SCASA: Shape Complementarity and Available Surface Area

SCASA is a Python package for calculating two geometric properties of protein complexes from PDB files: Shape Complementarity (SC) and Available/Buried Surface Area (ASA/BSA). It can be used as a standalone command-line tool or imported as a Python module.

---

What is Shape Complementarity?

Shape Complementarity (SC) (Lawrence and Colman, 1993) is a measure of the geometric "goodness of fit" between two protein interfaces. It ranges from 0 to 1, where values closer to 1 indicate tightly complementary surfaces. It has been widely used to characterise the quality of antibody/antigen and T-cell receptor/antigen interfaces.

SC is computed by:

1. Selecting the interface atoms from each complex (those within a distance threshold of the opposing surface)
2. Generating surface dots by randomly sampling a ConvexHull triangulation of each interface
3. Estimating a surface normal at each dot via PCA on its 10 nearest dot-cloud neighbours
4. For each dot on surface A, finding the nearest dot on surface B and computing the dot product of their normals
5. The final SC score is the mean of the medians of S(A→B) and S(B→A)

Comparison with CCP4 SC

SCASA implements the same Connolly molecular surface algorithm as CCP4 SC, translated directly from the original Fortran mds subroutine (Copyright Michael Connolly, 1986). The surface generation produces the same three types of surface dots:

• Convex — contact surface on each atom's VdW shell
• Toroidal — probe rolling between two adjacent atoms
• Concave — re-entrant patch where the probe nestles between three atoms

SCASA uses the same radii file (sc_radii.lib), probe radius (1.7 Å), dot density (15/Ų), and weight factor (0.5) as CCP4 SC. In place of the CCP4 trim band (which removes peripheral buried dots near accessible dots), SCASA uses an equivalent 1.5 Å inter-surface distance filter, which produces the same practical effect.

At default settings, SCASA gives SC ≈ 0.54 for 1FYT vs 0.56 from CCP4 SC — a difference of ~0.02, within the expected variation from floating-point differences and random surface sampling. Scores are directly comparable to CCP4 SC values.

To match CCP4 SC defaults exactly, use:

SCASA sc --pdb structure.pdb --complex_1 DE --complex_2 ABC --dot-density 15

What is (Buried or Available) Surface Area?

Buried Surface Area (BSA; reviewed by Ali et al., 2014) measures the total surface area (in Ų) of a complex that becomes buried upon binding to another complex. Available Surface Area (ASA) is the reciprocal — the total surface area remaining accessible to solvent.

Accessibility is defined relative to the solvent-accessible surface area (SASA), which imagines a probe sphere (representing a solvent molecule) rolling along the protein surface. Any region where the probe can pass unimpeded is considered accessible.

SCASA calculates ASA using the Shrake-Rupley algorithm implemented in Biopython. BSA is then derived by comparing the ASA of each complex in isolation against its ASA when bound.

---

Installation

Requires Python ≥ 3.9.

pip install .

---

Usage

Command-line tool

SCASA provides two subcommands: sc for shape complementarity and asa for surface area.

Common arguments (both subcommands)

Flag	Short	Required	Description
--pdb	-P	Yes	Path to PDB file of the complex
--complex_1	-C1	Yes	Chains of the first complex (e.g. DE or ABC). Multiple chains must be supplied as a single concatenated string
--complex_2	-C2	No	Chains of the second complex. Defaults to all remaining chains in the PDB file
--verbose	-v	No	Print additional progress messages

SCASA sc — Shape Complementarity

Flag	Short	Default	Description
--distance	-D	8.0	Interface cutoff in Å. Atoms with no neighbour within this distance of the opposing surface are excluded
--dot-density	-Dd	1.5	Surface dot sampling density (dots per Ų of interface area)
--plot	-pl	—	Generate a histogram plot of the SC function distribution

Example:

SCASA sc --pdb test/data/1FYT.pdb --complex_1 DE --complex_2 ABC

SCASA asa — Available/Buried Surface Area

Flag	Short	Default	Description
--level	-L	R	Granularity of output: S (whole complex), C (per chain), R (per residue), or A (per atom)

Example:

SCASA asa --pdb test/data/1FYT.pdb --complex_1 DE --complex_2 ABC --level R

---

Python module

SCASA can be imported and used directly in Python via the Complex class:

from scasa.scasa import Complex

complex = Complex(
    pdb_file="test/data/1FYT.pdb",
    complex_1="DE",
    complex_2="ABC",   # optional — defaults to all remaining chains
    distance=8.0,      # interface cutoff in Å
    density=1.5,       # dot sampling density per Ų
    verbose=True,
)

# Calculate shape complementarity
complex.sc()

# Calculate ASA/BSA (requires sub-PDB files written to tmp/)
complex.create_sub_pdbs()
complex.complex_sasa()

---

Interpreting SC scores

As a rough guide based on published literature:

SC score (SCASA)	Interpretation
0.70 – 0.80	Tightly complementary (e.g. antibody–antigen)
0.60 – 0.70	Typical protein–protein interface
0.45 – 0.60	Loosely packed or transient complex
< 0.45	Poor fit; may indicate a crystal contact rather than a biological interface

Note: these ranges apply when using the default density of 15 dots/Ų (--dot-density 15), which matches CCP4 SC. At the lower default density of 1.5, scores will be somewhat lower.

---

Contact

Tom Whalley