Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: Coevolutionary analyses require phylogenetically deep alignments and better null models to accurately detect inter-protein contacts within and between species

Fig. 3

Null distributions for coevolution statistics differ in their control of the false positive rate (FPR). Nominal FPRs for a given target FPR 0.1 % (Dashed orange line) are shown for the HisKA-RR sub-alignments. Left: Nominal FPRs using the empirical distribution of score ranks as the null distribution (i.e. using P empirical ). Right: Nominal FPRs assuming standardized scores have a standard normal null distribution (i.e. using P normal ). Blue lines indicate a loess fit for each method, 95 % confidence intervals are shown in gray. See Abbreviations and Table 1 for abbreviations

Back to article page