lectures/lecture12/lecture12.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lecture 12: Working with tabular data using `R` and `tidyverse`\n",
    "\n",
    "[Rasi Subramaniam](http://rasilab.org/)\n",
    "\n",
    "To interactively work with the code below, open [lecture12.ipynb](./lecture12.ipynb) in VSCode. Make sure to select the `kernel` for `R` so that you can execute `R` code. You should have already set this up following the software installation instructions [here](../../software/README.md)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`R` is the second programming language after `Python` that we will learn in this course. We will use `R` over the next 5 lectures.\n",
    "\n",
    "`R` is particularly well suited for reading, manipulating, and visualizing data in tabular and biological sequence formats. \n",
    "Many statistical tests are also available out of the box in `R`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While \"base\" `R` is used widely, I almost exclusively use `R` for its two excellent package collections:\n",
    "\n",
    "1. [Tidyverse](https://www.tidyverse.org/) - suited for tabular data\n",
    "2. [Bioconductor](https://www.bioconductor.org/) - suited for biology-aware analyses"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Today we will learn a few basic functions from `tidyverse` for working with tabular data.\n",
    "\n",
    "Unlike `pandas` which is a single package with lot of functionality, `tidyverse` is a collection of packages that are focused on specific tasks.\n",
    "\n",
    "1. [ggplot2](https://ggplot2.tidyverse.org/) - for plotting data\n",
    "2. [dplyr](https://dplyr.tidyverse.org/) - for filtering, aggregating, and transforming data\n",
    "3. [readr](https://readr.tidyverse.org/) - for reading and writing data\n",
    "4. [tidyr](https://tidyr.tidyverse.org/) - for cleaning and transforming data\n",
    "5. [stringr](https://stringr.tidyverse.org/) - for manipulating strings\n",
    "6. [purrr](https://purrr.tidyverse.org/) - for manipulating lists of R objects\n",
    "7. [forcats](https://forcats.tidyverse.org/) - for manipulating categorical data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can load all the above packages in one go:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "── \u001b[1mAttaching packages\u001b[22m ─────────────────────────────────────── tidyverse 1.3.0 ──\n",
      "\n",
      "\u001b[32m✔\u001b[39m \u001b[34mggplot2\u001b[39m 3.3.5     \u001b[32m✔\u001b[39m \u001b[34mpurrr  \u001b[39m 0.3.4\n",
      "\u001b[32m✔\u001b[39m \u001b[34mtibble \u001b[39m 3.0.4     \u001b[32m✔\u001b[39m \u001b[34mdplyr  \u001b[39m 1.0.2\n",
      "\u001b[32m✔\u001b[39m \u001b[34mtidyr  \u001b[39m 1.1.2     \u001b[32m✔\u001b[39m \u001b[34mstringr\u001b[39m 1.4.0\n",
      "\u001b[32m✔\u001b[39m \u001b[34mreadr  \u001b[39m 1.4.0     \u001b[32m✔\u001b[39m \u001b[34mforcats\u001b[39m 0.5.0\n",
      "\n",
      "── \u001b[1mConflicts\u001b[22m ────────────────────────────────────────── tidyverse_conflicts() ──\n",
      "\u001b[31m✖\u001b[39m \u001b[34mdplyr\u001b[39m::\u001b[32mfilter()\u001b[39m masks \u001b[34mstats\u001b[39m::filter()\n",
      "\u001b[31m✖\u001b[39m \u001b[34mdplyr\u001b[39m::\u001b[32mlag()\u001b[39m    masks \u001b[34mstats\u001b[39m::lag()\n",
      "\n"
     ]
    }
   ],
   "source": [
    "library(tidyverse)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Reading data\n",
    "\n",
    "Various options for reading and writing data are in package [`readr`](https://readr.tidyverse.org/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_1.tsv\")\n",
    "\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The tabular data structure is called a `tibble` in `tidyverse`, and is a souped-up version of the [data.frame](https://en.wikipedia.org/wiki/Data_frame) R data structure with additional nice features.\n",
    "\n",
    "The ` <- ` assignment operator is equivalent to the ` = ` assignment operator and can be used interchangeably. However, using the ` <- ` operator is more conventional."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plotting data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "ggplot(data, aes(x = kozak_region, y = mean_ratio)) +\n",
    "  geom_point()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Anatomy of a `ggplot2` plot**\n",
    "\n",
    " - Begins with `ggplot` function with a `tibble` argument as the first argument.\n",
    " - `aes` specifies the variables to plot.\n",
    " - `geom` specifies the type of plot.\n",
    " - `+` adds additional `layers` to the plot.\n",
    "\n",
    "\n",
    "**Key differences with `Python`**\n",
    "  - No need to specify variables within quotes.\n",
    "  - Indentation convention is different."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Change size of plot globally"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "options(repr.plot.width = 5, repr.plot.height = 3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## How do we show multiple experimental variables?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting a point graph with color"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "ggplot(data, aes(x = kozak_region,\n",
    "                 y = mean_ratio,\n",
    "                 color = insert_sequence)) +\n",
    "  geom_point()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting a line graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "ggplot(data, aes(x = kozak_region,\n",
    "                 y = mean_ratio,\n",
    "                 color = insert_sequence,\n",
    "                 group = insert_sequence)) +\n",
    "  geom_line()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plotting point and line graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "ggplot(data, aes(x = kozak_region,\n",
    "                 y = mean_ratio,\n",
    "                 color = insert_sequence,\n",
    "                 group = insert_sequence)) +\n",
    "  geom_line() +\n",
    "  geom_point()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ‘Faceting’ – Plotting in multiple panels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "options(repr.plot.width = 6, repr.plot.height = 3)\n",
    "\n",
    "ggplot(data, aes(x = kozak_region,\n",
    "                 y = mean_ratio,\n",
    "                 group = insert_sequence)) +\n",
    "  geom_line() +\n",
    "  geom_point() +\n",
    "  facet_grid(~ insert_sequence)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# In-class `ggplot2` exercises\n",
    "\n",
    "(20 min)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Make X, Y, legend labels into nice strings\n",
    "\n",
    "See https://ggplot2.tidyverse.org/reference/labs.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "ggplot(data, aes(x = kozak_region,\n",
    "                 y = mean_ratio,\n",
    "                 color = insert_sequence,\n",
    "                 group = insert_sequence)) +\n",
    "  geom_line() +\n",
    "  geom_point()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Add title to the above plot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Change the plot look to a `classic` theme\n",
    "\n",
    "See https://ggplot2.tidyverse.org/reference/ggtheme.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Change Y axis to log scale\n",
    "\n",
    "See https://ggplot2.tidyverse.org/reference/scale_continuous.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Change Y scale to go linearly from 0 to 5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Transforming data\n",
    "\n",
    "Uses functions from the [`dplyr`](https://dplyr.tidyverse.org/) package."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_1.tsv\")\n",
    "\n",
    "\n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Select specific columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "select(data, strain, mean_ratio, insert_sequence, kozak_region)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Combine operations using the ` %>% ` operator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_1.tsv\") %>% \n",
