Mercurial > repos > iuc > scater_plot_tsne

comparison scater-plot-dist-scatter.R @ 0:a30f4bfe8f01 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/scater commit 61f3899168453092fd25691cf31871a3a350fd3b"
author iuc
date Tue, 03 Sep 2019 14:30:21 -0400
parents
children 2b09ca1c5e41
comparison
equal deleted inserted replaced
-1:000000000000 0:a30f4bfe8f01
1 #!/usr/bin/env Rscript
2
3 # Plot the distribution of read counts and feature counts, side by side, then a scatter plot of read counts vs feature counts below
4
5 # Load optparse we need to check inputs
6
7 library(optparse)
8 library(workflowscriptscommon)
9 library(LoomExperiment)
10 library(scater)
11 library(ggpubr)
12 library(scales)
13
14 # parse options
15
16 option_list = list(
17 make_option(
18 c("-i", "--input-loom"),
19 action = "store",
20 default = NA,
21 type = 'character',
22 help = "A SingleCellExperiment object file in Loom format."
23 ),
24 make_option(
25 c("-o", "--output-plot-file"),
26 action = "store",
27 default = NA,
28 type = 'character',
29 help = "Path of the PDF output file to save plot to."
30 ),
31 make_option(
32 c("-l", "--log-scale"),
33 action="store_true",
34 default=FALSE,
35 type = 'logical',
36 help = "Plot on log scale (recommended for large datasets)."
37 )
38 )
39
40 opt <- wsc_parse_args(option_list, mandatory = c('input_loom', 'output_plot_file', 'log_scale'))
41
42 # Check parameter values
43
44 if ( ! file.exists(opt$input_loom)){
45 stop((paste('File', opt$input_loom, 'does not exist')))
46 }
47
48 # Input from Loom format
49
50 scle <- import(opt$input_loom, format='loom', type='SingleCellLoomExperiment')
51
52 #do the scatter plot of reads vs genes
53 total_counts <- scle$total_counts
54 total_features <- scle$total_features_by_counts
55 count_feats <- cbind(total_counts, total_features)
56 cf_dm <- as.data.frame(count_feats)
57
58 # Calculate binwidths for reads and features plots. Use 20 bins
59 read_bins <- max(total_counts / 1e6) / 20
60 feat_bins <- max(total_features) / 20
61
62 # Make the plots
63 plot <- ggplot(cf_dm, aes(x=total_counts / 1e6, y=total_features)) + geom_point(shape=1) + geom_smooth() + xlab("Read count (millions)") +
64 ylab("Feature count") + ggtitle("Scatterplot of reads vs features")
65 plot1 <- qplot(total_counts / 1e6, geom="histogram", binwidth = read_bins, ylab="Number of cells", xlab = "Read counts (millions)", fill=I("darkseagreen3")) + ggtitle("Read counts per cell")
66 plot2 <- qplot(total_features, geom="histogram", binwidth = feat_bins, ylab="Number of cells", xlab = "Feature counts", fill=I("darkseagreen3")) + ggtitle("Feature counts per cell")
67 plot3 <- plotColData(scle, y="pct_counts_MT", x="total_features_by_counts") + ggtitle("% MT genes") + geom_point(shape=1) + theme(text = element_text(size=15)) + theme(plot.title = element_text(size=15))
68
69 if (! opt$log_scale){
70 final_plot <- ggarrange(plot1, plot2, plot, plot3, ncol=2, nrow=2)
71 ggsave(opt$output_plot_file, final_plot, device="pdf")
72 } else {
73 plot_log_both <- plot + scale_x_continuous(trans = 'log10') + scale_y_continuous(trans = 'log10')
74 plot1_log <- plot1 + scale_y_continuous(trans = 'log10')
75 plot2_log <- plot2 + scale_y_continuous(trans = 'log10')
76 plot3_log <- plot3 + scale_y_log10(labels=number)
77 final_plot_log <- ggarrange(plot1_log, plot2_log, plot_log_both, plot3_log, ncol=2, nrow=2)
78 ggsave(opt$output_plot_file, final_plot_log, device="pdf")
79 }