NextSeq 500

Description

collection_metadata.txt
cell_counts.mtx
cell_counts_sctransform.mtx
pearson_residuals_sctransform.tsv

Data processing

Paired-end reads were combined to form a (cells x genes) UMI counts matrix using dropseqRunner (https://github.com/aselewa/dropseqRunner)
Transcripts were aligned to the genome (GRCh38) using STAR-solo (Dobin et al. 2013)
featureCounts (Liao et al. 2014) assigned each read to a genomic feature, and umi_tools (Smith et al. 2017) created a count matrix representing the frequency of each feature in the dataset
Demuxlet (Kang et al. 2018) was used to demultiplex the pooled single cell data, and assign to each cell a probability that the cell was a doublet as well as a maximum likelihood donor (cell line) assignment
Genes were removed that were detected in fewer than 10 cells, and cells were removed if they were assigned by demuxlet to have a posterior probability of doublet assignment greater than 0.3, were not assigned to a single cell line donor, had more than 25% reads mapping to mitochondrial genes, or had fewer than 300 unique genes detected. This left a total of 38,943 genes (of 60,668) and 230,786 cells (of 564,362)
Seurat v3 and SCTransform were used to normalize the expression data while accounting for overdispersion and cell cycle effects. Cell cycle scoring was performed with the Seurat package, assigning a S phase and G2/ M phase score to each cell. The difference between these two scores was regressed out during normalization of the expression data.
Genome_build: GRCh38
Supplementary_files_format_and_content: cell_counts.mtx contains the raw UMI counts for each cell (in the MatrixMarket format). The files cell_indices.tsv and gene_indices.tsv are used for interpreting the matrix files.
Supplementary_files_format_and_content: cell_counts_sctransform.mtx contains the expected UMI counts for each cell (in the MatrixMarket format) after normalizing the data using SCTransform (Hafemeister et al. 2019). These adjusted counts were used to aggregate pseudobulk
Supplementary_files_format_and_content: pearson_residuals_sctransform.tsv contains the Pearson residuals for each cell (in the MatrixMarket format) after normalization using SCTrasnform (Hafemeister et al. 2019). These Pearson residuals were used for visualization, cell type annotation, and pseudotime inference
Supplementary_files_format_and_content: experimental_design.txt describes the mapping of cell lines to their pooled collections. Columns are 'Exp' (experimental round, 1-3); 'CD' (collection day, 1-3); 'col' (collection, 1-6 for experiments 1 and 3, 1-7 for experiment 2), 'Line' (cell line), 'Day' (differentiation day)
Supplementary_files_format_and_content: collection_metadata.txt contains metadata for each collected sample. Rows are samples, columns are 'Line' (cell line), 'Differentiation Day' (differentiation day), 'Feeder Passage', 'FF Passage' (feeder free passage), 'Date' , 'CD' (collection day), 'col' (collection), 'orig.ident' (exp/collection day/collection), 'Experiment Batch', 'Diff Start Batch', "How many wells collected", "Confluence when collected as percent", "Day 0 Confluence (%)", "Beating on day of collection Y 1/N 0", "Day this well started beating", "% beating when collected", "First day beating for this line/diff start", "avg. day this line/diffStart started beating", "Networks+Stringy (0- none noted, 1--thin, 2- chunky, if no adjective-1)", "flash freezing or cryopreserve (flash-0, cryo-1)", "Sheets vs. clumps (0-no notes, 1- clumps, 2- sheets, 3-both)", "debris (0-none or not noted, 1-some, 2- lots)", "swirls/semi-circles/background (not noted-0, noted-1)"
Supplementary_files_format_and_content: cell_metadata.tsv contains metadata for each cell. Rows are cells, columns are 'cell', 'exp.grp' (experiment group), 'sample' (sample), 'diffday' (differentiation day), 'individual' (cell line), 'S.Score' (S score, computed in Seurat), 'G2M.Score' (G2/M score, computed in Seurat), 'CC.Difference' (G2M.Score - S.Score, regressed out during SCTransform normalization), 'demux.dbl.prb' (demuxlet posterior probability of doublet assignment), 'leiden' (leiden assignment from scanpy), 'type' (assigned cell type, based on leiden clustering), 'dpt_pseudotime' (inferred diffusion pseudotime)

Submission date

May 27, 2021

Last update date

Jun 03, 2021

Contact name

Joshua M Popp

E-mail(s)

jpopp4@jhu.edu

Organization name

Johns Hopkins University

Department

Biomedical Engineering