    "  select(strain, mean_ratio, insert_sequence, kozak_region)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above is the same as the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_1.tsv\") %>% \n",
    "  select(., strain, mean_ratio, insert_sequence, kozak_region)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `%>%` operator lets you `chain` different data analysis tasks together and makes the analysis logic easier to understand.\n",
    "\n",
    "Side note: You can create keyboard shortcuts for ` <- ` and ` %>% ` in VSCode as explained [here](https://code.visualstudio.com/docs/getstarted/keybindings). \n",
    "\n",
    "I use `Alt + -` for ` <- ` and `Alt + Shift + m` for ` %>% ` following `RStudio` convention."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can get a view of the transformed data using `print()` as the last step in a chain of commands"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_1.tsv\") %>% \n",
    "  select(strain, mean_ratio, insert_sequence, kozak_region) %>% \n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Filter rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_1.tsv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  filter(kozak_region == \"A\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>%\n",
    "  filter(kozak_region == \"A\", insert_sequence == \"10×AGA\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>%\n",
    "  filter(kozak_region == \"A\") %>% \n",
    "  filter(insert_sequence == \"10×AGA\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Arrange (sort) rows in a specific order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>%\n",
    "  arrange(mean_ratio)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create new columns using `mutate`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_2.tsv\") %>% \n",
    "  print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- data %>%\n",
    "  mutate(mean_ratio = mean_yfp / mean_rfp) %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Use mutate to modify existing columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>%\n",
    "  mutate(mean_ratio = round(mean_ratio, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Combine tables using `join` functions\n",
    "\n",
    "Variants: `inner_join`, `left_join`, `right_join`, `full_join`\n",
    "\n",
    "See [https://dplyr.tidyverse.org/reference/mutate-joins.html](https://dplyr.tidyverse.org/reference/mutate-joins.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "annotations <- read_tsv(\"data/example_dataset_3.tsv\")\n",
    "\n",
    "annotations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  inner_join(annotations, by = \"strain\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  left_join(annotations, by = \"strain\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  right_join(annotations, by = \"strain\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## You can combine `dplyr` and `ggplot2` functions\n",
    "\n",
    "But remember to use ` %>% ` in `dplyr` vs ` +` in `ggplot2`!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  mutate(mean_ratio = mean_yfp / mean_rfp) %>%\n",
    "  left_join(annotations, by = \"strain\") %>%\n",
    "  ggplot(aes(x = kozak_region, y = mean_ratio, \n",
    "             color = insert_sequence, group = insert_sequence)) +\n",
    "  geom_line() +\n",
    "  geom_point()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use `stringr` functions to manipulate string columns\n",
    "\n",
    "All functions are named nicely and begin with `str_`. I find them easier to use than the equivalent `Python` regular expression functions.\n",
    "\n",
    "See [https://stringr.tidyverse.org/reference/index.html](https://stringr.tidyverse.org/reference/index.html)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  mutate(mean_ratio = mean_yfp / mean_rfp) %>%\n",
    "  left_join(annotations, by = \"strain\") %>% \n",
    "  mutate(codon = str_extract(insert_sequence, \"[A-Z]{3}$\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# In-class data transformation exercises\n",
    "\n",
    "(20 min)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Create log2-transformed YFP/RFP ratio as a new column\n",
    "\n",
    "Google for `log2 R` to find the appropriate function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_2.tsv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Extract strain number from the `strain` column into a new column and sort numerically by strain number\n",
    "\n",
    "Extract the strain number using a `stringr` function.\n",
    "\n",
    "Google for `character to integer R` to find appropriate function to use in `mutate`.\n",
    "\n",
    "Then sort."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "annotations <- read_tsv(\"data/example_dataset_3.tsv\")\n",
    "\n",
    "annotations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Plot with the X axis as `kozak_region` but sorted by strain number\n",
    "\n",
    "This requires a bit more reading and discussion, but it is a good example of how to learn new `tidyverse` functions on your own!\n",
    "\n",
    "Use [fct_reorder](https://forcats.tidyverse.org/reference/fct_reorder.html) function from the [`forcats`](https://forcats.tidyverse.org/reference/index.html) package to sort `kozak_region` by strain number you created above in a `mutate` step and then feed it into `ggplot`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>% \n",
    "  mutate(mean_ratio = mean_yfp / mean_rfp) %>%\n",
    "  left_join(annotations, by = \"strain\") %>%\n",
    "  ggplot(aes(x = kozak_region, y = mean_ratio, \n",
    "             color = insert_sequence, group = insert_sequence)) +\n",
    "  geom_line() +\n",
    "  geom_point()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Grouping, summarizing, and reshaping data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use `summarize` to calculate stats across rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_2.tsv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[90m# A tibble: 1 x 2\u001b[39m\n",
      "  max_yfp max_rfp\n",
      "    \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m   \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m\n",
      "\u001b[90m1\u001b[39m    \u001b[4m9\u001b[24m365   \u001b[4m2\u001b[24m\u001b[4m3\u001b[24m866\n"
     ]
    }
   ],
   "source": [
    "data %>%\n",
    "  summarize(max_yfp = max(mean_yfp),\n",
    "            max_rfp = max(mean_rfp)) %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Other examples of summary functions:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- min()\t\n",
    "- mean()\n",
    "- sd()\t\n",
    "- dplyr::first()\t\n",
    "- dplyr::n()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use `group_by` to group rows by one or more columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- read_tsv(\"data/example_dataset_4.tsv\") %>% \n",
    "  print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[90m# A tibble: 74 x 4\u001b[39m\n",
      "\u001b[90m# Groups:   strain [16]\u001b[39m\n",
      "   strain    yfp   rfp replicate\n",
      "   \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m   \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m     \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m\n",
      "\u001b[90m 1\u001b[39m schp677  \u001b[4m4\u001b[24m123 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m661         1\n",
      "\u001b[90m 2\u001b[39m schp678  \u001b[4m4\u001b[24m550 \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m437         1\n",
      "\u001b[90m 3\u001b[39m schp675  \u001b[4m3\u001b[24m880 \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m323         1\n",
      "\u001b[90m 4\u001b[39m schp676  \u001b[4m2\u001b[24m863 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m668         1\n",
      "\u001b[90m 5\u001b[39m schp687  \u001b[4m4\u001b[24m767 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m995         1\n",
      "\u001b[90m 6\u001b[39m schp688  \u001b[4m1\u001b[24m274 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m927         1\n",
      "\u001b[90m 7\u001b[39m schp679  \u001b[4m2\u001b[24m605 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m840         1\n",
