--- title: 'Report breakdown by group' author: 'ampliCan' date: '`r format(Sys.time(), "%d %B %Y")`' output: html_document: toc: true theme: paper toc_float: true number_sections: true params: alignments: !r system.file('extdata', 'results', 'alignments', 'events_filtered_shifted_normalized.csv', package = 'amplican') config_summary: !r system.file('extdata', 'results', 'config_summary.csv', package = 'amplican') vignette: > %\VignetteIndexEntry{example group_report report} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r load data, message=F, warning=FALSE, include=FALSE} library(amplican) library(ggplot2) alignments <- data.table::fread(params$alignments) data.table::setDF(alignments) config <- data.frame(data.table::fread(params$config_summary)) height <- plot_height(length(unique(config$Group))) ``` *** # Description *** **Read distribution plot** - plot shows number of reads assigned during read grouping **Filtered Reads** - plot shows percentage of assigned reads that have been recognized as PRIMER DIMERS or filtered based on low alignment score **Cutting rates** - plot gives overview of percentage of reads (not filtered as PRIMER DIMER) that have cut **Frameshift** - plot shows what percentage of reads that have frameshift **Frameshift overlapping** - shows what percentage of reads have frameshift counting only deletions and insertions that overlap expected cut site (should be more accurate when controls are not available) **Read heterogeneity plot** - shows what is the share of each of the unique reads in total count of all reads. The more yellow each row, the less heterogeneity in the reads, more black means reads don't repeat often and are unique *** # Group Summary *** ## Read distribution ```{r plot_total_reads, echo=FALSE, fig.height=height, fig.width=14, message=F, warning=FALSE} ggplot(data = config, aes(x = as.factor(Group), y = log10(Reads + 1), order = Group, fill = Group)) + geom_boxplot() + ylab('Number of reads + 1, log10 scaled') + xlab('Group') + theme(legend.position = 'none', axis.text = element_text(size = 12), axis.title = element_text(size = 14, face = 'bold')) + coord_flip() ``` ## Filtered reads ```{r plot_F_per, echo=FALSE, fig.height=height, fig.width=14, message=F, warning=FALSE} config$F_percentage <- (config$PRIMER_DIMER + config$Low_Score) * 100/config$Reads config$F_percentage[is.nan(config$F_percentage)] <- 0 ggplot(data = config, aes(x = as.factor(Group), y = F_percentage, order = Group, fill = Group)) + geom_boxplot() + xlab('Group') + ylab('Percentage of filtered reads') + theme(axis.text = element_text(size=12), axis.title = element_text(size=14, face = 'bold'), legend.position = 'none') + ylim(0, 100) + coord_flip() ``` ## Indel rates ```{r plot indel percentage, echo=FALSE, fig.height=height, fig.width=14, message=F, warning=FALSE} config$indel_percentage <- config$Reads_Indel * 100/config$Reads_Filtered config$indel_percentage[is.nan(config$indel_percentage)] <- 0 ggplot(data = config, aes(x = as.factor(Group), y = indel_percentage, order = Group, fill = Group)) + geom_boxplot() + xlab('Group') + ylab('Percentage of reads (not filtered) that have indel') + theme(axis.text = element_text(size=12), axis.title = element_text(size=14, face = 'bold'), legend.position = 'None') + ylim(0,100) + coord_flip() ``` ## Frameshift ```{r plot_frameshift_per, echo=FALSE, fig.height=height, fig.width=14, message=F, warning=FALSE} config$frameshift_percentage <- config$Reads_Frameshifted * 100/config$Reads_Filtered config$frameshift_percentage[is.nan(config$frameshift_percentage)] <- 0 ggplot(data = config, aes(x = as.factor(Group), y = frameshift_percentage, order = Group, fill = Group)) + geom_boxplot() + xlab('Group') + ylab('Percentage of reads (not filtered) that have frameshift') + theme(axis.text = element_text(size=12), axis.title = element_text(size=14,face = 'bold'), legend.position = 'None') + ylim(0, 100) + coord_flip() ``` ## Heterogeneity of reads ```{r plot read heterogeneity, echo=FALSE, fig.height=height + 1, fig.width=14, message=F, warning=FALSE} plot_heterogeneity(alignments, config, level = 'Group') ``` *** # Alignments plots *** ```{r plot_alignments, results='asis', echo=F, message=F, warning=F} alignments <- alignments[alignments$consensus & alignments$overlaps, ] alignments$strand <- "+" # strand does not matter after consensus filtering src = sapply(unique(config$Group), function(i) { knitr::knit_expand(text = c( "## Group {{i}} \n", "### Deletions \n", paste('```{r del-{{i}}, echo = F, results = "asis", ', 'message=F, warning=F}', collapse = ''), 'amplican::metaplot_deletions(alignments, config, "Group", "{{i}}")', '```\n', "### Insertions", paste('```{r ins-{{i}}, echo = F, results = "asis", ', 'message=F, warning=F}', collapse = ''), 'amplican::metaplot_insertions(alignments, config, "Group", "{{i}}")', '```\n', "### Mismatches", paste('```{r mis-{{i}}, echo = F, results = "asis", ', 'message=F, warning=F}', collapse = ''), 'amplican::plot_mismatches(alignments, config, "Group", "{{i}}")', '```\n')) }) # knit the source res = knitr::knit_child(text = src, quiet = TRUE) cat(res, sep = '\n') ```