parameters.rmd

---
title: Determine sensitivity to parameters of Tigmint
author: Shaun Jackman
output:
  html_document:
    keep_md: true
params:
  abyss_fac_tsv:
    label: "Input TSV file of parameters and abyss-fac metrics"
    value: "abyss-fac.tsv"
    input: text
  samtobreak_tsv:
    label: "Input TSV file of parameters and abyss-samtobreak metrics"
    value: "samtobreak.tsv"
    input: text
  output_tsv:
    label: "Output TSV file of parameters and assembly metrics"
    value: "parameters.out.tsv"
    input: text
---

```{r setup, message=FALSE}
library(dplyr)
library(ggplot2)
library(ggrepel)
library(knitr)
library(readr)
library(scales)
library(stringr)
library(tidyr)

knit_print.data.frame <- function(x, ...) kable(x) %>% paste(collapse = "\n") %>% asis_output
abyss_fac_tsv <- params$abyss_fac_tsv
samtobreak_tsv <- params$samtobreak_tsv
output_tsv <- params$output_tsv

abyss_fac_tsv
samtobreak_tsv
output_tsv
```

# Read the data

```{r read-data}
abyss_fac <- read_tsv(abyss_fac_tsv)
samtobreak <- read_tsv(samtobreak_tsv)

metrics <- left_join(abyss_fac, samtobreak, by = c("Depth", "Starts")) %>%
	mutate(Label = ifelse(is.na(Depth), "ABySS + ARCS",
		paste0("d=", Depth, " s=", Starts)))
```

# Plot parameters and assembly metrics
```{r parameters}
ggplot(metrics) +
	aes(x = Total_breakpoints, y = Scaffold_NGA50, label = Label) +
	geom_point() +
	geom_text_repel(nudge_x = 20, nudge_y = 100e3) +
	scale_x_continuous(name = "Breakpoints", labels = comma) +
	scale_y_continuous(name = "Scaffold NGA50 (Mbp)", labels = unit_format(unit = "Mbp", scale = 1e-6)) +
	expand_limits(x = c(4400, 4900), y = c(0, 10e6)) +
	theme_minimal(base_size = 20)
```

# Table of parameters and assembly metrics
```{r parameters-table}
parameters_table <- metrics %>%
	arrange(Scaffold_NGA50) %>%
	transmute(
		Depth,
		Starts,
		`NG50 (Mbp)` = round(NG50 / 1e6, 2),
		`NGA50 (Mbp)` = round(Scaffold_NGA50 / 1e6, 2),
		Breakpoints = comma(Total_breakpoints),
		Reduction = max(Total_breakpoints) - Total_breakpoints,
		Reduction = str_c(Reduction, " (", percent(Reduction / max(Total_breakpoints)), ")"),
		Reduction = ifelse(is.na(Depth), NA, Reduction)) %>%
	replace_na(list(Depth = "Original", Starts = "Original"))

parameters_table
write_tsv(parameters_table, output_tsv)
```