Figure S5. Impact of counting statistics on the quantification of errors between replicates.

The reproducibility of inverse toeprinting was assessed using fully independent biological replicates. For each 3-aa motif, the fraction of the total number of reads between replicates that originated from replicates (A) NoAb1 and (B) Ery1 was calculated and binned according to the total number of reads. The SD of these fractions was calculated and plotted for each bin, as indicated by black points. The red curve indicates predicted SDs obtained using the same number of total reads for each bin, which had been randomly partitioned into two replicates for each sample, according to probabilities that were proportional to the total number of reads in each replicate. The close match between the predicted and measured SDs indicates that counting statistics are the main source of error. A threshold of 150 reads was chosen as the point where the SD drops below 0.05.