Fig. 3
From: Design of symmetric TIM barrel proteins from first principles
Summary statistics, energetics, and folding simulations for natural TIM barrels. a Summary statistics for a representative subset of TIM barrel structures derived from the CATH database. These statistics describe the residue composition and hydrogen bonding frequencies across different regions of the TIM barrel scaffold. Violin plots are used to visualize data distribution, while superimposed red histograms are used to visualize means. All p-values are calculated using the Welch two sample t-test. b,c,e,f General figure description: Right: Scatterplots describe the data distribution. A grey line divides the data into hydrophobic pores (<40 % polarity) and hydrophilic pores (≥40 % polarity). P values are depicted in green. All p-values are calculated using the Welch two sample t-test. Left: Violin plots depict the data distribution across hydrophobic and hydrophilic pores. b Variation of Rosetta energy scores against percentage pore polarity across all selected TIM barrel pores. Only pore residues are scored, and backbone hydrogen bonding terms are not included. c Variation of Rosetta energy scores against percentage pore polarity across all selected TIM barrel pores. Entire TIM barrel proteins are scored, and backbone hydrogen bonding terms are included. d The TIM barrel enzyme KLGP decarboxylase (PDB ID: 1kv8, chain A) is shown in white. The (αβ)3 foldon-like module chosen is shown in red. e Variation of Rosetta energy scores against percentage pore polarity for all selected (αβ)3 foldon-like modules of KLGP decarboxylase structural homologues. f Variation of the mean backbone RMSD (for the 5 ab initio modeled structures with the lowest Cα RMSD, compared to their (αβ)3 templates) against percentage pore polarity. This figure supports the application of the foldon hypothesis to TIM barrel proteins