--- title: 'Report breakdown by ID' 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') cut_buffer: 5 xlab_spacing: 4 vignette: > %\VignetteIndexEntry{example id_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$ID))) ``` *** # 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 **Deletions plot** - shows summary of deletions detected after alignments with distinction for forward (top plot) and reverse (bottom) reads, blue dotted lines represent primers as black dotted line represents cut site box, for deletions overlapping with cut site box there is distinction in color **Mismatches plot** - shows summary of mismatches detected after alignments split by forward (top plot) and reverse (bottom) reads, mismatches are colored in the same manner as amplicon **Insertions plot** - shows summary of insertions detected after alignments split by forward (top plot) and reverse (bottom) reads, insertion is shown as right-angled triangle where size of the insertion corresponds to the width of the triangle, size and transparency of triangle reflect on the frequency of the insertion *** # ID 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(ID), y = log10(Reads + 1), order = ID)) + geom_bar(stat='identity') + ylab('Number of reads + 1, log10 scaled') + xlab('ID') + theme(legend.position = 'none', axis.text = element_text(size = 12), axis.title = element_text(size = 14, face = 'bold')) + coord_flip() + geom_text(aes(x = as.factor(ID), y = log10(Reads + 1), label = Reads), hjust = -1) ``` ## Filtered reads ```{r plot_F_per, echo=FALSE, fig.height=height, fig.width=14, message=F, warning=FALSE} config$PRIMER_DIMER <- config$PRIMER_DIMER * 100/config$Reads config$PRIMER_DIMER[is.nan(config$PRIMER_DIMER)] <- 0 config$Low_Score <- config$Low_Score * 100/config$Reads config$Low_Score[is.nan(config$Low_Score)] <- 0 config_melt <- data.table::melt(config, id.vars = "ID", measure.vars = c("PRIMER_DIMER", "Low_Score")) ggplot(data = config_melt, aes(x = as.factor(ID), y = value, fill = variable, order = ID)) + geom_bar(stat='identity') + ylab('Percentage of filtered reads') + xlab('ID') + theme(legend.position = 'top', axis.text = element_text(size = 12), axis.title = element_text(size = 14, face = 'bold')) + coord_flip() + labs(fill = "") ``` ## Indel rates ```{r plot mut percentage, echo=FALSE, fig.height=height, fig.width=14, message=F, warning=FALSE} config$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(ID), y = indel_percentage, order = ID)) + geom_bar(stat='identity') + ylab('Percentage of reads (not filtered) that have indel') + xlab('ID') + theme(legend.position = 'none', axis.text = element_text(size = 12), axis.title = element_text(size = 14, face = 'bold')) + coord_flip() + geom_text(aes(x = as.factor(ID), y = indel_percentage, label = Reads_Indel), hjust = -1) ``` ## 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$frameshift_percentage)] <- 0 ggplot(data = config, aes(x = as.factor(ID), y = frameshift_percentage, order = ID)) + geom_bar(stat='identity') + ylab('Percentage of reads (not filtered) that have frameshift') + xlab('ID') + theme(legend.position = 'none', axis.text = element_text(size = 12), axis.title = element_text(size = 14, face = 'bold')) + coord_flip() + geom_text(aes(x = as.factor(ID), y = frameshift_percentage, label = Reads_Frameshifted), hjust = -1) ``` ## Heterogeneity of reads ```{r plot read domination, echo=FALSE, fig.height=height + 1, fig.width=14, message=F, warning=FALSE} plot_heterogeneity(alignments, config) ``` *** # Alignments plots *** ```{r plot_alignments, results='asis', echo=F, message=F, warning=F} alignments_cons <- alignments[alignments$consensus & alignments$overlaps, ] src = sapply(config$ID, function(i) { knitr::knit_expand(text = c( "## {{i}} \n", "### Deletions \n", paste('```{r del-{{i}}, echo = F, results = "asis", ', 'fig.width=25, message=F, warning=F}', collapse = ''), paste('p <- amplican::plot_deletions(alignments, config, "{{i}}",', ' params$cut_buffer, params$xlab_spacing)', collapse = ''), '```\n', "### Insertions \n", paste('```{r ins-{{i}}, echo = F, results = "asis", ', 'fig.width=25, message=F, warning=F}', collapse = ''), paste('p <- amplican::plot_insertions(alignments, config, "{{i}}",', ' params$cut_buffer, params$xlab_spacing)', collapse = ''), '```\n', "### Mismatches \n", paste('```{r mis-{{i}}, echo = F, results = "asis", ', 'fig.width=25, message=F, warning=F}', collapse = ''), paste('p <- amplican::plot_mismatches(alignments, config, "{{i}}",', ' params$cut_buffer, params$xlab_spacing)', collapse = ''), '```\n', "### Variants \n", paste('```{r var-{{i}}, echo = F, message=F, results = "asis", ', 'message=F, warning=F}', collapse = ''), paste('p <- amplican::plot_variants(alignments_cons, config, "{{i}}", ', ' params$cut_buffer)', collapse = ''), '```\n')) }) # knit the source res = knitr::knit_child(text = src, quiet = TRUE) cat(res, sep = '\n') ```