Fig. 2

Intrinsic disorder and compositional bias of C1CR-ICDs. a Disorder prediction for C1CR common receptors and group 1. The disorder propensity, ranging from 0 to 1 was predicted using IUPred2A (blue), ANCHOR (black) and Pondr-fit VSL2 (red) and is plotted as a function of residue number. The boundaries between the ECD (white background), TMD (yellow background) and ICD (orange background) were predicted using TMHMM v. 2.0. The sequence numbering includes the signal peptide. b Fractional differences in composition between the C1CR-ICDs (dark grey) or a set of IDPs (light grey), and a set of folded proteins calculated for each amino acid type (see methods for details). Negative values denote that the amino acid is less frequent than in folded proteins, and positive values denote that the amino acid is more frequent than in folded proteins. The arrows indicate the directions of “more” abundant than in folded proteins, and “less” abundant than in folded proteins. The error bars indicate the 90% confidence interval of the estimated frequencies, calculated using a per-protein bootstrapping procedure with 1000 iterations [105]. c IDDomainSpotter profiles of group 1 C1CR-ICDs. Profiles display scores for +Arg,Lys-Asp,Glu (green), +Cys (brown), +Pro (purple), + Ile,Leu,Val (red) and + Phe,Tyr,Gly (blue) calculated over 15 residue windows for each of the ICDs