4-normalizatio_libexerciseCGAP.R

## ----setup, include=FALSE--------------------------------------------------------------------------------------------
options(htmltools.dir.version = FALSE)


## ----xaringan-themer, include=FALSE----------------------------------------------------------------------------------
library(xaringanthemer)
solarized_dark(
  code_font_family = "Fira Code",
  code_font_url    = "https://cdn.rawgit.com/tonsky/FiraCode/1.204/distr/fira_code.css"
)


## /* From https://github.com/yihui/xaringan/issues/147  */

## .scroll-output {

##   height: 80%;

##   overflow-y: scroll;

## }

## 
## /* https://stackoverflow.com/questions/50919104/horizontally-scrollable-output-on-xaringan-slides */

## pre {

##   max-width: 100%;

##   overflow-x: scroll;

## }

## 
## /* From https://github.com/yihui/xaringan/wiki/Font-Size */

## .tiny{

##   font-size: 40%

## }

## 
## /* From https://github.com/yihui/xaringan/wiki/Title-slide */

## .title-slide {

##   background-image: url(https://raw.githubusercontent.com/Bioconductor/OrchestratingSingleCellAnalysis/master/images/Workflow.png);

##   background-size: 33%;

##   background-position: 0% 100%

## }


## ----all_code, cache=TRUE--------------------------------------------------------------------------------------------
library('scRNAseq')
sce.zeisel <- ZeiselBrainData(ensembl = TRUE)

# Quality control
library('scater')
is.mito <- which(rowData(sce.zeisel)$featureType == "mito")
stats <- perCellQCMetrics(sce.zeisel, subsets = list(Mt = is.mito))
qc <-
  quickPerCellQC(stats,
                 percent_subsets = c("altexps_ERCC_percent", "subsets_Mt_percent"))
sce.zeisel <- sce.zeisel[, !qc$discard]


## ----all_code2, cache=TRUE, dependson='all_code'---------------------------------------------------------------------
# Library size factors
lib.sf.zeisel <- librarySizeFactors(sce.zeisel)

# Examine distribution of size factors
summary(lib.sf.zeisel)
hist(log10(lib.sf.zeisel), xlab = "Log10[Size factor]", col = "grey80")
ls.zeisel <- colSums(counts(sce.zeisel))
plot(
  ls.zeisel,
  lib.sf.zeisel,
  log = "xy",
  xlab = "Library size",
  ylab = "Size factor"
)

###Exercise

#These are my answers to 1 and 2
#Question 1: are ls.zeisel and lib.sf.zeisel equal?
identical(ls.zeisel, lib.sf.zeisel)
#R= FALSE

#Question 2: are they proportional?
proportion=lib.sf.zeisel/ls.zeisel
table(proportion)
table(lib.sf.zeisel/ls.zeisel)
table(ls.zeisel/lib.sf.zeisel)
#R. 14589.58. The important point is in the "table" command, this will allow me to see if the factor is common across all samples.
#If I only get one value, then that means that all elements in the vector have the same factor, so I confirm that they are proportional


#I couldnt complete this one but that was the idea which I asked you before ending class, so I copied it from your text
## ----exercise_solution, cache=TRUE, dependson='all_code'-------------------------------------------------------------
## First compute the sums
zeisel_sums <- colSums(counts(sce.zeisel))
identical(zeisel_sums, ls.zeisel)

## Next, make them have unity mean
zeisel_size_factors <- zeisel_sums/mean(zeisel_sums)
identical(zeisel_size_factors, lib.sf.zeisel)





## ----all_code3, cache=TRUE, dependson='all_code2'--------------------------------------------------------------------
# Normalization by convolution

library('scran')
# Pre-clustering
set.seed(100)
clust.zeisel <- quickCluster(sce.zeisel)
# Compute deconvolution size factors
deconv.sf.zeisel <-
  calculateSumFactors(sce.zeisel, clusters = clust.zeisel, min.mean = 0.1)

# Examine distribution of size factors
summary(deconv.sf.zeisel)
hist(log10(deconv.sf.zeisel), xlab = "Log10[Size factor]",
     col = "grey80")
plot(
  ls.zeisel,
  deconv.sf.zeisel,
  log = "xy",
  xlab = "Library size",
  ylab = "Size factor"
)


## ----all_code4, cache=TRUE, dependson='all_code3'--------------------------------------------------------------------
# Library size factors vs. convolution size factors

# Colouring points using the supplied cell-types
plot(
  lib.sf.zeisel,
  deconv.sf.zeisel,
  xlab = "Library size factor",
  ylab = "Deconvolution size factor",
  log = 'xy',
  pch = 16,
  col = as.integer(factor(sce.zeisel$level1class))
)
abline(a = 0, b = 1, col = "red")


## ----'reproducibility', cache = TRUE, dependson=knitr::all_labels()--------------------------------------------------
options(width = 120)
sessioninfo::session_info()