Project files provided as supporting information to the manuscript "Detailed insight in the chignolin folding process from maximally informative low-resolution representations of its isocommittor hypersurfaces"

Project files provided as supporting information to the manuscript "Detailed insight in the chignolin folding process from maximally informative low-resolution representations of its isocommittor hypersurfaces" July 20, 2026 Alessia Guadagnin Pattaro, Roberto Menichetti, Raffaello Potestio ================================== The repository is organized as follows: - DATA_ANALYSIS: contains all the jupyter notebook to reproduce all the plots reported in the paper. The folder is further divided into the following subfolders. - 0_DATA: contains the raw/processed data needed to run the notebooks - 2_C_3_B_FREE_ENERGY: calculation of free energy-related plots (Figures 4 and 9) - 3_A_1_MEOW_FULL_FOLDED_UNFOLDED: analysis of the results of mapping entropy optimization workflow applied on the full trajectory, and on the folded and unfolded trajectories separated by FONC (Figures 5, 6) - 3_A_2_PCA: PCA analysis on triplets of amino acids (Figure 7). Warning: it takes about ~40 minutes to completely run - 3_B_1_MEOW_E_20IH: analysis of the results of mapping entropy optimization workflow applied to the transition region (Figures 8, S3, S4) - 3_B_2_MSD_KMEDOIDS: application of K-medoids clustering to the MSD matrices. Warning: for notebooks 0 and 1 use the dedicated conda environment sklearn_extra, provided in the folder with install instructions. - 3_B_3_ih_1_ih_12_SUBSETS: analysis of the division of IH_1 and IH_12 into two clusters according to MSD and d(D3, G7) (Figures 11, S16, S17, S19, S20) - 3_B_4_FREE_ENERGY_PROGRESSION: free energy landscapes of the whole committor range and IH_0 (Figures S13, S14, S15) - 3_B_5_FONC_H_BONDS: FONC decomposition (FONC.ipynb, generating Figures 10 and S18) and analysis of H-bonds formed by Thr6 (hbonds_thr6.ipynb, generating Figure S5) - 3_B_6_AF_MUTANTS: mutational analysis of CLN025, it contains the AF data, the molecular dynamics simulations up to the NPT equilibration phase, and the RMSD analysis of the last frame of each simulation compared to the crystal (Figures 12, S21) - SI_ANALYSIS_FULLTRAJ: RMSF of chignolin trajectory (Figure S1) - SI_ISOCOMM_ANALYSIS: analyses corroborating the validity of the relevance profile for the 20 IH case, contains the distance maps (distance_maps folder, Figures S6 and S7), the PCA and IPR analysis (PCA_IPR folder, Figures S10, S11, S12), the FONC measured per residue (per_residue_FONC folder, Figure S9), the RMSF of the CA atoms (RMSF_CA folder, Figure S8), and the generation of the heatmaps for the case of MEOW applied to 10 IH, 20 IH, and 30 IH (relevance_heatmap_panel folder, Figure S2) - EXCOGITO_RESULTS: the raw data obtained after the optimization of mapping entropy on the systems where it was applied (the prefix refers to the paper section and the DATA_ANALYSIS folder where the data was processed): - 3_A_1_FONC_FOLDED - 3_A_1_FONC_UNFOLDED - 3_A_1_FULLTRAJ - 3_A_1_SI_RMSD_FOLDED - 3_A_1_SI_RMSD_UNFOLDED - 3_B_1_MEOW_E_20IH - 3_B_3_MEOW_E_ih_1 - 3_B_3_MEOW_E_ih_12 - SI_MEOW_E_10IH - SI_MEOW_E_30IH All the notebooks can be run with ease by installing a conda environment with the provided yml file in the following way: conda env create --file chignolin2026.yml Then select the environment named chignolin2026 when choosing a kernel for your notebook. For any inquiry or issue contact [e-mail] or [e-mail]