These are some functions for doing the Internal Reference Standard / Sample Loading / TMM (EdgeR) Normalization.

The main function is sl_irs_tmm_normalization(), which also does the plotting showing how ion intensities shift by channel for each step.

The three inputs are 1) protein_df, which is the output of protein intensities. If you have multiple injections, perhaps the average of intensities should be taken beforehand. 2) tmt_exp_columns, which is a list. The list should be structured like this: list(c(1, 2, 3), c(4, 5, 6)) where the numbers refer to the columns in protein_df separating different TMT experiments. The last thing is tmt_common_channel_names, which designates the names of internal reference standards within each TMT experiment.

Note that the original code was modified (into functions) from an excellent guide by Phil Wilmart (https://github.com/pwilmart/IRS_normalization). The main thing these scripts do is wrap into functions, allowing a variable number of TMT experiments.

The other key difference is that we do not take the sum of IRS channels, rather the mean. This is in case there are different numbers of IRS channels (if for example you have one TMT experiment with just one IRS channel).