      "\u001b[90m 8\u001b[39m schp680  \u001b[4m1\u001b[24m175 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m902         1\n",
      "\u001b[90m 9\u001b[39m schp681  \u001b[4m3\u001b[24m861 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m659         1\n",
      "\u001b[90m10\u001b[39m schp683  \u001b[4m9\u001b[24m949 \u001b[4m2\u001b[24m\u001b[4m5\u001b[24m406         1\n",
      "\u001b[90m# … with 64 more rows\u001b[39m\n"
     ]
    }
   ],
   "source": [
    "data %>% \n",
    "  group_by(strain) %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use `group_by` + `summarize` for statistics by group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`summarise()` ungrouping output (override with `.groups` argument)\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[90m# A tibble: 16 x 3\u001b[39m\n",
      "   strain  mean_yfp mean_rfp\n",
      "   \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m      \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m    \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m\n",
      "\u001b[90m 1\u001b[39m schp674    \u001b[4m1\u001b[24m270    \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m316 \n",
      "\u001b[90m 2\u001b[39m schp675    \u001b[4m3\u001b[24m687.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m438.\n",
      "\u001b[90m 3\u001b[39m schp676    \u001b[4m2\u001b[24m656.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m223.\n",
      "\u001b[90m 4\u001b[39m schp677    \u001b[4m3\u001b[24m967.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m604 \n",
      "\u001b[90m 5\u001b[39m schp678    \u001b[4m4\u001b[24m378.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m630.\n",
      "\u001b[90m 6\u001b[39m schp679    \u001b[4m2\u001b[24m528    \u001b[4m1\u001b[24m\u001b[4m9\u001b[24m906 \n",
      "\u001b[90m 7\u001b[39m schp680    \u001b[4m1\u001b[24m117.   \u001b[4m1\u001b[24m\u001b[4m9\u001b[24m377.\n",
      "\u001b[90m 8\u001b[39m schp681    \u001b[4m3\u001b[24m705    \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m227 \n",
      "\u001b[90m 9\u001b[39m schp683    \u001b[4m9\u001b[24m364.   \u001b[4m2\u001b[24m\u001b[4m3\u001b[24m866.\n",
      "\u001b[90m10\u001b[39m schp684    \u001b[4m3\u001b[24m294.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m585.\n",
      "\u001b[90m11\u001b[39m schp685    \u001b[4m7\u001b[24m379    \u001b[4m2\u001b[24m\u001b[4m2\u001b[24m956 \n",
      "\u001b[90m12\u001b[39m schp686    \u001b[4m5\u001b[24m000.   \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m171.\n",
      "\u001b[90m13\u001b[39m schp687    \u001b[4m4\u001b[24m658.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m860.\n",
      "\u001b[90m14\u001b[39m schp688    \u001b[4m1\u001b[24m748.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m755.\n",
      "\u001b[90m15\u001b[39m schp689    \u001b[4m8\u001b[24m693.   \u001b[4m2\u001b[24m\u001b[4m2\u001b[24m650.\n",
      "\u001b[90m16\u001b[39m schp690    \u001b[4m3\u001b[24m535.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m594.\n"
     ]
    }
   ],
   "source": [
    "data %>% \n",
    "  group_by(strain) %>%\n",
    "  summarize(mean_yfp = mean(yfp), mean_rfp = mean(rfp)) %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`summarise()` ungrouping output (override with `.groups` argument)\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[90m# A tibble: 16 x 5\u001b[39m\n",
      "   strain  mean_yfp mean_rfp se_yfp se_rfp\n",
      "   \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m      \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m    \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m  \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m  \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m\n",
      "\u001b[90m 1\u001b[39m schp674    \u001b[4m1\u001b[24m270    \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m316    54     717 \n",
      "\u001b[90m 2\u001b[39m schp675    \u001b[4m3\u001b[24m687.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m438.   84.6   483.\n",
      "\u001b[90m 3\u001b[39m schp676    \u001b[4m2\u001b[24m656.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m223.  137.    380.\n",
      "\u001b[90m 4\u001b[39m schp677    \u001b[4m3\u001b[24m967.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m604   107.    423.\n",
      "\u001b[90m 5\u001b[39m schp678    \u001b[4m4\u001b[24m378.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m630.  111.    575.\n",
      "\u001b[90m 6\u001b[39m schp679    \u001b[4m2\u001b[24m528    \u001b[4m1\u001b[24m\u001b[4m9\u001b[24m906    33.9  \u001b[4m1\u001b[24m034.\n",
      "\u001b[90m 7\u001b[39m schp680    \u001b[4m1\u001b[24m117.   \u001b[4m1\u001b[24m\u001b[4m9\u001b[24m377.   27.7   700.\n",
      "\u001b[90m 8\u001b[39m schp681    \u001b[4m3\u001b[24m705    \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m227    90.8   469.\n",
      "\u001b[90m 9\u001b[39m schp683    \u001b[4m9\u001b[24m364.   \u001b[4m2\u001b[24m\u001b[4m3\u001b[24m866.  352.    515.\n",
      "\u001b[90m10\u001b[39m schp684    \u001b[4m3\u001b[24m294.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m585.   49.6   318.\n",
      "\u001b[90m11\u001b[39m schp685    \u001b[4m7\u001b[24m379    \u001b[4m2\u001b[24m\u001b[4m2\u001b[24m956   194.    973.\n",
      "\u001b[90m12\u001b[39m schp686    \u001b[4m5\u001b[24m000.   \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m171.   81.5   307.\n",
      "\u001b[90m13\u001b[39m schp687    \u001b[4m4\u001b[24m658.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m860.   80.9   199.\n",
      "\u001b[90m14\u001b[39m schp688    \u001b[4m1\u001b[24m748.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m755.  160.    203.\n",
      "\u001b[90m15\u001b[39m schp689    \u001b[4m8\u001b[24m693.   \u001b[4m2\u001b[24m\u001b[4m2\u001b[24m650.  667.   \u001b[4m1\u001b[24m045.\n",
      "\u001b[90m16\u001b[39m schp690    \u001b[4m3\u001b[24m535.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m594.   31.0   173.\n"
     ]
    }
   ],
   "source": [
    "data %>% \n",
    "  group_by(strain) %>%\n",
    "  summarize(mean_yfp = mean(yfp), mean_rfp = mean(rfp),\n",
    "            se_yfp = sd(yfp) / sqrt(n()), \n",
    "            se_rfp = sd(rfp) / sqrt(n())) %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `%>%` enables complex yet easily understandable data analysis pipelines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "annotations <- read_tsv(\"data/example_dataset_3.tsv\")\n",
    "\n",
    "annotations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`summarise()` ungrouping output (override with `.groups` argument)\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[90m# A tibble: 16 x 6\u001b[39m\n",
      "   strain  mean_yfp mean_rfp mean_ratio insert_sequence kozak_region\n",
      "   \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m      \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m    \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m      \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m           \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m       \n",
      "\u001b[90m 1\u001b[39m schp674    \u001b[4m1\u001b[24m270    \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m316      0.062\u001b[4m5\u001b[24m 10×AAG          G           \n",
      "\u001b[90m 2\u001b[39m schp675    \u001b[4m3\u001b[24m687.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m438.     0.180  10×AAG          B           \n",
      "\u001b[90m 3\u001b[39m schp676    \u001b[4m2\u001b[24m656.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m223.     0.131  10×AAG          F           \n",
      "\u001b[90m 4\u001b[39m schp677    \u001b[4m3\u001b[24m967.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m604      0.193  10×AAG          E           \n",
      "\u001b[90m 5\u001b[39m schp678    \u001b[4m4\u001b[24m378.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m630.     0.212  10×AAG          D           \n",
      "\u001b[90m 6\u001b[39m schp679    \u001b[4m2\u001b[24m528    \u001b[4m1\u001b[24m\u001b[4m9\u001b[24m906      0.127  10×AAG          A           \n",
      "\u001b[90m 7\u001b[39m schp680    \u001b[4m1\u001b[24m117.   \u001b[4m1\u001b[24m\u001b[4m9\u001b[24m377.     0.057\u001b[4m7\u001b[24m 10×AAG          H           \n",
