RSA, ASA, and BSA in Protein Structural Analysis

A concise guide for notes and quick reference

---

What are ASA, RSA, and BSA?

1. ASA — Accessible Surface Area

Definition
The surface area of an atom, residue, or protein that is accessible to a solvent molecule (typically modeled with a 1.4 Å water probe).

Key properties

• Unit: Ų
• Absolute value (not normalized)
• Depends on structure and environment

Interpretation

• Large ASA → exposed residue
• Small ASA → buried residue

---

2. RSA — Relative Surface Accessibility

Definition
ASA normalized by the maximum possible exposure of that residue type.

$$
RSA = \frac{ASA_{observed}}{ASA_{max}}
$$

Key properties

• Unitless (0–1 typically)
• Comparable across residue types

Typical thresholds

• RSA < 0.2 → buried
• RSA ≥ 0.2 → exposed

---

3. BSA — Buried Surface Area

Definition
Surface area lost when two molecules form a complex.

$$
BSA = ASA_{monomers} - ASA_{complex}
$$

Key properties

• Measures interface size
• Unit: Ų

Interpretation

• Larger BSA → more extensive interaction

---

Why do we care about them?

Structural biology

ASA and RSA are fundamental descriptors of protein architecture. Residues with low RSA values are typically buried in the hydrophobic core and contribute to folding stability, whereas highly exposed residues often participate in solvent interactions or molecular recognition. Early work linking solvent exposure to protein stability demonstrated that buried surface area correlates with folding energetics ^[1].

Protein–protein interactions

BSA is one of the most widely used quantitative measures of interface size. Larger interfaces are generally associated with more stable complexes, although geometry and chemistry also matter. Janin and Chothia showed that interface area correlates with interaction strength and specificity in protein complexes ^[2]. Today, BSA is routinely used in docking scoring and interface classification.

Antibody–antigen studies

In immunology, RSA and BSA provide complementary information for mapping epitopes and paratopes. Antigen residues that are solvent‑exposed in the unbound state (high RSA) but become buried upon binding (high BSA) are strong epitope candidates. Structural analyses of antibody–antigen complexes consistently show that epitope residues are enriched in exposed positions prior to binding ^[3].

Bioinformatics and machine learning

ASA, RSA, and BSA are widely used as features in computational models for predicting binding sites, interaction partners, and mutational effects. Many interface prediction algorithms rely on solvent accessibility as a primary signal because exposed residues are more likely to participate in interactions. Reviews of structural bioinformatics methods highlight solvent accessibility as a core feature in predictive pipelines ^[4].

---

Utility cases

Epitope definition

A common structural workflow for epitope mapping involves combining exposure and burial metrics:

1. Compute RSA on the antigen in the unbound state.
2. Compute BSA in the antigen–antibody complex.
3. Classify residues with high RSA and significant BSA as epitope residues.

This approach underlies many structural epitope databases and benchmarking studies ^[5].

Docking validation

Interface area is frequently used as a sanity check for docking models. Complexes with extremely small BSA values are often artifacts, whereas biologically meaningful interfaces typically exceed ~1000 Ų for protein–protein interactions. CAPRI assessment studies use interface metrics including BSA to evaluate docking quality ^[6].

Mutation analysis

Solvent accessibility provides context for interpreting mutational effects. Mutations at buried sites (low RSA) tend to disrupt packing and stability, while mutations at interface residues (high BSA) often alter binding affinity. Large-scale mutagenesis studies have shown that structural environment, including solvent exposure, strongly predicts functional impact ^[7].

---

How to compute ASA, RSA, and BSA

Using DSSP (ASA + RSA)

Run DSSP:

# Install 
conda install -c salilab dssp

# structure repair
mkdssp structure.pdb structure.dssp

Sometimes mmCIF files cause DSSP issues due to:

• nonstandard residue names
• missing auth vs label chain IDs
• alternate conformations

conda install -c conda-forge gemmi
gemmi convert input.cif output.pdb

• ACC column → ASA (Ų)
• RSA = ASA / MaxASA

You can compute RSA using a reference table ^[8].

