SUPPLEMENTARY DATA S1
Structural model of mature rSpVaP and reproduction of Figure 7
=================================================================

This package contains the AlphaFold 3 output for the mature SpVaP
protein, the secondary-structure assignment derived from it, and the
scripts used to generate the panels of Figure 7.


1. CONTENTS
-----------

SpVaP_mature.fasta
    Amino-acid sequence of the mature protein used as input for the
    structure prediction (132 residues, Asp1-Thr132). This is the
    sequence remaining after removal of the 20-residue signal peptide
    predicted by SignalP 5.0 from the 152-residue precursor encoded by
    transcript SIMPER PX899338.

SpVaP_mature_AF3_model0.pdb
    Top-ranked AlphaFold 3 model, converted from mmCIF to PDB format.
    Chain A, residues 1-132, 1076 atoms.
    NOTE: following the AlphaFold convention, the B-factor column
    contains per-atom pLDDT values, NOT crystallographic temperature
    factors. High values indicate high model confidence.

SpVaP_mature.dssp
    Secondary-structure assignment calculated with mkdssp 4.2.2 from
    the coordinates above, in legacy DSSP format.

reproduce_Fig7c.pml
    PyMOL script that regenerates the two molecular views of panel c.
    Self-contained: the DSSP assignment and the camera matrices are
    embedded in the script.

reproduce_Fig7d.py
    Python/Matplotlib script that regenerates the secondary-structure
    topology diagram of panel d. Self-contained.

AlphaFold3_output/
    Unmodified output of the AlphaFold 3 prediction, comprising:
      fold_spvap_job_request.json ......... input sequence and seed
      fold_spvap_model_0..4.cif ........... the five sampled models
      fold_spvap_summary_confidences_*.json  pTM, ranking score,
                                            fraction disordered,
                                            clash flag, per model
      fold_spvap_full_data_0.json ......... PAE matrix, per-atom pLDDT,
                                            contact probabilities
      msas/ ............................... multiple sequence alignments
      templates/ .......................... structural templates
                                            retrieved by the prediction
      terms_of_use.md ..................... AlphaFold Server terms


2. HOW TO REPRODUCE FIGURE 7
----------------------------

Panel c (molecular views)
    Place reproduce_Fig7c.pml in the same folder as
    SpVaP_mature_AF3_model0.pdb, then run:

        pymol -cq reproduce_Fig7c.pml

    or, from within PyMOL:

        @reproduce_Fig7c.pml

    Outputs Fig7c_i.png and Fig7c_ii.png (2000 x 2000 px, 300 dpi).
    The script prints the PyMOL version used.

Panel d (topology diagram)
    Requires Matplotlib (pip install matplotlib). Run:

        python reproduce_Fig7d.py

    Outputs Fig7d.png (400 dpi) and Fig7d.tif (600 dpi), 180 mm wide,
    and prints the beta-strand boundaries to the terminal.

Panel b (secondary-structure composition)
    The values plotted for the model are the proportions of each DSSP
    class over the 132 residues, grouped as follows:
        helix       H + G + I      3 residues     2.3 %
        beta-sheet  E + B         62 residues    47.0 %
        turn        T             23 residues    17.4 %
        unordered   S + P + coil  44 residues    33.3 %
    The values plotted for the experimental data are the CD
    deconvolution results reported in the manuscript.

Regenerating the DSSP file from the coordinates
        mkdssp SpVaP_mature_AF3_model0.pdb SpVaP_mature.dssp \
               --output-format dssp


3. SUMMARY OF THE MODEL
-----------------------

Input                       132 residues (mature protein)
pTM                         0.92 (identical across all five models)
Ranking score               0.92
Mean pLDDT                  94.7  (min 75.8; max 98.6)
Residues with pLDDT < 70    0 of 132
Fraction disordered         0.00
Steric clashes              none
Pairwise Ca RMSD (5 models) 0.07 - 0.15 A

Secondary structure (DSSP)  12 beta-strands; one three-residue 3-10
                            turn (Asn41-Gln43); no alpha-helix
Disulfide bond              Cys8 - Cys130 (S-gamma to S-gamma 2.03 A)
Fold                        beta-trefoil

Structural templates retrieved by the AlphaFold prediction:
    PDB 3PG0  ThreeFoil (designed beta-trefoil)   - three hits
    PDB 3VSF  Ricin B lectin                      - one hit


4. SOFTWARE VERSIONS
--------------------

AlphaFold 3     (AlphaFold Server)
mkdssp          4.2.2
PyMOL           3.2 (open-source build)
Matplotlib      3.10
Python          3.12


5. REFERENCES
-------------

Abramson J, et al. Accurate structure prediction of biomolecular
    interactions with AlphaFold 3. Nature 2024;630:493-500.
    doi:10.1038/s41586-024-07487-w

Kabsch W, Sander C. Dictionary of protein secondary structure: pattern
    recognition of hydrogen-bonded and geometrical features.
    Biopolymers 1983;22:2577-2637. doi:10.1002/bip.360221211

Hunter JD. Matplotlib: a 2D graphics environment. Computing in Science
    & Engineering 2007;9(3):90-95. doi:10.1109/MCSE.2007.55

Schrodinger LLC. The PyMOL Molecular Graphics System.
