GEO Accession viewer

NCBI > GEO > Accession Display

Not logged in | Login

GEO help: Mouse over screen elements for information.

Sample GSM5618816

Query DataSets for GSM5618816

Status

Public on Apr 05, 2024

Title

Sample 2a [SKG Low Rep 2]

Sample type

SRA

Source name

naive CD4 T cells

Organism

Mus musculus

Characteristics

strain: Balb/c
tissue: lymph node
genotype: SKG
subgroup: SKG_Low
cell type: naive CD4 T cells

Extracted molecule

total RNA

Extraction protocol

Negatively selected CD4 T cells from the lymph node were sorted for CD62LhiCD44loCD25- and the 10% lowest (GFPlo) expressing T cells. Cells were washed, pelleted and immediately flash frozen using dry ice in ethyl alcohol. Sample was processed with Illumina TruSeq Stranded mRNA kit (Cat#: RS-122-2103) for library preparation.
RNA libraries were prepared for sequencing using standard Illumina protocols

Library strategy

RNA-Seq

Library source

transcriptomic

Library selection

cDNA

Instrument model

Illumina HiSeq 2500

Data processing

Bulk RNA Seq: The raw fastq files were clipped and filtered using fastq-mcf v.1.04.636 to remove low quality reads and bases, homopolymers, and adapter sequences. The filtered reads were aligned using the STAR v.2.4 with the default settings to the mm10 transcriptome to produce count matrices for each sample.
Single cell RNA seq: The raw fastq files were aligned using CellRanger v3.0.1 and 3.0.2 software with the default settings to the mm10 transcriptome with the addition of the sequence for the eGFP transcript and the vdj GRCm38 v 3.1.0 reference for the GEX and TCR fastqs, respectively.
We filtered out 721 cells with less than 100 or more than 3000 genes detected and filtered out 14,388 genes detected in less than 3 cells. We also filtered out 1,066 cells with more than 10% of total counts (UMIs) mapping to mitochondrial genes and 1008 cells determined to be contaminating B cells based on CD19 expression.
The raw counts were normalized to 10,000 counts, log1p-normalized and scaled.
For technical and batch correction, we regressed out total UMI counts and % counts mapping to mitochondrial genes and used combat for batch correction with each sample as a batch.
We identified 1119 highly variable genes (excluding all Trav and Trbv genes to avoid clustering cells based on expression of those genes) which were used with the default settings in scanpy v.1.4.3 for PCA analysis followed by leiden clustering after nearest neighbor detection and UMAP projection. This analysis identified 13 clusters which we collapsed into 9 cell sub-types based on differential gene analysis.
For each 10x well alignment output from cellranger, we used velocyto v.0.17.17 to create a loom file with the spliced, unspliced, and ambiguous counts with the Dec. 2011 GRCm38/mm10 repeat masking gtf file from the UCSC genome browser 70, 71. The loom files across all wells were merged and then subsetted to all cells in the T.4NNr4a1 cluster.
The resulting object was used to run the dynamical model from scvelo v.0.2.1 with the default settings to uncover the RNA velocity to predict the latent time for each cell.
We used we used a Gaussian mixture model with the GaussianMixture tool from sklearn v.0.23.1 to deconvolute four underlying individual Gaussian distributions from the latent time distribution for cells from the T.4NNr4a1 cluster.
For trajectory inference between the four distributions (“Stage 1” – “Stage 4”), we used the graph-based tool PAGA within scvelo to predict velocity-inferred transitions among the clusters.
Genome_build: mm10
Genome_build: vdj GRCm38 v 3.1.0
Supplementary_files_format_and_content: comma separated text file with raw gene counts for every gene and every sample
Supplementary_files_format_and_content: h5ad file with filtered, processed, and cluster annotated data from all eight combined single cell RNA seq libraries
Supplementary_files_format_and_content: h5ad file with cells from T.4 Nr4a1 cluster from single cell RNA seq data post trajectory analysis with scvelo
Supplementary_files_format_and_content: comma separated text file with clonotype and TCR chain assignment for each cell barcode in the indicated library

Submission date

Oct 08, 2021

Last update date

Apr 05, 2024

Contact name

Yang Sun

E-mail(s)

yang.sun@ucsf.edu

Organization name

UCSF

Department

Rheumatology