---

Using Biopython (RSA directly)

from Bio.PDB import PDBParser, DSSP

structure = PDBParser(QUIET=True).get_structure("X", "structure.pdb")
model = structure[0]
dssp = DSSP(model, "structure.pdb")

for key in dssp:
    chain, res_id = key
    rsa = dssp[key][3]
    print(chain, res_id[1], rsa)

---

Using PDBePISA (BSA)

Upload a structure to the PDBePISA web server or analyze assemblies provided by the PDB entry. The server computes interface areas, solvation energies, and assembly predictions.

Typical outputs include:

• Total interface BSA
• Per‑residue BSA
• Interface energetics and assembly stability estimates

Quick tutorial (based on the official PDBePISA guide)

1. Go to the PDBePISA web interface.

2. Upload a PDB/mmCIF file or enter a PDB ID.

3. Run the analysis to generate biological assembly predictions.

4. Open the “Interfaces” section to view:

   • Interface area (BSA)
   • Hydrogen bonds and salt bridges
   • Solvation energy gain

5. Export tables for downstream analysis if needed.

For a full step‑by‑step walkthrough, see the official tutorial PDF:
https://www.ebi.ac.uk/pdbe/sites/default/files/documents/service_tutorials/PDBePISA.pdf

---

Max ASA reference

RSA requires a reference maximum exposure value.
Commonly used dataset:

Tien et al., 2013 — PLoS ONE ^[8:1]
Maximum allowed solvent accessibilities of residues in proteins.

MAX_ASA_1LETTER = {
    "A":129, "R":274, "N":195, "D":193, "C":167,
    "Q":225, "E":223, "G":104, "H":224, "I":197,
    "L":201, "K":236, "M":224, "F":240, "P":159,
    "S":155, "T":172, "W":285, "Y":263, "V":174
}

MAX_ASA_3LETTER = {
    "ALA":129, "ARG":274, "ASN":195, "ASP":193, "CYS":167,
    "GLN":225, "GLU":223, "GLY":104, "HIS":224, "ILE":197,
    "LEU":201, "LYS":236, "MET":224, "PHE":240, "PRO":159,
    "SER":155, "THR":172, "TRP":285, "TYR":263, "VAL":174
}

---

Summary

Metric	Measures	Typical use
ASA	Absolute solvent exposure	Structural analysis
RSA	Relative exposure	Buried/exposed classification
BSA	Surface buried on binding	Interface analysis

---

1. Chothia, Cyrus. “Hydrophobic bonding and accessible surface area in proteins.” Nature 248 (1974): 338–339. ↩︎

2. Janin, Joël & Chothia, Cyrus. “The structure of protein–protein recognition sites.” J. Mol. Biol. 212 (1990): 403–408. ↩︎

3. Davies, David R. & Cohen, George H. “Interactions of protein antigens with antibodies.” Proc. Natl. Acad. Sci. USA 93 (1996): 7–12. ↩︎

4. Zhou, Huan-Xiang & Qin, Sanbo. “Interaction-site prediction for protein complexes: a critical assessment.” Brief. Bioinform. 8 (2007): 70–81. ↩︎

5. Sela-Culang, Inbal, Kunik, Vered & Ofran, Yanay. “Structural analysis of B-cell epitopes in antibody:protein complexes.” Structure 21 (2013): 1806–1813. ↩︎

6. Lensink, Marc F. & Wodak, Shoshana J. “Docking and scoring protein interactions: CAPRI 2009.” Proteins 78 (2010): 3073–3084. ↩︎

7. Worth, Catherine L., Gong, Sungsam & Blundell, Tom L. “SDM—a server for predicting effects of mutations on protein stability and malfunction.” Nucleic Acids Res. 39, W215–W222 (2011). ↩︎

8. Tien, Matthew Z., Meyer, Austin G., Sydykova, Dariya K., Spielman, Stephanie J. & Wilke, Claus O. “Maximum allowed solvent accessibilities of residues in proteins.” PLoS ONE 8 (2013): e80635. ↩︎ ↩︎