      "\u001b[90m 8\u001b[39m schp681    \u001b[4m3\u001b[24m705    \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m227      0.183  10×AAG          C           \n",
      "\u001b[90m 9\u001b[39m schp683    \u001b[4m9\u001b[24m364.   \u001b[4m2\u001b[24m\u001b[4m3\u001b[24m866.     0.392  10×AGA          G           \n",
      "\u001b[90m10\u001b[39m schp684    \u001b[4m3\u001b[24m294.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m585.     0.160  10×AGA          B           \n",
      "\u001b[90m11\u001b[39m schp685    \u001b[4m7\u001b[24m379    \u001b[4m2\u001b[24m\u001b[4m2\u001b[24m956      0.321  10×AGA          F           \n",
      "\u001b[90m12\u001b[39m schp686    \u001b[4m5\u001b[24m000.   \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m171.     0.236  10×AGA          E           \n",
      "\u001b[90m13\u001b[39m schp687    \u001b[4m4\u001b[24m658.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m860.     0.223  10×AGA          D           \n",
      "\u001b[90m14\u001b[39m schp688    \u001b[4m1\u001b[24m748.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m755.     0.084\u001b[4m2\u001b[24m 10×AGA          A           \n",
      "\u001b[90m15\u001b[39m schp689    \u001b[4m8\u001b[24m693.   \u001b[4m2\u001b[24m\u001b[4m2\u001b[24m650.     0.384  10×AGA          H           \n",
      "\u001b[90m16\u001b[39m schp690    \u001b[4m3\u001b[24m535.   \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m594.     0.172  10×AGA          C           \n"
     ]
    }
   ],
   "source": [
    "data %>% \n",
    "  group_by(strain) %>%\n",
    "  summarize(mean_yfp = mean(yfp), mean_rfp = mean(rfp)) %>%\n",
    "  mutate(mean_ratio = mean_yfp / mean_rfp) %>%\n",
    "  left_join(annotations, by = \"strain\") %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## %>% and + if you want to plot\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`summarise()` ungrouping output (override with `.groups` argument)\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlgAAAFoCAIAAAAElhK7AAAACXBIWXMAABJ0AAASdAHeZh94AAAgAElEQVR4nOzdd1wT9/8H8M9dNglb9pIlhKVIq7hHtdaBVcCBilacHa5q22/tsNrWtv7aOlsVV1VcVcBBa7Uu3BVRkL1BAdkrO5e7+/0RSxERwwhhvJ+P/pF8buSdPpAXd/cZGE3TCAAAAOipcF0XAAAAAOgSBCEAAIAeDYIQAABAjwZBCAAAoEeDIAQAANCjQRACAADo0SAIAQAA9GgQhAAAAHq0rhSEH374obOzc0FBga4LAQAA0H10pSAsLy/Pzc0lCELXhQAAAOg+tBuESqVyx44ds2bNCgkJ2bZtm1QqbWbnxMTEt99+u7CwUKslAQAAAA1pNwh37dr14MGDlStXfvzxxxkZGVu3bn3ZnlKpdOvWrTDxKQAAgA6mxSCsrq6+cuXKokWLBgwY4OvrGxYWdvfu3aqqqiZ3Dg8Pt7Ky0l4xAAAAQJO0GITJyck0Tfv6+qrfenl5YRiWmJj44p7//PNPcnLy/PnztVcMAAAA0CSm9k5dUlJiZGTE5XLVbzkcjkAgqK6ubrRbXV3dr7/+unr16vo9GyouLr579676dXV1NYvF0l7BAAAAeiAtBqFMJmOz2Q1buFyuTCZrtNuvv/46aNAgHx+fJrvJZGRkbNy4seEZtFEqAACAHkuLQcjn8xUKRcMWgiD09fUbtsTGxubk5Gzfvv1lJ3Fzc1u7dq369f79+x89eqSNUgEAAPRYWgxCExOTuro6pVKpvi4kCKK2ttbc3LzhPg8fPiwvL589ezZCSN1ldPny5YMHD16zZo16B2tr68DAQPXr6OhoGEQIQFenpOgygrBms3FM16UAgBDSahCqu8k8evTotddeQwglJiZiGObj49Nwn3nz5gUHB6tfV1ZWfvHFF5999pmjo6P2qgIA6EoFQazOyT9aWq6iaQGDsdrO+jMHOxYGeQh0TItBaGRkNGTIkN27dyOEFApFeHj4uHHj9PT0EELnzp1DCAUEBBgbGxsbG6v3V184WlhYmJiYaK8qAIBOUDQKSc28VF2jfismyfX5TxQU/Z2Tg24LA0C7A+o/+OADLy+vH3/88ddffx08ePDChQvV7Xfu3Llz545WPxoA0KlcqampT8F6Pz4pqiJUOqkHgHpYF5rMJTQ0NCIiIisry8XFRde1AABaZkfR02VZuS+2TzI1HmRg4MLjuvC4LjyeAZPR8bWBHk6Lt0YBAKCeCbPp3zYxldUxlf8NLzZns1x4XFceT52L6oA0esmxALQL+PECAHSEt0yMDRnMWvK5G6H+BvpHhH1y5fJcmTxFKk2VyHLl8n/qxLdrRQ13M2YynXhcJy7Xicd14nKceFxPvp7V88OUAWg1CEIAQEfIl8sViMIwVP80xpXHOyLs48TjOvG4yPi/PZUUXahQqNMxV65IkUhTpdIEsSReJG54whfTUf26FbXVqFRPlYQzlwtDOnomCEIAgNYVK5RvJ6cpSGq3mwuOUKFCKdTjTell2mTwsHHsxXQkaPqJ/L90VF9BJrY5HQsVyvcyc85VViGEODi+ytZ6g6M9jOjoaSAIAQDaJSLJCUmphQrld04Oi6wsWncSFtb6dDRiMp1fSEdHHldF04HJaXH/7qygqO8fFyKEYERHTwNBCADQIpKmQ1IzEsWSMCuL/9nbtu/JX5aO+XJ5tkyeJZVny2TZMnmWTPbohXQ0ZDJ6sVg5Mnmjc/5cWPSpvS10Xu1RIAgBAFq0Kjvvj8rq4UYGv7o6dcwnsjDMlcdz5fHGN5iZQ0XTj5+/dsyVyx+JJY0PppGSpPPk8r4CfsdUCzoDCEIAgLbsKHq6veipmx7vtJeQg2t3+o7mMZu6dtz3tHRhRvZz+2EIIWTOhuXeehZd/mgCALqxv6qqV2XnmbKY57yFxp1yIODbvUzMXljidLyJMQzM6GkgCAEA7S9VIp2ZmoFj6KSnuyuPp+tymtaLxTrq0effLKQRQv31BfvdYeKqHqcz/pkGAOjSSpTK8UmpdSryoNB1lJGhrstpzhhjo8yB/S9U1RQplB583pvGxjCSsAeCIAQAtCcZRU1NTn8sV6zrbRdqYf7qA3TNiMmcYd5L11UAXYJbowCAdkMjFJaedbdONM2s17re9rouBwCNQBACANrN/3Lzj5dVDDE0OCR0hVuMoKuAIAQAtI8DJWWbHhc5crlRnu5cnQ6WAKBF4IcVANAOrtfULc3MNmAyznoLYRwe6FogCAEAbZUulU1JTqNpFOnp7sXX03U5ALQM9BoFALRJJaGanJRWrVLt7OM8xthI1+UA0GJwRQgAaD0lRQenpGfJZB/b2yy1ttR1OQC0BgQhAKCVaIQWZmRfq6mdaGq80RGWLgJdFQQhAKCV1uc/Plxa1l9fcMLDjQGL2YIuC4IQANAav5dVbMh/Ys1hn/ES8hmweh/owiAIAQAtdqu2bm56loDB+NPbw5YDazWArg16jQIAWiZPLg9MSVfR9O8ebrCALegGIAgBAC1QpyInJ6WVKYltrk6Te5m8+gAAOj24NQoA0BRB00Ep6ckS6SIri2U2VrouB4D2AUEIANDU8qzcS9U1b5kY/9rHWde1ANBuIAgBABr54XHhruISD77ecQ83JgyWAN0IBCEA4NWiyivX5hZYstnnvT0MmTBYAnQrEIQAgFeIF4nnpmdxcDzay92ey9F1OQC0M+g1CgBoTpFC+XZympQkj3q4+Rvo67ocANofBCEA4KVEJDkxKbVIofzBqfdM8166LgcArYBbowCAppE0PTs1M1EsmW9p/rG9ja7LAUBbIAgBAE1bmZ13rrJquJHBrj4uuq4FAC2CIAQANGFb4dMdRU/d9XinvYRsHAZLgO4MghAA0Nj5qurVOXmmLOZZb6ExE3oSgG4OghAA8JwUiTQkNQPH0ClPd1ceT9flAKB18LceAOA/T5XKCUmpdSrykLDPSCNDXZcDQEeAK0IAwDMyipqanP5YrljX236OhZmuywGgg0AQAgAQQoii0ezUzH/qRNPNe33Z207X5QDQcSAIAQAIIfS/3PzoisohhgaH3F2hkyjoUSAIAQBo/9PS/3tS5MjlRnm6c3D4tQB6FviJB6Cni62pfTcrx4DJOOstNGezdF0OAB0NghCAHi1dKpuanE7TKNLT3Yuvp+tyANABGD4BQM9VSagCklKrVaqdfZzHGBvpuhwAdAOuCAHooeQUNTk5NVsm/5+97VJrS12XA4DOQBAC0BPRCC3KyL5dKwo0M/3W0UHX5QCgSy24NVpdXX316tWioiI2m+3h4TF48GAGg6G9ygAA2vNV/uOI0nI/fcEhd1eYUhv0cJoG4a5du9asWSORSOpb7OzsduzYMXnyZO0UBgDQlhNlFV/nP7HhsM94Cfnw5yzo8TS6Nfrnn3++++677u7u4eHhV69ejYmJ+frrr1ks1tSpUy9fvqztEgEA7ehmbd289CwBg/GHt4cNh63rcgDQPYym6VfuNGzYMJFIFBcXx2L9N8ZIJpP5+/v36tWrw7IwNDQ0IiIiKyvLxQWWCQWgNfLkcv8HjyoJVbSXe4Cpia7LAaBT0OiKMD4+fvr06Q1TECHE4/FCQkLu37+vncIAAO2sVkUGJKWVKYktLo6QggDU0ygIGQxGkxeOMplMqVS2d0kAgPZH0HRQSlqKRLrC1voDGytdlwNAJ6JRZ5mBAwcePXp05cqVfD6/vlEikURGRnp6emqtNgBAm6RKpL+VlBUqlC48bp5ccbm6dryJ8Y/OvXVdFwCdi0ZB+OWXX77xxhs+Pj7vvvuur68vl8tNSUnZvn17SkrK8ePHtV0iAKAVjpaWz8/IUlL/3ctx5HKPebgxMRgtAcBzNArC4cOHR0VFLVmy5KOPPqpv1NPT27Zt24wZM7RWGwCglUqVxNLMnIYpiBBSUJQeA+bQAKAxTccRBgQEjBs37vr162lpaQRB2Nvbv/HGG8bGxlotDgDQOtdqakUk2aixWKl8KJIMMBDopCQAOq0WzCzDZrPHjBkzZswY7VUDAGgXMop6SXvjdAQANBeE/fr1c3R0jI6O7tevXzO7JSQktHdVAIA28dNv4rKPg+M+Av6L7QD0cPDAAIBuyJuvN9608ZOLbx3tjZmw8hoAjTX3r6L+Ug+u+QDoWooVynt1IiaG2XHYpQThxuOtsrOeY2Gu67oA6Iw0uiKcOXPmP//882L7X3/9tXTp0vYuCQDQJjRCCzOyKwnVD069c/1fkwwb9OC1fqEW5jBsAoAmNXdFWFNTU1NTgxA6ceLEiBEjLCwsGm6lafr48ePHjx/ftWuXdmsEALTE5ifF56uqxxgbrbS11nUtAHQBzQXhli1b1q9fr3793nvvNbmPv79/+xcFAGitFIn0s7wCYyZzv7sLLDQIgCaaC8KRI0eqX6xfv37atGkeHh4Nt2IYZm5uPm3aNO0VBwBoEQVFzUrLlFPUQXdXOw5H1+UA0DW8IgjVWRgTE7N06dLRo0d3UFEAgFb5KCf/kViywMpiunkvXdcCQJehUWeZ+/fvN5mCsbGxML4egE7iYlXNjqKnzjzuZhdHXdcCQFei0aAimqY3b9586dKlRosu5efnl5WVaacwAEALlBPEO+lZDAyLEPbRZzB0XQ4AXYlGQbhz587Vq1ebmZkplcq6ujp7e3uZTFZWVmZtbb1jxw5tlwgAeKWFGdlPlcqvHe39DfR1XQsAXYxGQXjgwIGRI0f+/fffpaWltra2586d8/b2/vvvv2fNmjVw4MBmDlQqleHh4bdv36ZpetCgQQsXLtTT02u0T0lJyc6dO1NTU/X09Pr16xcWFmZoaNj6LwRAz7OruORsRdUQQ4NP7W11XQsAXY9GzwgzMzOnTp3KZDJtbGx69ep1+/ZthNDYsWPnzJmzYcOGZg7ctWvXgwcPVq5c+fHHH2dkZGzdurXRDjRNf/311yKRaPXq1YsXL05OTv75559b/WUA6IGyZfKPcvINmYwIYR8GrDUIQMtpOvEg9e9k9g4ODllZWerXI0eOXLRo0csOqa6uvnLlyieffDJgwACEUFhY2IYNG6qqqkxMTOr3yczMfPLkyf79+3v16qX+lB9//FGhUHCg5zcAGiBoenZahpgkI4R9enPhXw0AraHRFaGXl9epU6dUKhVCSCgU3rhxQ91eXFwsl8tfdlRycjJN076+vvUnwTAsMTGx4T5isdjT01OdggghDodD07T6gwAAr/RF3uN7deJgM9PZFma6rgWArkqjK8IVK1bMmDHDwcHh5s2bkyZNmjlzZnBwsI+Pz9atWwcNGvSyo0pKSoyMjLhcrvoth8MRCATV1dUN9/Hz8/Pz81O/FolEZ86c8fLy4vP/WymmuLj47t276tfV1dUsFqtFXw+AbuxGbd2PT4psOexwNxdd1wJAF6ZREE6fPp2iqKNHj9I0PX369LNnzx49ejQyMtLJyWnbtm0vO0omk7HZ7IYtXC5XJpM1ufPKlStzc3OZTOaXX37ZsD0jI2Pjxo0Nz6BJwQB0ezUqVWhaJk3Th4R9YHElANpC038/M2fOnDlzpvr1kSNHNmzYUF1d7e3t3czDPD6fr1AoGrYQBKGv33Tf7rVr11ZUVPzxxx/r16/fvHlz79691e1ubm5r165Vv96/f/+jR480LBiA7u3dzJwCuWKtg+0oI+hlDUCbvDoI5XK5m5vb+++///HHH9c3Ojs7v/JAExOTuro6pVKpvi4kCKK2ttbc/LkV0TIzMxUKhbe3t7m5ubm5ubu7e1xcXHx8fH0QWltbBwYGql9HR0cTBKHxVwOg2zpYUna8rMJPX7DOwV7XtQDQ5b26swyXyzU1Nb1z505LT63uJlN/DZeYmIhhmI+PT8N97t69u3XrVpqm1W+VSiVFUTCOEIBm5Mnly7Nz+QzGEWEfNiwwAUCbaXRrdMeOHW+++eYvv/yydOlShsazNxkZGQ0ZMmT37t0IIYVCER4ePm7cOPWA+nPnziGEAgICRo0aFR0dvXXr1jfeeIMgiKioKD6f3/wgfQB6MopG89Oz61RkuJuLmx5P1+UA0B1g9VdjzRg5cmR+fn5BQYFAILCxsWm0NT09/WUHymSy8PDwO3fuMBiM4cOHL1iwgMlkIoTUj/3UvWASExMPHz785MkTDocjFArnzp374keohYaGRkREZGVlubhAHznQQ63Pf/JV/uO3e5mc9hLquhYAugmNgrD51XfrhzdoGwQh6OHiROIhDx4Zs5iPXvO1YMNQIgDah0a3Rjss6gAALyMhydmpmSqaPuDmCikIQDvSaGYZAIDOfZCVmyWTrbC1nmBqrOtaAOhWIAgB6AKiyit/Kynz5OttdHLQdS0AdDcQhAB0dkUK5eLMbA6OHxX24eHwbxaAdgYzMwHQqVE0mpueWUmotrg4+gj4rz4AANBC8NclAJ3aT4VFV6prxxobLbe11nUtAHRPr74ijIuLy8jIsLa29vf3f3F9eQCA9jwUSz7PK+jFYh0UusIUMgBoSXNBKJPJJk+efOnSJfVbGxubffv2jRs3rkMKA6Cnk1PUvLRMJUXvFDpbPb+QCwCgHTUXhN9///2lS5eGDRsWEBBQWFi4e/fuqVOnZmdnW1vDLRoAtG51Tn6SRLrE2jLYzFTXtQDQnTUXhCdOnBAKhVevXlXPL+rr6zt//vyYmJjFixd3VHkA9FB/VVXvLHrqwuP+6Nxb17UA0M0111kmNzd3+PDh9bNsT548GSFUVFTUEXUB0IOVE8T89GwGhkUI+wg0nuYeANA6zV0REgQhEAjq35qYmCCESJLUelEA9GA0QmHp2SVK5XdODgMNml7IGgDQjmD4BACdyy9FT2Mqq4YZGnxk1/QyLACA9gVBCEAnkiaVfZyTb8hkHBb2YWAwYgKAjvCKcYQZGRnHjx9v2JKamtqoZebMme1fFwA9j4KiZqVmyChqn3sfBy5H1+WAzuv+/fuvv/76kiVLdu3a1ZGfe+3atWvXrq1cudLIyKgjP1fbXhGEMTExMTExDVuio6Ojo6MbtkAQAtAuPst7nCCWhFqYh5ib6boWAJpw7dq19evXv/POOz0oCDdv3txhdQDQw8XW1G5+UuTI5e5wddJ1LQA0JpfLuVyurqvQGrrrmDNnDkIoKytL14UA0M6qCcL+Thx+9ea16hpd1wK6gLi4OITQkiVL1G/79u07bty48+fP+/v7c7lcExOTsLAwkUik3ioSiT755BMnJyc2m21jY/P+++/X1tbWnyo1NTUoKMjY2FhPT2/IkCGnTp2q3zRw4MARI0bcunWrf//+Dg4Ofn5+9cExY8aMjvy+2tZcZ5kpU6Zs27ZN20kMAFiamfNYrvjMwW6EkaGuawFd0r1792bNmhUUFHTq1Kng4OD9+/d/9tln6k3z5s3bvn37tGnTfvrpp0mTJoWHh9c/z0pISPD39799+/bSpUu//PJLmqaDg4M3bdpUf9r8/PyxY8fa2dm9++67mzZtmjZtGkLo119/XbVqVcd/Ry1qJiQRQvPmzat/u3HjRj8/P20nczPgihB0S/uKS9DVm6/dT1BSlK5rAV3Di1eECKErV67U7+Dt7W1nZ0fTdG1tLYZhy5cvr9/07bffcjicp0+f0jT9+uuv29jYlJeXqzeRJDl69Ggul1tTU0PT9MCBAxFC27Ztqz923bp1CKG8vDytf8OO1YLhEwUFBfHx8doIYwB6rFyZfGV2Hp/BOCLsw4LxEqC1HBwcRo0aVf/Ww8OjpKQEIUTTNJPJPHPmzK1bt2iaRgitXbtWLpdbWlpmZGTExcXNnDmTy+WKxWKxWCyVSmfNmiWXy+/evas+j0AgeO+993TyjToSjCMEQGdUND0nLVNEkttcHPvo8XRdDujCTE2fm5mdyWQSBIEQMjQ03LVrV1VV1dChQ83MzAIDA8PDw+vq6hBCaWlpCKGffvpJv4GFCxcihEpLS9XnsbGxYfSASf5ghXoAdGZD/pM7daKpvUzDrCx0XQvo2rCX304ICwubNm3ahQsXYmNjL1++HB0d/d133929e1edlCtXrgwKCmp0iKurq3bL7WQgCAHQjVu1dRsfF9pw2HvcXHRdC+i2CgoK4uLi3nzzzeDg4ODgYITQ4cOH586de/r0aV9fX4QQQRBDhw6t3//Ro0dXrlzpaUEIt0YB0AExSb6TnkXT9EF3V1MW/D0KtOXJkyfTpk2LjIysb3FxcUEIsdns/v3729nZHT16tKysTL1JJpMtWbJk48aNBgYGTZ5Nfd2pftbYnbziX2BeXl79hGo5OTkIoUbzqyGYWQaAlnsvMzdbJl9jZ/OGcbeaoQN0NoMHD1ZPxhYXF+ft7V1SUhIeHm5jY/P2228zmcydO3dOmTLl9ddfnzVrFpvNPnnyZHp6enR0NI/X9BNrc3NzhNCGDRtmzJjx1ltvdexX0aZmepS2/QztC4ZPgO7hVFkFunrT694DGUnquhbQJb04fKLR2LbZs2fX/3IuKiqaNWuWubk5g8EwMTGZMmVKenp6/Z7Xr18fNWoUj8czNDQcMWLEhQsX6jcNHDjQzc2t4WnLy8v9/f1ZLNbixYu19d10AaNfHnhbtmzRJAhXrlypYWS2UWhoaERERFZWlvrSHoCuqFCh7Hv/oZSk7vn19ebr6bocAECzt0Y7LOEA6CEoGs1Ny6wiVDtcnSAFAegkNOosQ1HUyzadP3++/YoBoJv74Unh1ZracSZG79lY6boWAMAzrwjCHTt2ODg4MJlMOzu7n3/+GSH0559/BgYGDho0qF+/fnZ2dhMmTOiQOgHo8h6IxF/lPzZjsX5zd4UpZADoPJq7NXr06NFly5axWCwfH5/i4uLVq1dnZWXt2bOHJEkcx/l8voGBwYABAzqsVgC6LilJzU7LVFL0Pk8XSzZb1+UAAP7TXBDu3LnT2Ng4Li7O2dmZoij1asgCgeDEiRNjx45lsVgdViUAXd2qnLx0qew9G6sAUxNd1wIAeE5zt0aTkpICAgKcnZ0RQjiOf/rppwihkJCQCRMmQAoCoLnzVdV7ikuEerz/c+qt61oAAI01F4S1tbVmZmb1bx0cHBBClpaWWi8KgG6kTEnMT89iYthBYR89BszlBECn04K5ndRzkOM4/EsGQFM0QmEZWaVKYpNz79f1BbouBwDQBEg1ALRoW2HxH5XVw40MPrS11nUtAICmQRACoC2pEumnuQVGTOZh9z4MWHQXgM7qFbdGk5KS9u7d27DlwYMHjVrUCzkCABpSUNSstEwZRe13d7XncnRdDgDgpZqba7SZlR4bauYM7QvmGgVdyKrsvC2Fxe9Ymh9w71lLuwHQ5TR3Rbhnz54OqwOA7uTv6pqthcVOPO5WFydd1wIAeIXmghDueQKguWKFMqK0/LFCYc5i/lpcwsCwCGEfAyZD13UBAF4BlsYGoB1crKoJSkkXk2R9S5ilxSADfR2WBADQEPQaBaCtRCQ5Nz2zYQoihP6uqVG8fNkWAEDnAUEIQFvdqKkrVRKNGp/IFfdEYp3UAwBoEbg1CkBrEDSdKJbcqxPdE4kvVdU0uY9IRTbZDgDoVJoLwkOHDgUEBBgbG3dYNQB0ZtkyuTr57tWJHool8n/vfHKbmkEUR8hHwO/YAgEArdHcrdH33nvPwsLizTff3LVrV0lJSYfVBEAnUacib9bW/fC4MCApzeL2Pdd/4menZW4tLL5TJ7LmsEMtzLe4ON7w9a4ZMnCJdePJ6Ffb2dhyYN1B0K3IZDKBQJCQkNCwZdGiRSYmJsbGxmFhYXV1dRqeSiwW8/l8ExMTpVLZoq1SqfSrr75yc3PjcDjm5uaTJ0++du1aa7/QM81dEZaXl1+4cCEqKmrt2rXvv//+oEGDgoKCAgMD1ctQAND9EDSdKZXdqq27WSuKF4vTJNL62SKs2OxJpiZ++nw/fcEQAwMT1nP/dra6ONpw2DuLSp4qlbYc9nJb61UwuSjQORWBmO22ZF5VVdXnn38ukUgaNr7//vuXLl06ePAgh8NZtWrV/PnzIyMjNTlbZGSkUqmUSqUXLlwICAjQcKtUKh01alRmZuaqVav69u1bUVFx7Nix0aNHnzhxYtq0aa3+as3NLFOPIIirV69GRUWdOXOmpKSkf//+gYGBgYGBQqGw1R/cCjCzDNCGYoXyVl3dzdq6eJEkXiSuv+EpYDD6Cvh++nw/gcBPX+DJ19PkbAqK4sAKLUC3SJK8cZW8eY0W1WECfdx/KHPUmDYm4po1a3766Sf164cPH/br1w8h9PTpUzs7u5MnT06dOhUhdP78+YkTJxYWFlpbv/qvwLFjx+rr6yclJb3++utHjx7VcOvq1av37t177949Nzc3dQtN00FBQffv3y8oKNBwNrQXaRSE9SiKunXrVlRUVHR0dEFBgbu7e1BQ0DfffNO6z24pCELQLkqUyjiROF4kjhdJ7tTVVRIqdTsDw9z0eH4CgZ8+f6ihQT8BH2bKBl2R6o/T5PUrDVsYAwYxg0Lacs7CwsKKioqMjIyZM2fWB+GJEydmzZpVV1fH5/MRQlKp1MDA4MCBA6GhoSkpKZ6envWHi8Xi8vJyR0dH9Vt1gh48eDA+Pj48PLysrExP778/NF+2ValUmpqarlixolHo5OXlxcbGzpgxg8fjte7btazXKI7jw4YNGzZs2ObNm+Pj46OioiIjIzssCAFoSEKSjxVKew6bz3jF7C0SknwolqiTL14sTpVI6zfV3/Acamgw2MAAFs4FXQhdVdlEo6iOvHG1USN57w7u3Q/rZf7i/pixCdLgDz5bW1tbW9tGjTk5Oebm5uoURAjp6ekZGxuXlJQoFIpx48ZNnTp1+/btCCGpVDpx4kQOh3Px4kX1nkePHmUwGJMmTbKxsdm8efPZs2dnzpxZf9qXbU1KShKLxaNGjWpUhqOjY33Etk7rh0/4+fn5+fl9++23bfl4AFqhTkWuzsnb/7SUQghHaJ6l+c8ujkbM/36YSZpOl8riReJ4sTheJIkTiZTUszsf+gzGEEMDdWuhQbAAACAASURBVPINNzS0YLfbExQAOpjy/75GGk/aQOzb2WQ7Z/0mxOW2rgCRSNToIkwgEIhEIg6Hc/bs2TFjxrBYrG+//TYgIKC0tLRhl5aIiIgxY8YYGhoOHTrUxMTk2LFjDYPwZVsrKioQQiYmJurdYmJiGj4+PHnyZHBwcOu+CIwjBF3P4szsE2UV6tcUQgdKympJcruLU7z42Q3PW7V11apnNzyZGNZHj+cnEAw11B9iaCDU08PhfifoFnDvfuiFZ1u0REznZL24M+bojOkbNHGWV91QaYaRkZFUKm3YIpfL1UHVv3//ixcvjhkz5tSpU1wu99q1a5aWz3pWp6amJiQkLFu2DCHEZDLHjx9/8uTJ6upq9VC9ZraqLz3LysrU5xk+fPjDhw8RQgqFwt/fv9XfAkEQgi4nTSqrT8F6UeWVUeX/3SZy5fEmmBoP0NcfYCDwFfCh9wrolliz3mmilSSV//c1XV3VsA0T6LPnL0Wcdl4X09rauqKiQiaTqa8LFQpFeXl579691Vu9vb19fHxu3LixZMmSht1nDh8+jBBavHjx4sWLEUI0TVMUFRUVtWDBgua3enp64jj+zz//jBs3DiFkYGCgflSZmJjYxi8CvyBAF5MllTXZ3l/AX9/b/ryPR+WQgZkD+0cI+yy3tfI30IcUBD0Lg8EMmYca9D1BXC5zZmi7pyBC6M0330QIXbnyrGPO5cuXcRwfPXo0QkipVAYGBhYXFx87diwiIuKjjz5S70PT9LFjxyZOnJjwr8TERDs7u2PHjr1yq7Gx8YQJE3bu3FlT89xcTps2bWrjF4ErQtDF9GI1/VTvJxfHkUaGHVwMAJ0Q7uDIXvMF9TCOrqzAjE1w39eavinaZhYWFsHBwcuWLcMwTCKRLF++fNGiRQYGBgRBBAUFpaWlxcbG2tnZWVlZTZgwgcFgfP/999evXy8oKAgPD/fy8qo/T2ho6Pfff19SUpKRkdHMVktLyy1btgwcONDf3//DDz90cXEpKirau3dvVVWVmZlZW77Iq/9YjouLi4iIuHLlSqN7wQDohCmLyXnhKZ+7Hm+wgVb+qQPQFWF8PmPoSObbwYzho7WUgmp79uwZMWJESEjI0qVLg4KCNm/ejBBiMBhCofDq1at2dnYIoREjRsTExNjY2CCEjhw54uTkNHbs2IYneeeddyiKOnHiRPNbEULOzs4PHjwYMmTIunXrxo8f/8UXX/j6+t64cWPgwIFt+RbNjSOUyWSTJ0++dOmS+q2Njc2+ffvUN2d1AsYRgpu1dVOT0ysIwpDJqP13SmtHLve0lztM7AkAaJ3mbo1+//33ly5dGjZsWEBAQGFh4e7du6dOnZqdna3JrAEAtLsDJWVLM7MpGv3i6rTAyuJ8VXWuTOHI44w3MebCg0AAQGs1F4QnTpxQX94yGAyEkK+v7/z582NiYtSdeQDoMDRC6/Mfr89/os9gnPByG29ijBCaItCjVQpMoIcgBQEAbdBcEObm5oaFhTH+HWUyefJkhFBRUVFH1AXAvxQUFZaRfbS03JHLjfEWevD1kFymOhtJPohDNI0wjNHvNebbQYin0VygAADQSHNBSBCEQCCof6seJkmSsNYo6DiVhGpKctrN2rqBBvpnvITqiWCIU8eopH8XgqFp8mEcrZCz5i7UZKYoAABoBO4pgc4rRSJ9LT7hZm1dkJnp1X5e6hSkS0v+S8F/UalJ9FO4VwEAaA0IQtBJXaquGfowKV+uWG5r9buHO+/fB4F0ZXmT+9MlxR1YHQCg+3jFgPqMjIzjx483bElNTW3U0nCyVADaxf6npUszc2iEdvZxXtpw8XdCSTU1jyJCiIg8jic/wr37MoReiNvK1VgAAD1Qc+MINVzksEUrGrYFjCPsCeo7iBozmZFe7qPqJ4tRKsh7d8jYy3RdLcIQavRDJxBgXD26ogwhhHAct++Ne/fD+/lhAv0Orh8A0OU0d0WoniMAgA4jp6j56VnHyyqceNwYbw+hHg8hREsl5K3r5O1YJJUiNocxZATu1VcVdZwufzYJPdbLjBW6ELO0okufko8SqLRkKj+Xys9Ff5x+loh9+2t1cg0AQJfWshXqdQuuCLu3EqVyclJanEjsb6B/xktozmbRYhF55yZ58xqSyxCXy/AfyhjxBqbHRwghkqQy0+iqSszEFHd1R8zn/qSjqyqp1CQqKYEqyFMPscAdHHHvfri3L2YI85ECAJ7T3BXhlClTRo8evXz58lafXalUhoeH3759m6bpQYMGLVy4UE+v8WAvsVi8d+/euLg4uVzu7Oy8YMECNze3Vn8i6KKSJdJJSakFcsU0s14Hha7c2hrVX1fJe7cRQWB8AT5mPHPoSNRwCVAGAxd6vexsmIkpY+hIxtCRdHUVlfJInYhUfi46F4VZWOLevgzf17BebZqlFwDQbbziGeG8efN+++039dvvvvsuMjLy/v37mp9927ZtCQkJS5cuZbFYe/futbW1/fTTTxvts2HDhsLCwvnz5+vp6Z06dSo3N3f37t0Nxy/WgyvC7urv6pppKem1KnK5rdXPxob0rWvkP7eQSoXpGzCGjmQMGY5Y7DZ+BF1TTSUnUukpVE6WelHvZ4nYrz9mZtEeXwIA0FW1YBmmgoKC+Ph4zfevrq6+cuXKJ598MmDAAIRQWFjYhg0bqqqq1APz1aqqquLj49evX69eX9HFxWXOnDk3b9586623NP8g0KXtfVr6XmYOhqHfrMxmx99SJcQjisJMTBlDRjD8hyBm04sutRRmZPzsGlEiodJTqKQEKjONvHSevHT+WSJ6eGE2du3yWQCArkWL6xEmJyfTNO3r66t+6+XlhWFYYmLiqFGj6vcpLy83MTGpvxfK5/P19PQaLboIuqv6DqImDPxEacGwP0+QNI1ZWDFGvMHwfU1LM4hifD7DbwDDbwCSSsm05OcS0cQUF3rhPr64gyNMUgNAz6HFICwpKTEyMuJyueq3HA5HIBBUV1c33MfNze3AgQP1b+/cuSMSiVxdXetbysrKHj16pH5dV1fHZMJKwt2EhCRnp2WeqahyUimjr192k9RhVjaMYaO0F4GN6ek9S0SZlExNptJTqPQU8lYseSsWMzbBPbwhEQHoIbSYKzKZjM1+7tEOl8uVyWRN7kzTdExMzG+//ebv7+/n51ffnpKS8r///a/+LY8HA6W7g6dK5eS4h/cJlX9Nxan4m+bWNoxpM3F3T92kDu/fRCSUVHYm9eghmfzoWSIaGeOePrjQC3d2hTUuAOiutBiEfD5foVA0bCEIQl+/iQHOhYWFW7Zsyc3NnTZt2vTp0xtucnNzW7t2rfr1/v37668OQVdF04nJSQElFU9YnGlPH++rLefPW4S79NF1WQghhFhsXOiFC72YgQSVnUE9ekimJj1LRD4fd/PAfXzxPkL073osAIDu4RVBmJeXVz+hWk5ODkKo0fxq6OVTrJmYmNTV1SmVSvV1IUEQtbW15ubmjXZLSkr65ptvXF1dt2/fbmNj02irtbV1YGCg+nV0dDRBEK/+TqBzomkqKeH8P3dnOQpFTPZHtRXfDx2M2znouqymsFjPElFFUFkZVFoylfyIfBBHPohDenoMd8+XJSJdW0PXVGMmpjB+H3RLMpnMzMzs5s2b6u6N6pbly5dHRkbSND116tQtW7YYGGj0wy8Wiy0sLDgcTklJSaN7h1KpdNOmTceOHcvPzzc0NPT39//www9Hjhyp4eGt8IogvH79+vXr1xu2hISENNrnZUGo7ibz6NGj1157DSGUmJiIYZiPj0/DfZRK5Y8//jhkyJBly5ZpOKMb6HpIkkyIJ69c2MvTX+H5GgOhg1a9QkcN1XVZGmA+S0Q0ZTr1OJ969JB69PBZIvJ4DKEXLvTC3T0Rm02LRapTx6i0ZPVxuI8vK3AGLJEIdKhYocyTyx24XFtOW3NCraqq6vPPP5dIJA0b33///UuXLh08eJDD4axatWr+/PmRkZGanC0yMlKpVEql0gsXLgQEBNS3S6XSUaNGZWZmrlq1qm/fvhUVFceOHRs9evSJEyemTZv2ysNbR4tTrBkZGQ0ZMmT37t0IIYVCER4ePm7cOPWA+nPnziGEAgICEhISampqBg8enJaWVn+gmZmZmRkMdu4WlEry3m3y+hVVXe06t74/OrqZMPEoL48RRl1tehccx3s74b2d0KSpzxIxKeFZIrLYuEsfuqqCLi2p35169JAgSdbchTosGfRYVYRqaWbOyfIK9dtJpiZ73Jwt23bZtGbNmp9++qlR49OnTw8dOnTy5El1FP34448TJ04sLi62trZ+5QkjIiICAgKSkpKOHTvWMMm++OKL9PT0e/fu1Y8mCAsLCwoKWr16dXBwcP310ssOb53mBtTTNN3GqzSZTBYeHn7nzh0GgzF8+PAFCxaou32qH/tt3Ljx999/j4iIaHRUSEjIi9edCAbUdy0yGXn7OnnrGi2RiLl684e9eY7JceFxY7w93PS6RY8nmqbyc6nkRCo5ka6pbnIX9odrMQvLJjcBoD1Tk9NPV1Q2bBltbPi3jxfehl/nhYWFFRUVGRkZM2fOfPjwofrW6IkTJ2bNmlVXV8fn8xFCUqnUwMDgwIEDoaGhKSkpnp6e9YeLxeLy8nJHR0f126dPn9rZ2R08eDA+Pj48PLysrEx9jaRUKk1NTVesWPHNN980/PS8vLzY2NgZM2ao+0u+7PBWa+6KcNiwYXv27BEKha0+O4/HW7FixYoVKxq1b9y4Uf1i+vTpjXrHgK6OlojJm9fI2zfUE4SWjhoXaGoVL5ENMTQ47eXei9U+A+R1D8NwR2fc0RlNmkpevqj6+48Xd6GrKiAIgfacKq+kG6/DgooUykYpiBC6Ul373ZMnfZrqdT+llylLgwseW1tbW1vbRo05OTnm5ubqFEQI6enpGRsbl5SUKBSKcePGTZ06dfv27QghqVQ6ceJEDodz8eJF9Z5Hjx5lMBiTJk2ysbHZvHnz2bNn1Y/YkpKSxGJxw7Hmao6OjvUh2szhrdZcEN66dcvX13ft2rWffvopq9v8/gJaQ9fWktcvk//cRoQS0+Mz3pyQ0m/ApMzcJxLZO5bmu/u4sNvyF2mnhWGYk1PTWwy62h1g0KWEpGaoNF414fPcx0221w71ZzFb2RFaJBI1GtImEAhEIhGHwzl79uyYMWNYLNa3334bEBBQWlp67dq1+t0iIiLGjBljaGg4dOhQExOTY8eOqZOsoqICIVQ/+1hMTEzD254nT54MDg5u5vBWay4Iz549u2LFinXr1v3+++979+719/dvyyeBboyuriJvXCX/uY1UBCbQx0e8wRw28i+pfEZqhkhFrutt91Vve13XqEW4gxNmaU2XFDdsxPgCzKpxL2gA2tFGJwfqhSAsViq3FT59cedF1hbO/05v0hCnDX+eGhkZSaXShi1yuVwdY/3797948eKYMWNOnTrF5XKvXbtmafns7khqampCQsKyZcsQQkwmc/z48SdPnqyurjY2NlZfXJaVPVthbfjw4Q8fPkQIKRSK+gBq5vBWf5HmgjAgIGDMmDHffffdpk2bhgwZ8t5773333XdNTocNeiy6soKMvUTG3UUUhRmbMIaOZAwcglisrYXFq3PymRh2WNhntkV37/rEYLDmzCcO76dL//0FxGLTErEq6jgzKATmpgFa8pFdE39p0QjF1Ynv1IkaNnrz9X5xddbkFmiLWFtbV1RUyGQy9XWhQqEoLy/v3bv3sw/19vbx8blx48aSJUsadp85fPgwQmjx4sWLFy9GCNE0TVFUVFTUggULPD09cRz/559/xo0bhxAyMDBQP4xMTEzU5PDWfxNaA9nZ2ePHj0cI2dnZnT59Ou95mpyhXcyZMwchlJWV1WGfCNSomhrV338Sxw8Rf52jKsqeNRYXEccPyf+3Qv7xMsX3X6nu3qJVKpqmVRS1LDMHXb1pevPu9ZpanRbesUiSzM5Uxd0hc7NoiVixdZP842XEyaM0Rem6MtCz5MnkfvcT0NWb6v+87j1Ik0jb5czq67OHDx+q35aUlDAYjJiYGPXbP/74g8Vi1dbW0jStUCgmTJjg7Ox87NgxPp+/Zs0a9T4URTk4OEycODGpATs7uzfeeEO9w6RJkywtLaurqxt+7qxZsxBCJ0+efOXhraPRzDLOzs5//vlnVFTUnDlzpkyZ8mKUtj6HQadH5WYTB3Yj5bNJgsjrV5lvjKMe51PpKYimMUtrxvDR9ROEikkyJDUzprLKlcf7w0fo2qOmxMNx3NkVoWcz5bIWvEuEbyfj7iAMMQNnwnUh6DC9uZx7/fveqqvLkckduJxhhgZM7fz4WVhYBAcHq0eBSySS5cuXL1q0yMDAgCCIoKCgtLS02NhYOzs7KyurCRMmMBiM77///vr16wUFBeHh4V5e/60nGhoa+v3335eUlFhaWm7ZsmXgwIHqQfQuLi5FRUV79+6tqqpSj6l75eGt+yKaTrGWkJDw448/ymQyLy+vhp1iQfujaSo3m64sxwyNcGfX9lqHqJVIUnX8UH0KIoSQilBdiEEIYda2zFFjce9+9b/iixTKgKTUh2LJG8aGpzzdjXr2DOkYX8Ba9AERvoO8dwdhGHPqDMhC0GFwDA0zNBhmqPUZjvbs2bN8+fKQkBAmkxkSEvLzzz8jhBgMhlAo3LFjh52dHUJoxIgRMTExycnJCKEjR444OTmNHTu24UneeeedjRs3njhxYsWKFc7Ozg8ePFi/fv26deuqqqqsrKymTJny1VdfhYaGanJ4675Fc+MI1Wpra7/88stffvmFzWZ/+eWXa9as0dUSED1hHCFdW0sc2kMXPuvfhZmYsuaEaXGdPJUKEUpaJkOEEhHEfy/kcqRUIIKgykqoh00sxcwcO54xZnzDlkSxZFJSaqFCGWZlsatP+z+N6KJoUR2xeztdXsoYOpIZEKjrcgAATXhFpB05cmTNmjUlJSWjRo0KDw/vxgnUSahOHK5PQYQQXVVJROxnr/oUNTkrBEkipYKWyxFBvCTPZEipRARBK+RIoUCEklYokEKOCIJWKpFchlp9W9u0V8N30RWVc9IyZSTV7TuIthSmb8BasozYtY28eQ3hOHNi4ycLAACday4IR44cGRsba2xsvHfv3jZ1yAGaoSvLqZzMxo1VlcTeXxGP2/BC7VmeUVSLP4PBQGwOxuVi+vrIxBTj8RCLjVisZy+YLIzHQ2w2YrEwDhdxODSpUh3e/+IH4Tb/pd3WwuIPs/PYOH7Eo0+IeXfvINpymL4Ba/EyInwbef0KwnDmhMm6rggA8JzmgjA2NjY4OHjHjh0WFhYdVlBPRotETbZTBbnPXuE44nAxDgcTCJCJKcblIhYLsTkYl4dYLMRiYTy9Zy+46jxjY1wuYnMQi4VxOIjDbc2iemPGqy7+O20KjRCGGIOHY+YWCCGSpldk5/1S9LQXixXt5T5U+w8kuijM0JC1eBmxexsZewnhGPOttk6NCABoR80F4ZkzZyZPhr9eOw5mYoow7MXblcyZoQx3T8Tm6GQlPMboN5G+PnnzGl1ZgRkZMwYOYQwZgRASkWRIasYfldWefL0Yb4/eXE7H19aFYIZGz7Lw6t8Ix5lvTtR1RQCAZ5oLQkjBjlZXizGZ9PNrLuIOjoy+frpcHh3DGAMGMwYMbthWpFBOSkpNEEvGGBud9HTr4R1ENYQZGbOWLCd2byMvX8BwvFFvIwCAruju1yt4HpWZrgzfTqtUuJ19fT973N2TOTtMlyn4rwyp7I/K6mSJlEboXp34tfjEBLFkgZXFnz4ekIKaU2chZmKq+vs8efkvXZcDAEBI83GEQKvIuLuqqOMIx1kh8/C+/ZFMSleUIyPjzrDQeTlBzE3L+qvq2UpDQj29fIVcDh1EWwszMlbfI1Vd/BMhjPHGOF1XBEBPp/tLjZ6OplV/n1edOoo4XNbC9/C+/RFCiKeH2Tl0hhRECL2T/l8KIoTSpFIFRUUI+0AKthpmbMJa9D5maKi6+Ad55aKuywGgp4Mg1CmKUkWdIC+dxwwN2UuW446dbphmulT2Z2XjVWcpGllx2rTaNcBMzViLl2MGhqoLMeTVv3VdDgA9Gtwa1R2lgog4QGWkYpbWrLClmKGRrgv6j4QkE8XSh2JxzAspqPZYrmiyHWgO62XGWrKM2L1d9dc5hGGMkWN0XREAPZRGQSgWiz/77LPo6OiSkhLi+T6NCCbdbhVaVEcc2E0XPcFd+rBCF6Km1gnrSJWE6qFY/FAseSiSPBSLM6Wy5sfq23FgsEQ7wHqZP3te+Nc5xGIzhgzXdUUA9EQaBeHy5csPHDjg7e09ePBgXU002p3QpSXEgV10dRWj/wBmcIhORgcWK5TxYnGqRJoikcWLxWkSaf2fM2wcc+fr+QkEfvp8P33B+vzHl6prGx7bX18wzKhTPL/sBjAz82fXheci1ZMV6LoiAHocjVLt9OnTixcv3r17t7ar6QmogjzVwXBaImEMGcEMCOywFQnUyRcvEseLJHEiUanyvyt7AyZjsKGBnz7fTyDw5Ot58/nsBotWHxb2mZ2WeeXfLBxgIDgqdIM5tdsRZmahno9UdTYSYRhj0DBdVwRAz6JREEqlUhhc3y6o5ETi+CFEksyp0xn+Q7X3QSqazpDK4kXieLE4VSJ7IBZXEar6rVZs9hhjIw8+z08g8NMXePD1mok1Szb7cl+vJIk0Ryaz53D6CQQ4hGB7w8wsWAvfJ/ZsV505hTBMqz8bAIBGNArCvn37pqenT5wIk0K1CXnzmiomGrFYrLkLcaHXqw9oCSVFZ8meJV+8SPJAJJY1mCnbis2eZGqivtX5ur7Assm1LJrlzdfz5uu1a8ngOZiVtXr9QtXpkwghyEIAOoxGQbhu3bo5c+b0799/1KhR2i6oe6Jp1flzZOwlTI/PfGcx7uDYzL4Xq2p+KizKksptOewFVhZzLc2bvACrVZFJEon6Vme8WJwhlZH/9lpiYlgfPZ6fQODJ53nw9QYbGJiy4MluF4BZ2bAWfaDcs111+iTCcMbAwa8+BgDQZhr9fty2bRtCaPTo0Q4ODqamptjzz4fu329i4VbwH5WK+D2CSnyAmfZihb2L9WpuoaKDJWXvpGepX+fJ5Tdq65Il0v9z7o0QKlYoU6XSFIlUnXwNu7ewMMxNj1ffvcVPX8DrBLOygVbArG3Yiz5Q7tmhij6BWCxG/9d1XREA3Z9GQVhSUmJvb29vb48QIklSyyV1LzIpcXAPlZeD2Tmw5i/B+IJm9pWS1LKs3EaNPz4pihOJM6SyEqWyvrH57i2gS8OsbdkL31fu+UV18gjCEMMXshAA7dIoCBMSErRdR7dEV1cR+3fSZaW4hzdr1jzEesWTuSSJRESS6jX/GoqtqbXmsCeaGvsKBL4Cvq8+31HX4w6BVmE2dqz5S4j9O1W/H0EIY/i+puuKAOjO2nQDLTY2dswYmA6jaXRJMbFzC11Wynh9ECt0wStTECHEVN9zfuHSbquLU9Gg12O8Pb52tA80M4UU7AlwB0dW2LuIxVL9HkElxOu6HAC6M42uCGma3rx586VLl5QN7s4hhPLz88vKyrRTWNdGZWUQEfuQQsEYM545VtNl53wEfEs2u+S5/8k0F2e83ctEG0WCTk6dhcS+ncSJwywMezYhOwCgvWkUhDt37ly9erWZmZlSqayrq7O3t5fJZGVlZdbW1jt27NB2iV0OGX9PFXkMIcScEdqim1osDOuvL/izsqpBG/aDU28HWPy9p8J7O7EW/JuFTCbu6aPrigDohjS6NXrgwIGRI0cWFxenpKTQNH3u3LnS0tKLFy8qlcqBAwdqu8Suhbx2SXXyCGIwWe8sbumjnUMlZX9WVjlxuQutLN8wNpxvaX7d13u5rZWWSgVdAt7biRW2FDEYxJEDVGqSrssBoBvSKAgzMzOnTp3KZDJtbGx69ep1+/ZthNDYsWPnzJmzYcMGLVfYdVCUKvp31fmzmECf/e5KvI+wRUffrRMtzszRZzBifDz2uDlf6uu13911mCFM6QkQ7ujMmrsI4TgRsZ9KS9Z1OQB0N5p2lqH+nabEwcEhK+vZQLeRI0f+/TcspYYQQkipJA7tJe/exCysWB+sxqxtWnR0qZKYlpJOUNRRjz5CPZ6WagRdF+7qxpq3CGE4cXgfZCEA7UujIPTy8jp16pRKpUIICYXCGzduqNuLi4vlcrkWq+siaKmE2PsLlZaMO7uy312JGRm36HCCpqenphcqlF852k8yhX4xoGm4q/t/WZieoutyAOg+NArCFStW3Lp1y8HBIS8vb9KkSffu3QsODt6wYcPnn38+aNAgbZfYydGVFcQvP1MFebh3P1bYUsRr8fXcyuy86zV1b/cy+czeThsVgm4D7+POmrsQYRgRsZ/KytB1OQB0ExoF4fTp048dO+bn50fT9PTp02fNmhUZGblu3TojIyP17Gs9FvW4gPjlZ7qinDFkBGv2fMRktfQMh0vLfi166qbHO+jeByaHAa+EuwlZcxcimiYOhlM5mbouB4DuAGvd+vI5OTnV1dXe3t6cDlypPDQ0NCIiIisry8XFpcM+tBlUahJx9CAiVczJQa1bQ+6hWDLkwSMmht3t7+MBazsAjVHpqcThvQjDWfOX4M6uui4HgK6tBTPL1NbWXrhw4dChQ6WlpQKBwNPTsyNTsLMh790hDu9DNMUKmde6FCxVEm8npckp6oC7K6QgaBHc3YMVMg9RJHFgN5WbpetyAOjaNA3CTZs2WVpavvXWW/PmzcvKyjp27JiDg8OZM2e0WlwnRdOqv8+rIo8hLpe16APcx7cV51DR9IzUjCcKxZe97YLMTNu9RtDt4V59WSHzEKkiDuymcrN1XQ4AXZhGQRgZGfnJJ58MHjx479696pbRo0c7OztPnTr14sWL2iyv8yFJ4vhh8tJ5zMSU/d6HeG+n1p3mw+y82JraAFOTLx3s27dAKUoQqQAAIABJREFU0HPg3v1Ys95BKhVxYBeVB1kIQCtpFISbN28ePHjwpUuXZsyYoW7x8fG5cuWKu7v7xo0btVleJ6NQEL+FUwn3MVt71vsfYmbmrTtNRGn59qKnffR4h4Su0EEGtAXu3Y8VMg+pVMRv4dTjAl2XA0CXpFEQPnz48O233260Hi+Px5s2bVpiYqJ2Cut06Lpa5a4tVGYa7urOXrwME+i37jwJYsmSzGwBgxHl6W7EhIXjQVvhPr6smXORUkns+5V+AlkIQItpFIRsNrumpubFdrlc3mg9iu6KLn1K/PIzXVzEeG0ga/4S1NpeQpWEKjA5XUZSB9xdPaGDDGgneN/+zKAQpJAr9/1KFz7WdTkAdDEaBeHo0aMPHjxYUVHRsLGkpOTYsWP+/v7aKawToXKylL9uoWtrGGPGM6fNRgxG686joulpKel5cvnnDnbB0EEGtCvGawOZwSFILlfu/QWyEIAW0SgIv/nmG7FY3K9fv6+++gohdOrUqQ8++EAoFD59+nT9+vXaLVDXqKQEYv8upFQwp87QfGXBJn2Uk3+1pvZNE6N1vWEGGdD+GK/5M4PUWfgrXfRE1+UA0GVoFIRCoTA2Ntba2vqnn35CCG3duvWXX37R19c/ffr00KFDtVyhLpE3rxFHDiAcZ81bzBg4uC2nOlpavqWw2JXHO+HhxsCghwzQCsbr/szAmUguI/btpEuKdV0OAF2Dpp01+vXrd+/evYyMjLS0NIIgHBwc+vfvz+zGfT0oSnU2