Learning how to graph flow cytometry data in R.... day one

I found this very interesting tutorial here:
http://bioinformatics.ca//files/public/Flow_2013_Module2.mp4

It goes through some of the key concepts to analyse flow cytometry data in R.

These include:

• vectors
• binding two vectors into a matrix
• lists
• mylist <- list(“first” = x, “second” = y)
• x and y are vectors that were defined earlier. 
• Thus the list here contains vectors. 
• INTERESTING AND IMPORTANT
• defining within a list uses double square brackets:
• mylist[[1]]
• can call a part of a list by it’s name!
• mylist[[“first”]]
• access by name rather than by index number can be very useful if multiple names are the same or if index numbers change.
• length(mylist)
• subsetting by index or name.
• Fundamental to R
• variables
• functions
• syntax of commands
• round brackets for functions; square brackets for subsetting
• data structures
• nest functions within functions - innermost brackets first.
• Google the error messages!
• warning messages rather than errors

Comment: I did this with Jo a little bit and made some notes but I don’t know where they are.

functions that are useful for subsetting for flow cytometry data:

• which - returns the index which adhere to the key concepts. 
• intersect
• union

Visualising flow cytometry data

• plot(density(a))

Key package - flowCore - it's a Bioconductor Package

Installing flowCore:
> source("http://bioconductor.org/biocLite.R") 

> biocLite("flowCore")

When we import an FCS file into R using read.FCS, it creates a special kind of object called a flowFrame. This seems to act like a kind of data frame with lots of metadata and the like. 

The data from this flowFrame can be converted into a matrix which can be useful. 

Another key package - flowViz

Installing flowViz:

> biocLite("flowViz")

this allows plotting of data but is quite slow!

Using "@" symbol to explore an object. 

I am having some problems with flowVis. It's giving an "Error: subset out of bounds" message. 

If I extract the data then I can plot the data nicely:

E <- exprs(data)

plot(E[,1], E[,2], 

     pch = ".", 

     ylim=c(0,1000000), 

     xlab="FSC-A",

     ylab="SSC-A",

     main="Forward Scatter vs Side Scatter") # this works

This gives a graph like this:

IT'S SLOW BUT IT IS PROGRESS...