Our OP² single-cell transcriptomics pipeline is a bioinformatics analysis workflow used for single-cell RNA sequencing data.
It allows you to analyze your RNA sequencing data using this gold standard analysis pipeline.
You get insights into the quality of your data, expression profiles of your cells, differential expression levels of multiple genes, cell annotations and identities, and gene enrichment analysis.
The workflow processes raw data from FastQ inputs, aligns the reads, generates counts relative to genes and performs extensive quality-control on the results.
These results are made available to you via two interactive reports, and a data package with all essential intermediate files to perform more in-depth data analysis.
The pre-processing workflow processes your raw sequence data until QC approved aligned data.
Next, the post-processing workflow enables you to review the biological meaning of your data via a statistical analysis approach.
Droplet-based (e.g. 10X Genomics)
Compressed raw FastQ files (R1 and R2)
Chemistry specific barcodes (V2 or V3)
Reference transcriptome (hg19 or hg38 or mm10)
Extract cell barcodes to retrievesingle cell information
Cell barcodes + UMI’s
Reads with low-quality cell barcodes are discarded
STARsolo aligns reads to reference transcriptome
Alignment statistics: read depths, per base, GC content, …
Load cell-gene count matrices
Produce high count matrix
Identification of cells from empty droplets
Removal of barcode-swapped pseudo-cells
Downsampling of the count matrix
Identification of low quality libraries
Number of UMI's, low expressed genes and percentage of mitochondrial
Automated mean absolute deviation (MAD) thresholding
Removed matrix QC values with MAD above 3
Library size normalisation to remove technical biases
Identify highly variable features
Select most variable genes that contain useful information about the biology
Remove genes that contain noise
Integrate Seurat objects
Format object to perform statistical analysis
Linear transformation to give equal weights to all genes
Avoid highly-expressed gene to dominate
Shift gene expression values to cell mean of 0
Shift gene variance values to cell mean of 1
Linear dimension reduction
Principal components analysis (PCA) is performed to denoise and compact the data prior to post-processing.
Select components based on the Elbow Plot
Construct K-Nearest neighbor graph on Euclidean distancein PCA space
Refine by Jaccard similarity
Cluster cells by modularity optimization Louvain algorithms
Non-linear dimension reduction
t-distributed stochastic neighbour embedding (t-SNE) is widely used for visualizing complex single-cell data sets. This is useful as it improves speed by using a low-rank approximation of the expression matrix; and reduces random noise, by focusing on the major factors of variation.
Assign cell types
Unbiased cell type recognition from single-cell RNA sequencing data, by leveraging reference transcriptomic datasets of pure cell types to infer the cell of origin of each single cell independently.
Identify cell markers
If a cell in the test dataset is confidently assigned to a particular label, we would expect it to have strong expression of that label’s markers. This can be useful if you want to find markers that are conserved between a treated and untreated condition for a specific cell type or group of cells.
Identify differential expressed genes
After identifying conserved markers, a comparative analyses is performed on the differences induced by stimulation/treatment. We take the average expression of all clusters and generate the scatter plots, highlighting genes that are identified in previous step.
Check over-repressentations of genes or gene products across conditions.