SCUMIKit

A streamlined pipeline for processing UMI-tagged single cell RNA-Seq data. The pipeline does the following workflow in three simple commands.

Prerequisites

Read alignment and exact UMI collapsing: SRAToolkit, SAMtools, HomerTools ,BEDtools, UMItools, Bowtie and BWA

UMI sequencing error correction: CTK, CZPlib

Matrix imputation: impute

Multicore computing: doMC and parallel

Before running the pipeline, we need to prepare the index for the aligners.

For BWA, run the following command:

bwa index -a bwtsw Mus_musculus.GRCm38.cdna.all.fa

For Bowtie:

bowtie-build Mus_musculus.GRCm38.cdna.all.fa GRCm38

Usage

Install the prerequisite programs and set up the paths in SCUMI.sh
Then simply run

sh ./SCUMI.sh

A directory with resulting BED files will be created in the designated directory. Please run the following command to create the count matrix from the UMI-collapsed BED file in the directory. The output is an R object file (count_matrix.rda) and a CSV (count_matrix.csv) of the count matrix.

Rscript /path/to/BEDcounter_collapsed.R

If you want to collapse the file using multiple processors without CTK or UMItools. Please use the following command. The values after the R script is the number of cores.

Rscript /path/to/BEDcounter.R 12

To correct the count matrix using the k-NN-based single cell expression noise model, please run the follows. The output is an R object file (count_matrix_imputed.rda).

Rscript /path/to/NoiseCorrection.R count_matrix.rda

Reference Transcriptomes

The reference transcriptomes can be downloaded from Ensembl.

References

Bose, S., Wan Z., Carr A., Rizvi, A.H. et al. Scalable microfluidics for single-cell RNA printing and sequencing. Genome Biol. 16:120.

Islam, S., Zeisel, A., Joost, S., La Manno G. et. al. Quantitative single-cell RNA-seq with unique molecular identifiers. Nat. Methods 11:163-166.

Zhang, C., Darnell, R.B. 2011. Mapping in vivo protein-RNA interactions at single-nucleotide resolution from HITS-CLIP data. Nat. Biotech. 29:607-614.

Acknowledgements