I am new to the world of fold change and gene expression in general.
I've been tasked with creating a volcano plot comparing 2 separate groups of individuals. Let's call them Group S and Group H. Each group has between 80-200 individuals, and each individual has all the genes.
To obtain fold change for each gene between groups, I have been calculating it as:
{log_2(mean(gene_expression in group S))}/{log_2(mean(gene_expression in group H))}
The gene_expression value is coming from values obtained using the SST-RMA algorithm in TAC software, a method summarized briefly on page 213 of the User Manual. Because our .CEL files don't have the group assignment data, I pull the expression values from the software, add them to our .csv dataset, and calculate the fold change in R using the above equation.
I am not sure if it is truly appropriate to use this SST-RMA value to calculate foldchange like this, and if so, how to go about obtaining a sensible cutoff for up and downregulated. Can anyone recommend an approach or a resource to consult? Thanks in advance for any help you can provide.
You should be using specialized methods such as limma (R/Bioconductor) rather than doing manually for what established software exists.