Figure 1
Statistics of sequence reassignment of randomly selected continuous protein chain fragments from benchmark-set MX structures. Grey and contoured histograms represent cases where newly assigned sequences match or differ, respectively, from the reference model for test fragments of (a) 10 and (b) 20 amino acids. The vertical dashed line depicts a standard threshold used by checkMySequence for outlier identification in cryo-EM models. The ordinate axes of the plots show −log(p-value); higher values correspond to lower p-values and more reliable sequence assignments. Frequency histograms are shown for clarity, but the sets presented in each panel are strongly unbalanced. The number of test fragments with reassigned sequences that do not match the reference model is 1% of the overall number of test fragments in the benchmark set. |