search

MoseleyBioinformaticsLab / manuscript.peakCharacterization

0 stars 1 forks source link

highlight best method for sample based on RSD #21

Closed rmflight closed 2 years ago

rmflight commented 2 years ago

Given the current graphs, I think we might need to first filter and report things by all peaks, and then peaks present in 95%+ of scans before we report the final results in the table.

And then highlight the best method for a sample in the graph.

rmflight commented 2 years ago

So, my solution was to leave the graphs to speak for themselves.

However, I did modify the table to put everything in the same order as the graphs, a horizontal line to separate each sample from the other, and bolded the fitlersd numbers to make them easier to compare to, and then bolded any other samples numbers that I thought were equivalent or better than filtersd (and the identical compatriots in doublenorm and singlenorm_int)

rsd_plot_table