tail-tools analyse-tail-counts: produces a variety of CSV files as output.
read_count.csv contains read counts.
tail_count.csv contains counts of reads with tails. This is gives an indication of how accurate mean or median tail lengths are. Means and medians from low tail counts will have a high standard error.
tail.csv contains average tail lengths.
tail_sd.csv contains the standard deviation of tail lengths.
tail_quantile_50.csv contains the median tail lengths.
tail_quantile_25.csv and tail_quantile_75.csv provide the other quartiles, and can be used as a more robust measure of tail length variability than the standard deviation.
counts.csv contains various outputs, in a special Nesoni format. This was the original way analyse-tail-counts produced its results. If you've installed the Nesoni R component, it can be loaded in R with:
library(nesoni)
x <- read.grouped.table("counts.csv")
This turned out to be annoying way to do things. Hence the CSV files described above are now also created.