Visualization for single-cell multi-omic data
ChAIR-Viewer is developped for visualization single-cell multi-omic data. For ensemble data: bedgraph file (coverage signal), bedpe file (loop), bed file (peak/chromhmm/gene annotation) and more. For single cell data: fragment file. ChAIR-Viewer provides a variety of ways to sort cell, including: sorted by RNA UMI count, sorted by ATAC fragment count, sorted by contact/PET count, sorted by maximum contact distance, ordered of hierarchical clustering (in RNA/ATAC/PET data) and more.
ChAIR-Viewer is developped by R, some package are needed.
RColorBrewer
pheatmap
plotrix
dplyr
GenomicRanges
ggpubr
ggplot2
patchwork
1. Only the two longest and two exon most transcripts are retained (Convenient output of gene annotation track) (Optional)
python getMaxlengthRNA.py ~/cellranger-7.1.0/genome/refdata-gex-mm10-2020-A/genes/genes.gtf | grep -v "^#" | awk '$3=="exon" || $3=="UTR"'|perl -lane '$id="#";$id=$1 if(/transcript_id "(.+?)"/);$name="#"; $name=$1 if(/gene_name "(.+?)"/); print "$F[0]\t".($F[3]-1)."\t$F[4]\t$id\t$name\t0\t$F[6]\t$F[2]"' > mm10.genome.max.bed
else
cat ~/cellranger-7.1.0/genome/refdata-gex-mm10-2020-A/genes/genes.gtf | grep -v "^#" | awk '$3=="exon" || $3=="UTR"'|perl -lane '$id="#";$id=$1 if(/transcript_id "(.+?)"/);$name="#"; $name=$1 if(/gene_name "(.+?)"/); print "$F[0]\t".($F[3]-1)."\t$F[4]\t$id\t$name\t0\t$F[6]\t$F[2]"' > mm10.genome.max.bed
The readInfo.R script is used to organize the input files, and needs to be modified by the user according to your actual situation.
mkdir alldata
Rscript readInfo.R mm10.genome.max.bed mouseBrain mouseBrain.chromstat.hmm.bed
The program searches the current directory for the following files and stores all results as an RData object.
| file | description |
|---|---|
| mouseBrain.clean.ATAC.mouseBrain.fragments.unsorted.tsv | scATAC fragment file |
| mouseBrain.clean.ATAC.mouseBrain.q001_peaks.final.bed | peak file (3 columns) |
| mouseBrain.clean.ATAC.mouseBrain.bedgraph | scATAC coverage signal |
| mouseBrain.clean.ATAC.mouseBrain.RPKM.bedgraph | scATAC coverage signal by using RPKM normalization |
| mouseBrain.clean.PET.bedpe.mouseBrain.bedpe | scPET file (11 columns: chrom1, start1, end1, chrom2, start2, end2, barcode, readID, score, strand1, strand2) |
| mouseBrain.clean.PET.bedpe.mouseBrain.clusters.bedpe | loop file (no given anchor model) (7 columns: chrom1, start1, end1, chrom2, start2, end2, contact count) |
| mouseBrain.clean.PET.bedpe.mouseBrain.clusters.twopeaks.bedpe | loop file (no given anchor model) (overlap with peak) (7 columns: chrom1, start1, end1, chrom2, start2, end2, contact count) |
| mouseBrain.clean.PET.bedpe.mouseBrain.bedpe.ipet.loops | loop file ( given anchor model) (10 columns: chrom1, start1, end1, chrom2, start2, end2, contact count, anchor1 coverage, anchor2 coverage, VC_sqrt score) |
| mouseBrain.clean.GEX.mouseBrain.bedgraph | scRNA coverage signal |
| mouseBrain.clean.GEX.mouseBrain.RPKM.bedgraph | scRNA coverage signal by using RPKM normalization |
| mouseBrain.clean.dup.GEX.mouseBrain.bedgraph | scRNA (contain mapped reads from intergenic) coverage signal. (If there is no mouseBrain.clean.dup.GEX.mouseBrain.bedgraph file, you can copy the mouseBrain.clean.GEX.mouseBrain.bedgraph file) |
| mouseBrain.clean.dup.GEX.mouseBrain.RPKM.bedgraph | scRNA (contain mapped reads from intergenic) coverage signal by using RPKM normalization. (If there is no mouseBrain.clean.dup.GEX.mouseBrain.RPKM.bedgraph file, you can copy the mouseBrain.clean.GEX.mouseBrain.RPKM.bedgraph file) |
| mouseBrain.clean.GEX.mouseBrain.fragments.unsorted.tsv | scRNA fragment file |
| mouseBrain.clean.dup.GEX.mouseBrain.fragments.unsorted.tsv | scRNA (contain mapped reads from intergenic) fragment file. (If there is no mouseBrain.clean.dup.GEX.mouseBrain.fragments.unsorted.tsv file, you can copy the mouseBrain.clean.GEX.mouseBrain.fragments.unsorted.tsv file) |
Rscript plotData.R
黄加祥 (Jiaxiang Huang, 12207142@zju.edu.cn)
黄星宇 (Xingyu Huang, xingyu.huang@zju.edu.cn/huang182@live.cn)