Fig. 3: Proteins with native NCLEs misfold more often, biased towards NCLE regions.

a Odds ratio (Equation 6.1 in Supplementary information) between proteins containing non-covalent lasso entanglements (NCLEs) and misfolding. Shown is the contingency table, odds ratio between the presence of native NCLEs in a protein and that protein being non-refoldable (i.e., having one or more significant cut-sites), and the \(p\) value calculated from the Fisher Exact test. b Odds ratio between proteins native NCLE regions (\({n}_{{{{\rm{prot}}}}}=264\), \({n}_{{{{\rm{residues}}}}}=87,716\)) and significant cut-sites (indicating misfolding) determined by logistic regression. ***, **, * indicate, respectfully, conditions where the two-sided \(p\) value are below the levels of significance of 0.001, 0.01, 0.05. c Probability density function(s) of the solvent accessible surface area (SASA) for the native NCLE region of the proteome (blue) and the non-NCLE region (transparent with black outline) before Propensity Score Matching (left) and after (right). Propensity Score Matching controls for differences in solvent exposure by constructing a matched dataset of non-NCLE regions with similar SASA as native NCLE regions. d Same as (b) except calculated from the propensity score matched dataset (\({n}_{{{{\rm{prot}}}}}=264\), \({n}_{{{{\rm{residues}}}}}=68,886\)). Data are presented as mean values with 95% confidence interval.