Supplementary Dataset for "Structural and Biophysical Basis for PFAS Binding by Human Sterol Carrier Protein-2"

Boltz-2 co-folding predictions of human SCP2 (C-terminal domain residues 425–547 of the SCP2 gene product, UniProt P22307) bound to different PFCAs, including PFDA, PFDoA, PFHxDA, PFOS, PFTeDA, PFTrDA and PFUnDA, as well as Palmitate as a positive control to idenfity the binding pocket in human SCP2 in comparison with the structure of Aedes aegypti SCP2 bound to palmitate (PDB ID: 2KSI). For the protein-ligand complex predictions, two sets of parameters were employed for all cases. The first default set was performed with 3 recycling steps (--recycling_steps = 3) and 5 diffusion samples (--diffusion_samples = 5). The second exhaustive (r10_d25) approach utilized 10 recycling steps (--recycling_steps = 10) and 25 diffusion samples (--diffusion_samples = 25). All predictions were executed using a Multiple Sequence Alignment (MSA) obtained from the MMseqs2 server with the --use_msa_server flag. The generated outputs are organized into the output directories following the Boltz-2 output structure, where [parameters] correspond to either default or r10_d25, [ligand] correspond to Palmitate, PFDA, PFDoA, PFHxDA, PFOS, PFOSx2 (predictions with 2 PFOS molecules), PFTeDA, PFTrDA and PFUnDA, and [number] corresponds to the number of diffusion samples (5 for default and 25 for r10_d25): boltz_results_humanSCP2_[ligand]_[parameters]/ ├── lightning_logs/# Logs generated during training or evaluation ├── msa/# Contains the MSA generated automatically via the mmseqs2 server ├── predictions/# Contains the model's predictions ├── humanSCP2_[ligand]_[parameters]/ ├── affinity_humanSCP2_[ligand].json # The affinity scores (affinity_pred_value, affinity_probability_binary, affinity_pred_value1, affinity_probability_binary1, affinity_pred_value2, affinity_probability_binary2) ├── confidence_humanSCP2_[ligand]_model_[number].json # The confidence scores (confidence_score, ptm, iptm, ligand_iptm, protein_iptm, complex_plddt, complex_iplddt, chains_ptm, pair_chains_iptm) ├── humanSCP2_[ligand]_model_[number].cif # The predicted structure in CIF format, with the inclusion of per token pLDDT scores ├── pae_humanSCP2_[ligand]_model_[number].npz # The predicted PAE score for every pair of tokens ├── pde_humanSCP2_[ligand]_model_[number].npz # The predicted PDE score for every pair of tokens ├── plddt_humanSCP2_[ligand]_model_[number].npz # The predicted pLDDT score for every token └── plddt_humanSCP2_[ligand]_model_[number].npz # The predicted pLDDT score for every token └── processed/# Processed data used during execution