krxzA9M3YM1fgtm06RouUSxZpO4g4wUdZIB2MQYMQohWRZ0gwncwg0Kogly6tgYzNWMMHIwZGum6OgA6o5b9UnZzc3Nzc9NSKZ2ISkWcOEw9eoiZmrEWLMVMzdpysipCFZiSLiOp4x5uXtBBBmgfY8BgpFSqzkURh/bUN5I3rrDCluKOnWJ6QgA6leaCUCAQaHIKsVjcTsV0DlIpcTCcys/F7Xsz31mC8fltORlJ07PTMnNl8rUOttPNe7VXjQA0D+/nh86fRSrVf01Kper4YfbHX7a6txcA3VVzQSiRSBBCbm5ujo6OFEWRJNlRVXUsiqJrazB9fcRk0VWVxL6ddEUZ7t2PNTO0FatJNPK/3IK/qqrHGhtt6A0zyICOQ+dkPZeC6saaarq4ELNz0ElJAHRazQXhiBEjbty4kZGRoa+vP3PmzOnTp9vZda/ujhSluvwXef0KUioRjuMubnTRY1oiYQwdyZw0FbW5S0tUeeVPT4p6czlHPfpABxnQkWiCeFk7/CAC0EhzvUavXbtWWFi4ZcsWFov10UcfOTg4DB06dPv27SUlJR1Wn1ap/v6TvPQXUs8JQFFUZhotkTDeeIsZENj2FHwklsxNz+LieKSney9WW68sAWgR3Lapv1mZTNzKusNrAaCze8XwCSsrqxUrVty+fTs/P/+HH36Qy+XLly+3sbEZPXp0eHh4ZWVlx1SpFXIZGXv5xWbcvnfbz12tUgWmpEtJ8oC7a399jR61AtCOMEvrF1cHw0zNEIerk3oA6Mw0XYbJ3t7+o48+un//flZW1oYNGyoqKpYsWWJpaTl+fJvGmOsQXVWJmnrqSZeXtvHMFI1mp2bmyOQf2dvMgA4yQEeYAYHMSVMxcwvEYmNWNlgvc7r0qerPHrl0GgDNavGYNhcXl88++ywwMPDLL788derUX3/9pY2yOgLvJSMZ9NrUTRQhtDav4HxV9RvGht86Qq8EoDsMBmPYKMawUep3tFRC/LKZvHEVMzJmDB2p08oA6FxasEI9Qig3N3fjxo19+/b18PCIjo5W3yDVUmXahhmb4E6Nx1RhfD7u7tmW00ZXVG56XOjA5Rz3cGNCBxnQaWB6fNaCdzGBviommkpK0HU5AHQiGgVhYWHhzz//PGDAAGdn588//5zP52/btq2oqOjy5cuLFi3Sdonaw5wRiplb/PdeT485c15bBg6mS2XvQAcZ0FlhJqas+UsQi0UcP0zl5+q6HAA6i+ZujZaVlZ08efL48eO3bt2iadrX1/eHH36YOXOmvX03GRKHGRmzV/6PSkumy0qRgSHu4YW14b5otUoVkJRapyJ/c3f1gw4yoFPCbO1Zs+cTB/eoDu5hvbeq1YtLA9CdNBeE1tbWJEkKhcIvvvgiKCjI2dmZpmn0wlQyGk5A00kxGLhX37afhqLRnLTMbJl8tZ3NPEv45QI6L9zdkzlluirqOLF/J+v9D1u9xDQA3Qamzramt2n2iKuZM7Sv0NDQiIiIrKwsF5dON1/iZ3kFGwsKRxsbXvDxhEeDoPNT/XmGjL2M2dqzlyxHbLauywFAl5q7IpwxY0aH1dGlnamo+q6g0B46yICugzl+MhLVkQ/iiCMHWPMWIbxl/eYA6E6aC8Ljx493WB1dV4ZUNjc9k4PjkZ7uZtBBBnQVGMYMnkXX1VLpKarTvzMDZ+q6IAB0Bv4MbBMRSU5NTqtTkb/2cX4NOsiAroXBYIUuwCytyX9uNznLEgA9BARh61E0mpWamSaVrbS1ng8dZEBXxOWxwpZihkaq82fJh3G6rgYA3YAgbL2v8h/HVFYNMTT4wam3rmsBoJUwQyPWvEWIxVadPEplZ+q6HAB0AIKwlc5WVH1b8MSOw4nydGfj0EEGdGGYjR1rznxE08ThfXRJsa7LAaCjQRC2hrqDDBvHI73czdnQQQZ0ebibBzNwBpLLiP276NoaXZcDQIeCIGwxEUkGpqTXqshfXJ1ehw4yoLtgvD6IMfpNuraGOLgHKRW6LgeAjgNB2DI0QvPTs1Il0mU2VmFWFq8+AICug/nmRIbfALroCRFxAFGUrssBoINAELbMhvwnkeWVgw31f3R21HUtALQ3DGMGheCu7lRGqirqhK6rAaCDaDcIlUrljh07Zs2aFRISsm3bNqlU2sye06dPz83t1DPin6us2pD/2IrNPukBHWRAN8VgsOaEYVY2ZNwd8spFXVcDQEfQbhDu2rXrwYMHK1eu/PjjjzMyMrZu3drkbiKRaN++fXK5XKvFtFGmVDY3LYuF46e9hNYcmJsRdF9cLitsKWZkrLr4Bxl/T9fVAKB1WgzC6urqK1euLFq0aMCAAb6+vmFhYXfv3q2qqmq02/79+2fPnn3+/HntVdJ2YpIMTEmvUal2uDoNMIAOMqCbwwwMWWH/3969x0VR7n8A/85ekL3BckcQVlDEFBUBBQTBlcgKSRJNLpnHSwckzS6maZ5T/kxfnlMmloVSaV4QEQXTjkfDl8BGonAOhYpgWSLCS7kod3aXvczvj7F1W8Er7INnvu8/fO08Ozvz2QX3yzwzzzzJYGmpPZSp/7WKdByE+lc/FsILFy4wsxgyiz4+PhRFlZeXm6z2wgsvpKamvvPOO/2X5DHRAAuqLld0di12cV6EF8ggdqCcBvNfWQQAmr076Ot1pOMg1I/uddPtx3Tjxg2pVGppacksDho0SCwWNzc3m6xmb29vb2/f20YaGhrOnTvHPG5ra+Px+jFwb9ZfvZbd2BRsJUkd7mn+vSNECsfTi/9Somb/Hs2ObfzX3qKkNqQTIdQv+rGuKJVKiz/Pc2ZpaalUKh9qIxUVFe+++65hUSAQ9E24B5bX3PJB9TVnC4tsvIMMYh+ObwCvqUmbd0yzY5vF4jfA7P8BETKDfiyEIpFIrf7TsFyNRiORPNx02MOGDVu6dCnzOCsr6/z5832W7wFcUaniL17iABwY7e2KF8ggVuI+/Szd3qY7U6TZ8xV/YQpwuaQTIdTH+rEQ2tratrW1dXd3M8eFGo2mtbXV0fHhZmlwd3efN28e8/jkyZPd3d19H7QXHTrdC+crb2q020cMm2xtZbb9IjTQ8GbMoltb9JUXNAcy+HFzAWefRv9b+vFiGeYyGcMZvvLycoqixo4d23977EM0wMJLly90ds1zdvyrizPpOAgRxeHw4+dRLkP0P/9He3JAX+CN0CPoxyNCqVQaEhKyfft2AFCr1enp6dOmTRMKhQBw9OhRAIiOju6/vT+mjTW1Bxqagqwk20cMI50FoQFg0CD+gmTN55/oTh6nJFbcoFDSgRDqM/07oH7JkiU+Pj4ff/zxF198MWnSpEWLFjHtxcXFxcXF/brrx3GyueVvV2qcLPgHR48cxMG70CEEAEBJrPjzk0Eg1B45pP+lknQchPoMRdM06QwPau7cuXv37v3111+HDx/ef3upVqkn/Le8Vas9Oc4nTIqnBhH6E/2Vy5qvvgAu1yJ5GeUyhHQchPoAHu78iVKvj62oatJotnh5YhVE6G4cj+H8l16G7m7Njm10s+mNohB6EmEhvIMGWFD1a1l7x1wnx8V4gQxCveCM8+M9O51ub9Ps3AbKXu+kj9CTgsCNWgYUDU0fbbp1Sal0sbC4olTvb2gaLxZt98YLZBC6F+6USLq1VXdaodn1JX/Ra0Dilk8I9RVW//rWqNTPnquo7PrjZjcU2PB4R8aMEuAFMgjdDy96Jt3aoq84pzmwlx8/DwcXoicXq7/x51X9eqcKAgANegBLvI8aQg+Cw+HHz+O4D9WXl2mPf0c6DUKPjr2FsEalLmhpNWls1Wr/ddP0tuAIoZ7x+by/JFH2DrqCPN1pBek0CD0i9hbCm1ptz+2antsRQnejRCL+gsWUSKw9mqOvOEc6DkKPgr2F0NPSkt/TWY2RQry/PkIPgbKz589PAi5Pk7lLX1NNOg5CD429hdCax33H3dWkMUxqNc1WSiQPQk8uyk3GfykRtFrtN9vppkbScRB6OOwthADwwVD3v8ncRFwuAHApKt7RIXvUSC5e/IbQw+OMHc+LiqE7OzU70ujODtJxEHoIrB4+waeo//Nwf3+oW41a7WxhgaMmEHoc3MlyuqVZV1Sg/WY7/69LgY9TeKInA371A5eiPCwtsQoi9Ph401/k+IzT11zVHMiAJ+c+xojl8NsfIdR3KIof9wpH5qE/95P2X4dJp0HogWAhRAj1KT6f95e/Ug6Ouh/ydUUFpNMgdH9YCBFCfYwSivgLFlNiifa7XP2FctJxELoPLIQIob5H2drx5ycBn6/Zv1t/9QrpOAjdCxZChFC/oIa48xP+Ajqd9pt0urGBdByEeoWFECHUXzhP+fBiZtNdnZodaXRHO+k4CPUMCyFCqB9xA0O4YVPpWzc1O7dDdzfpOAj1AAshQqh/8Z6fwR0/ga6t0ez7BvR60nEQMsXqO8sghMyBonizE+i2Vn3lBU36VtBp6bZWysGJGz6V4zWSdDiE8IgQIWQGXC7/lYUgFOmvXNbXVNMtzfpfqzRffaH7qZR0MoSwECKEzIK+dQu6Ok0atYcPgkZDJA9CBtg1ihAyB31NT6MJVUrtnq/BwZESiUEkpkRiEIlu/ysUAU4Fg8wCCyFCyCw43B6bdZcuwqWLPTxBUXcqokhEiSQgFlNCEYjElEhEiSUgFFEiMfD5j5VKq9EVntKVldDtbZSjM29KJMdn7GNtED2BsBAihMyBM3wE8Pig/VNHKGUt5ae8CWoV3dZGt7WCsotWKkHZRbe3QVsrreyCri59Q/29JrLg8SmBAIRCSiAEK2tKYgUCISUQgEBIWVlTVlYgEFJiCfQyvYzmQIa+vIwGoADoa1c1e77izU7kBgT27XtHAxwWQoSQOVC2drxnp2u/y73TxOPz5sylpDYAQDkN7vWVWi3d1XmnRv5RMpkWuq0V2lrppkZap7vX7nl8ysqKsrI2LpO0SqkvLwMA4x5Y7dEcrq8/8PC7kUXwh40QMhPuZDk1xF3/3xK6tYVycOSGhFF2Dvd/GY9HWVmDlfV9ThhqNLSyC9rb6LZWuut2mYT2VrqtDZRdtw8uq3+//+5USrqpkXLuvTCj/zlYCBFC5sPxGMbxGNYvm+bzKb41WFlTrm69rqPR0J0d0NlBd3TQXR36ivP68z/3uKl+SYgGKiyECCHW4PMpqQ1IbZiDS4770O7KCtPTlk6DKVs7IukQKTiOECHEUpSdA296jFEDDQIBP+4VHLbBNnhEiBBiL27wZI77UN1P/6HbWjmOzpygEEosIR0KmRsWQoQQq1Gubrx7nFZELIBdowghhFgNCyFCCCFWw0KIEEKI1bAQIoQQYjUshAghhFgNCyFCCCFWw0KIEEKI1Z68cYTp6el2dngDJITQE0Amk8XFxZFOge6HfnKcOnVqxIgRff4JWFpaymQyqVTa51t+ZAKBQCaTWVtbkw5yh1AolMlkVlZWpIPcIRKJZDKZWCwmHeQOiUQik8lEIhHpIHdYWVnJZDKhUEg6yB3W1tYymUwgEJAOcodUKpXJZJaWln2+5YiICNJfnOj+KPoeM14OPAUFBY2NjX27zerq6qysrODg4LCwsL7d8iO7fPnyoUOHwsLCgoODSWe5rbKy8siRIxEREQEBAaSz3Hb+/Pljx45NmzbN19eXdJbbysrK8vLyoqKifHx8SGe5raSkJD8/PyYmxtvbm3SW23788ceioqLZs2d7enqSznKbQqEoLi6Oj493d3fv2y07OjqGh4f37TZRn3vCukanTJnS59ssKSnJysp66qmnZs+e3ecbfzQKheLQoUM+Pj4DJ9L3339/5MiRcePGDZxIlpaWx44d8/Pzi42NJZ3ljry8vIkTJ0ZFRZEOcptKpcrPzw8ODo6IiCCd5baWlpaioqLQ0NCQkBDSWW5raGgoLi4ODw8fOH/nIXPCi2UQQgixGhZChBBCrPaEnSNECCGE+hYeESKEEGI1LIQIIYRYDQshQgghVnvChk/0B5VKNXfuXD6fv3v3bh6P8AeSmJjY3t7OPKYoys7OLioqiuzwgNbW1v379589e7a5uVksFo8fP/7ll192dHQklWfZsmVXrlxhHnO5XAcHh4iIiNjYWII/u9WrV1+4cMGkUSKRZGRkEMnD6DFVTEzMggULiOQBALVanZOTo1Ao6uvrRSKRt7f3jBkzxowZQyrP4sWLx4wZk5KSYtwYFxc3Y8aM+Ph4UqmQ+WEhhNOnT2u1WrVaXVZWNnHiRNJx4LnnnouMjAQAtVpdVFS0a9cuJyen0NBQImFu3ry5cuVKAIiNjR08eHBTU9PRo0dXrFiRmppK8F48AQEBL774IgBoNJqLFy9mZ2f//vvvq1atIpUHAHx8fF544QXjFj6fTyqMwciRI2fOnGnc4uLiQiqMWq1+77336urqZsyY4eHh0dbWplAo1qxZs2LFioEzoBCxExZCyM/PnzhxYnV1tUKhGAiF0NbWdvjw4czj0aNHl5WVlZaWkiqEaWlpALB582aJRMK0hIaGLlmyZN++fSZ/R5uTjY2N4TDCz8/P3d39448/LikpIfjjs7e3DwoKIrX33tja2g6cVBkZGbW1tZs2bXJ1dWVann766Y0bN3799deTJk2iKIpsPMRmbD9H2NzcfP78+eDg4IkTJ5aUlKjVatKJTFlYWFhYWBDZdXNzc2lp6cyZMw1VEACEQuGbb75JsDvrbqGhofb29gqFgnQQ1CutVnv8+PHp06cbqiAAUBS1YMGCxMREjUZDMBtCbD8iLCws5HA4EydOtLOz+/bbb0tKSiZPnkw61G1arbagoKCuri45OZlIgN9++42m6VGjRpm0D6gqCAAcDsfT07OmpoZgBpqmdTqdcQuHwyF+lEPTdHd3t2GRw+GQOpNaXV2tUqnu/s1xcnJycnIiEomh0+lUKhXBAGggYHshLCgo8PX1FQqFo0aNkkgkhYWFxAvhvn37MjMzAUCv1wNAYGCgoafUzJRKJQAMqOkdeiOVSuvq6ggGKCwsLCwsNG6ZP38+cyKToOLi4lmzZhkWXV1dmb5u82trawMAQ9dCaWnpunXrDM+uXLmS1GnCvLy8vLw8IrtGAwerC+G1a9d+//336dOnAwCXy/X39y8qKuro6CD71R8VFTVt2jQA0Ol0N27c2LNnz4cffmj8rWE2zHRCLS0t9vb2xu1KpbKjo8PBwcH8kXrT2dlJdtYqf3//xMRE4xaTD42IMWPGvPLKK4ZFUn3sAMDMcNTS0sIsjh49OjU1FQC0Wu3y5ctJpQKAwMBAk79XPvjgA0JZEDGsLoT5+fkAsHXr1q1bt8IfUzMWFxczF22SwszWxjz29PTkcrnr169vbGw0f+FhZsmpqqoyOSTduXNnVVXVp59+auY8vaFp+tKlS2SnrJJIJKQO3O9BIpEMkNmX3N3dKYr65Zdf/Pz8AEAoFDK/XYaRMKTY2NiYdP5zOGy/coKF2FsIaZpWKBQBAQHz5s0zNK5du1ahUJAthCaY/kkiZ5ukUun48eNzcnLkcrlhstn6+vqCggLjDjfiCgoKmpqacNa3gUwsFgcEBPz73/+Ojo42nrg4JyeHYCqEGOwthBUVFQ0NDa+99prh8AsA5HL5wYMHm5ubbWxsSAVramq6ePEiAOh0upqamszMTH9/f1L9bElJSe+8885bb70VHR3t6Oh4/fr13NzcIUOGmIxOM7OWlhbmI1KpVFVVVTk5Oc8888yIESMIRmpubi4vLzdpHDt2LPHrZQaOV1999e23316+fHlMTMzgwYNv3rz5/fffd3R0kO3TRgjYXAgLCwudnZ1NJjePiIjIzs4uKiqKjo4mFezEiRMnTpwAAIqipFJpSEiI8TGrmbm4uGzevDkjIyMrK6ujo8Pe3n7KlClz5swhewue0tLS0tJSAOByuU5OTvHx8WQLMwCUl5ffXQgPHjxI8LTcQOPs7JyampqZmblv37729nZbW9vAwMCEhIRPPvmEdDTEdjgNE0IIIVbD08IIIYRYDQshQgghVsNCiBBCiNWwECKEEGI1LIQIIYRYDQshQgghVsNCiBBCiNWwECLzWbRoEUVR1dXVxo11dXUjRoygKGrDhg39t+vhw4fHxMT03/Z74+vra3LTBoTQQMPeO8uggaC2tlYul1++fDk1NXXZsmWk4/Q9Dw8P0hEQQveBhRARU1NTI5fLq6ur09PTX331VdJxeqBSqZj5gx5Zbm5uX4VBCPUT7BpFZFy9ejU8PPzq1au7d+82qYI//PBDZGSklZWVSCQKCws7duwY015bW0v15PDhw8wKOTk5kydPtra2lkqlEyZMSEtL6/EOgnq9PiEhgcPhbN++/e5ng4KCpkyZcvr0aX9//5EjRzKNlZWVs2bNsrW1FYlEoaGhhw4dMn7J4cOHfX19LS0tvb29t23b9u677xqmtDTpGu3trTFrPvvss8ePHw8ODhYIBHZ2dgsXLuzo6HjYDxYh9LDwiBARcOXKFblcfv369QMHDpjcL/vo0aMzZ8708vJasWIFRVHZ2dlRUVFpaWnJyclSqfSjjz4yrNne3r5+/Xoej8fMO5GVlRUXFxcSErJ27VqKonJzc1NSUlQq1Ztvvmm8fZqmFyxYkJmZ+dlnnyUlJfUYr7q6OjIyMjIy8qWXXgKAn3/+OTw8XCQSJScnW1tbHzlyZNasWf/4xz9WrFgBAJmZmYmJiT4+Pu+9915DQ8Py5ct7m07hHm+NWaGkpCQhIWH16tVr1qw5cuRIenq6WCzesmXLo3/QCKEHQSNkLgsXLgSAkydPurm5AUBISIherzdeQa1WDx482MPDo7293dDi7+8vEolu3bplsrXY2FgA+PLLL5nF6OhoFxcXjUbDLGo0GgcHh+eee45ZHDZs2IwZM/R6PXP0+cknn/QWMjAwEAA+/fRTQ8uECRNcXV0bGxuZRZ1ON3XqVEtLy5aWFpVKZW9v7+Pj09XVxTx75swZiqJEIhGzOG7cuHHjxj3IWxs3bhwAnDp1yrDfMWPGuLm5PdAnixB6DNg1iswtNjb21q1bvr6+P/744+bNm42fUigU169fT0lJMXQtWlhYJCUldXZ25ufnG6/5z3/+89ChQwsXLly0aBHTkpqaWlRUZJgfqr6+vru7u7u72/hVS5cu/fLLL9esWWNymGhCLBanpKQwjy9dulRaWhoXF2dpadnR0dHR0dHV1ZWQkKBSqc6cOfPDDz80NTUlJSUJBAJm/cDAQLlcfvc2H+StyWQy49eOGjXqxo0b98iJEOoT2DWKzI3D4Zw8edLV1XXs2LGrVq2Sy+Xjx49nnqqqqgIAwyLDy8sLAOrq6gwt+fn5q1evDggI+Pzzzw2NHh4e2dnZ77//fmVl5dWrVxsbG00mxT1+/LharQaAioqKeyd0dXXlcrnM48rKSgDYtGnTpk2bTFarr6/v7OwEgNGjRxu3e3t7nz171mTlB3lrdnZ2xs/yeDyNRnPvqAihx4eFEJnbt99+GxQUBABpaWnx8fFxcXFlZWUikQgAmAM4kws1mWJjmOG2trZ2zpw5Uqn04MGDgwYNMqw2d+7cjIyMCRMmhIWFeXl5BQUFmQwcVKvVGzZsqKqq2r1799GjRx9w7mWmFL3xxhtMT6wxLy+vrKws42wMDqeHjpYHeWs4nT1CRGAhRObGnCAEgLi4uO+++y4jI2Pp0qU7duwAAHd3dwC4ePFiSEiIYf2ffvoJAEaNGgUA3d3ds2bNunnz5vHjx2UymWGd3377LSMjIzY29uDBg4ZGpVJpvN9p06atWrXqxo0bhw8fXrJkydSpU5nqe2/MXjQaTWhoqKHx3Llzp06d8vLycnR0BICamhrjwOfOnbt7O/d9awghUvAcISLp888/l8lkO3fu3L9/PwBEREQIhcLPPvusq6uLWUGpVO7cudPNzW3SpEkAsGzZsrNnz65bty4yMtJ4O83NzQDAXG/CyMvLa2hoMF6HORpzdnb++9//XlNTs3bt2gdJ6Ofn5+bmtm/fPsPWlEplUlLShg0brKysQkNDORzOzp076T/GaZw5c6aoqOju7dz3rSGESMEjQkSStbX1nj17pkyZkpycHBQUNHTo0I0bN77++uuBgYGxsbFCoTAjI6O6ujo3N5fL5R44cGDbtm02NjYURW3cuNGwkbCwMH9/f09Pzy1btjg6Ojo7OxcUFOzdu9fFxaW6urqwsDA8PNx4p6+x1zodAAABvklEQVS//vrXX3+9efPml19+eezYsfdOyOPx0tLSYmJiJkyYkJCQYGFhkZ2dXVVVlZubKxAIhgwZkpKSsnXr1ujo6JiYmNraWibD3eP/bGxs7vHW+urzRAg9CtKXrSIWYYZPXLlyxaR91apVABAcHMwMfsjMzPTz8xs0aJBEIpHL5Xl5ecxqK1eu7PF3eN26dTRNV1VVPfPMM2Kx2N7efs6cOZWVlbt27RKLxXK5nP5j+IRhj3l5ecweTcZv0DQdGBjo7e1t0qhQKORyuUAgsLa2Dg8PP3HihOEpnU730UcfeXp68vl8Dw+PLVu2zJ8/39HRkXnWMHyC0dtbY9b09/c33mliYiL+D0XIDCi6p1tvIIQe2fPPP19bW9vjmUKE0ACE5wgRenSdnZ3Ozs6GQYcAUF9fX1RUNHXqVIKpEEIPBc8RIvToRCKRXC5PT0+naTowMLCpqWnbtm08Hu/tt98mHQ0h9KCwaxShx6JUKjds2LB///5r164xZxDXr1/PjJRHCD0RsBAihBBiNTxHiBBCiNWwECKEEGI1LIQIIYRYDQshQgghVsNCiBBCiNWwECKEEGI1LIQIIYRY7f8BHoUgKOCaEugAAAAASUVORK5CYII=",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 180,
       "width": 300
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "options(repr.plot.width = 5, repr.plot.height = 3)\n",
    "\n",
    "data %>% \n",
    "  group_by(strain) %>%\n",
    "  summarize(mean_yfp = mean(yfp), mean_rfp = mean(rfp)) %>%\n",
    "  mutate(mean_ratio = mean_yfp / mean_rfp) %>%\n",
    "  left_join(annotations, by = \"strain\") %>%\n",
    "  ggplot(aes(x = kozak_region, y = mean_ratio, \n",
    "             color = insert_sequence, group = insert_sequence)) +\n",
    "  geom_line() + \n",
    "  geom_point() +\n",
    "  theme_classic() +\n",
    "  labs(x = \"Kozak region\", y = \"Mean YFP / RFP ratio\", color = \"Insert\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Re-shape data using `pivot_longer` and `pivot_wider`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "\u001b[36m──\u001b[39m \u001b[1m\u001b[1mColumn specification\u001b[1m\u001b[22m \u001b[36m────────────────────────────────────────────────────────\u001b[39m\n",
      "cols(\n",
      "  strain = \u001b[31mcol_character()\u001b[39m,\n",
      "  yfp = \u001b[32mcol_double()\u001b[39m,\n",
      "  rfp = \u001b[32mcol_double()\u001b[39m,\n",
      "  replicate = \u001b[32mcol_double()\u001b[39m\n",
      ")\n",
      "\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[90m# A tibble: 74 x 4\u001b[39m\n",
      "   strain    yfp   rfp replicate\n",
      "   \u001b[3m\u001b[90m<chr>\u001b[39m\u001b[23m   \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m     \u001b[3m\u001b[90m<dbl>\u001b[39m\u001b[23m\n",
      "\u001b[90m 1\u001b[39m schp677  \u001b[4m4\u001b[24m123 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m661         1\n",
      "\u001b[90m 2\u001b[39m schp678  \u001b[4m4\u001b[24m550 \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m437         1\n",
      "\u001b[90m 3\u001b[39m schp675  \u001b[4m3\u001b[24m880 \u001b[4m2\u001b[24m\u001b[4m1\u001b[24m323         1\n",
      "\u001b[90m 4\u001b[39m schp676  \u001b[4m2\u001b[24m863 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m668         1\n",
      "\u001b[90m 5\u001b[39m schp687  \u001b[4m4\u001b[24m767 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m995         1\n",
      "\u001b[90m 6\u001b[39m schp688  \u001b[4m1\u001b[24m274 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m927         1\n",
      "\u001b[90m 7\u001b[39m schp679  \u001b[4m2\u001b[24m605 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m840         1\n",
      "\u001b[90m 8\u001b[39m schp680  \u001b[4m1\u001b[24m175 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m902         1\n",
      "\u001b[90m 9\u001b[39m schp681  \u001b[4m3\u001b[24m861 \u001b[4m2\u001b[24m\u001b[4m0\u001b[24m659         1\n",
      "\u001b[90m10\u001b[39m schp683  \u001b[4m9\u001b[24m949 \u001b[4m2\u001b[24m\u001b[4m5\u001b[24m406         1\n",
      "\u001b[90m# … with 64 more rows\u001b[39m\n"
     ]
    }
   ],
   "source": [
    "data <- read_tsv(\"data/example_dataset_4.tsv\") %>% \n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How do you plot a histogram of both `yfp` and `rfp` as separate panels?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>%\n",
    "  pivot_longer(c(\"yfp\", \"rfp\"), names_to = \"channel\", values_to = \"value\") %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data %>%\n",
    "  pivot_longer(c(\"yfp\", \"rfp\"), names_to = \"channel\", values_to = \"value\") %>%\n",
    "  pivot_wider(names_from = \"channel\", values_from = \"value\") %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`pivot_longer` is useful for handling distinct measurements without listing them explicitly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 180,
       "width": 300
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_data <- data %>%\n",
    "  pivot_longer(c(\"yfp\", \"rfp\"), names_to = \"channel\", values_to = \"value\")\n",
    "\n",
    "plot_data %>%\n",
    "  ggplot(aes(x = value)) +\n",
    "  facet_wrap(~ channel) +\n",
    "  geom_histogram(bins = 30) +\n",
    "  theme_bw()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use `purrr::map` and `purrr::unnest` to read and work with tables from many files\n",
    "\n",
    "We use `list.files` and `enframe` to go from a list of files to a `tibble` quickly. Also check out `deframe`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "vscode": {
     "languageId": "r"
    }
   },
   "outputs": [],
   "source": [
    "data <- list.files(\"data/channel_data/\", full.names = T) %>%\n",
    "  enframe(\"number\", \"file\") %>%\n",
    "  mutate(data = map(file, read_tsv)) %>%\n",
    "  unnest(data) %>%\n",
    "  mutate(channel = str_extract(file, \"[^/]+(?=.tsv)\")) %>%\n",
    "  select(-number, -file) %>%\n",
    "  print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# In-class exercises\n",
    "\n",
    "(20 min)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Re-shape data read above from two separate files to create columns for `yfp` and `rfp`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Calculate ratio of YFP to RFP"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Calculate the `mean` and `standard error` of the ratio between replicates"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Combine data with annotations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Plot the mean and standard deviation of the YFP-RFP ratio as a function of the Kozak region.\n",
    "\n",
    "Hint: You have to use `geom_errorbar()` in addition to `geom_point()` and `geom_line()`."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "4.0.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}