---
title: "2 UMAP before Harmony"
author: "Yifang Liu"
date: "`r Sys.Date()`"
output:
  rmdformats::html_clean:
    code_folding: hide
    fig_width: 10
    fig_height: 10
    highlight: kate
    thumbnails: false
    lightbox: true
    gallery: true
---

```{r knitr_init, echo=FALSE, cache=FALSE}
library(knitr)
library(rmdformats)

options(max.print = 200)
opts_chunk$set(echo = TRUE,
               cache = FALSE,
               prompt = FALSE,
               tidy = TRUE,
               comment = NA,
               message = FALSE,
               warning = FALSE,
               dev = c('png', 'pdf'),
               fig.width = 10,
               fig.height = 10,
               fig.align = "center",
               fig.path = '2_PDF_2020-02-05_UMAP_before_Harmony/',
               dpi = 72)
opts_knit$set(width = 75)
```

```{r setup}
set.seed(123)

npc <- 20
# theta1 <- 2
# theta2 <- 5
# theta <- c(theta1, theta2)
resolution <- 0.1
pt_size <- 1
# alpha <- 0.8

# Suppress loading messages
suppressPackageStartupMessages({
  library(Matrix)
  library(dplyr)
  library(tidyverse)
  library(Seurat)
  library(cowplot)
  library(Rcpp)
  library(harmony)
  library(SoupX)
})
```

```{r UMAP}
EGFP <- readRDS("Data/2020-02-05_EGFP_seurat_obj_before_Harmony.Rds")
object <- EGFP
dims <- c(1, 2)
reduction <- "umap"
cells <- colnames(x = object)
data <- Embeddings(object = object[[reduction]])[cells, dims]
data <- as.data.frame(x = data)
dims <- paste0(Key(object = object[[reduction]]), dims)
object[['ident']] <- Idents(object = object)
group_by <- "ident"
data[, group_by] <- object[[group_by]][cells, , drop = FALSE]
data[, "LibraryID"] <- object[["LibraryID"]][cells, , drop = FALSE]
data_G0 <- subset(data, LibraryID == "G0")
data_G1 <- subset(data, LibraryID == "G1")
# group_color <- c("#0000EE","#9d009d","#ff7f0e","#ff0078","#05e259","#35bbf8","#c4af00","#686864","#9467bd","#006c00","#1b8c8b","#8d532e","#9f5084","#f7b6d2","#7f7f7f","#c7c7c7","#bcbd22")
```

# UMAP plots of combined and separated G0 and G1 {.tabset}

## UMAP combined with legend

```{r UMAP_combined_with_legend}
# range(data$UMAP_1)
# range(data$UMAP_2)
ggplot(data = data) +
    geom_point(
      mapping = aes_string(
        x = dims[1],
        y = dims[2],
        color = "ident"
      ),
      shape = 16,
      size = pt_size
    ) +
    # scale_color_manual(values = alpha(group_color, alpha)) +
  coord_cartesian(xlim = c(-10, 5), ylim = c(-14, 8)) +
  theme_cowplot()
```

## UMAP combined

```{r UMAP_combined}
ggplot(data = data) +
    geom_point(
      mapping = aes_string(
        x = dims[1],
        y = dims[2],
        color = "ident"
      ),
      shape = 16,
      size = pt_size
    ) +
    # scale_color_manual(values = alpha(group_color, alpha)) +
  coord_cartesian(xlim = c(-10, 5), ylim = c(-14, 8)) +
  theme_cowplot() +
  theme(legend.position = "none")
```

## G0

```{r UMAP_G0}
data <- data_G0
ggplot(data = data) +
    geom_point(
      mapping = aes_string(
        x = dims[1],
        y = dims[2],
        color = "ident"
      ),
      shape = 16,
      size = pt_size
    ) +
    # scale_color_manual(values = alpha(group_color, alpha)) +
  coord_cartesian(xlim = c(-10, 5), ylim = c(-14, 8)) +
  theme_cowplot() +
  theme(legend.position = "none")
```

## G1

```{r UMAP_G1}
# Do some extra work because the number of cluster in G0 is different from
# the number of cluster in G1
gg_color_hue <- function(n) {
  hues = seq(15, 375, length = n + 1)
  hcl(h = hues, l = 65, c = 100)[1:n]
}
n <- 10
cols <- gg_color_hue(n)
cols <- cols[c(1, 2, 3, 4, 5, 8, 9)]
data <- data_G1
ggplot(data = data) +
    geom_point(
      mapping = aes_string(
        x = dims[1],
        y = dims[2],
        color = "ident"
      ),
      shape = 16,
      size = pt_size
    ) +
  scale_color_manual(values = cols) +
  coord_cartesian(xlim = c(-10, 5), ylim = c(-14, 8)) +
  theme_cowplot() +
  theme(legend.position = "none")
```

# Notes

2020-02-05:

  * UMAP before Harmony.

Sun Dec 1, 2019:

  * UMAP plots (after SoupX) of combined and separated EGFP and TSC.

Tue Oct 29, 2019:

  * use SoupX fixed 0.45 to remove ambient RNA.

Mon Oct 7, 2019:

  * Add more sequence depth.

Mon, Sep 30, 2019:

  * remove genes: EGFP, Tsc1, gig. Then perform integrate analysis of EGFP, TSC1.

Fri, Sep 20, 2019:

  * First version for integrate analysis of EGFP, TSC1.

# Session Info

```{r sessioninfo, message=TRUE}
sessionInfo()
```