Here, we use compass to characterize cellular metabolic states based on single-cell RNA-Seq. Downstream visualization was conducted based on the author's analysis examples.
compass
The input gene expression matrix can be a tab-delimited text file (tsv) containing gene expression estimates (CPM, TPM, or similar scaled units) with one row per gene, one column per sample. Here, the author's data (rename as th17_geneuexpression_matrix.tsv) is used as the test data.
1)Here is a simple example of running Compass. For more parameters, please refer to https://yoseflab.github.io/Compass/user-guide.html.
source activate compass
compass --data th17_gene_expression_matrix.tsv \
--num-processes 20 \
--species homo_sapiens \
--output-dir compass_outdir
2)Visualize the results of Compass
step 1: run compass_plot.py
source activate compass
python ../script/compass_plot.py \
--reaction_penalties compass_outdir/reactions.tsv \
--metadata ../data/reaction_metadata.csv \
--group_info group_info.tsv \
--subsystem Glycolysis/gluconeogenesis \
--outdir compass_plot_outdir \
--name Glycolysis_gluconeogenesis
help info:
----
--reaction_penalties reactions.tsv from the result of compass
--metadata reaction metadata provided by the author
--group_info a tsv contains group info. The first column corresponding to barcodes, The second column corresponding to group
--subsystem subsystem used to plot, such as Glycolysis/gluconeogenesis. More subsystem can be found in the reaction metadata provided by the author
--plot_type plot type, anno or no(t plot anno). Default: no
--outdir output dir
--name name, just like subsystem, special symbols need to be removed
step 2: run compass_plot_update.R
source activate r4.1_env
Rscript ../script/compass_plot_update.R \
--wilcox_res compass_plot_outdir/wilcox_res.tsv \
--subsystem Glycolysis/gluconeogenesis \
--xlab_name KD_vs_WT \
--annotation_text_size 2 \
--outdir compass_plot_outdir \
--name Glycolysis_gluconeogenesis
help info:
----
--wilcox_res wilcox result from running compass_plot.py
--subsystem subsystem, need to be consistent with the subsystem parameter used for running compass_plot.py
--xlab_name xlab name of compass_plot.pdf from compass_plot.py
--annotation_text_size annotation text size, default: 3
--outdir output dir
--name name, need to be consistent with the name parameter used for running compass_plot.py
1)wilcox_res.tsv: Comparison results between two groups using Wilcoxon rank sum test
2)Glycolysis_gluconeogenesis_update.pdf: Scatter plot displays the comparison results between two groups of subsystems
3)Glycolysis_gluconeogenesis_anno_update.pdf: Scatter plot displays the comparison results between two groups of subsystems, with metabolic process names attached
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent