RustSASA

RustSASA is a Rust library for computing the absolute solvent accessible surface area (ASA/SASA) of each atom in a given protein structure using the Shrake-Rupley algorithm[1]. It can be used in Rust and Python!

Features:

• 🦀 Written in Pure Rust
• ⚡️ 3X Faster than Biopython and ~120% faster than Freesasa
• 🧪 Full test coverage
• 🐍 Can be used in Python

Using in Rust 🦀

use pdbtbx::StrictnessLevel;
use rust_sasa::{Atom, calculate_sasa, calculate_sasa_internal, SASALevel};
let (mut pdb, _errors) = pdbtbx::open(
             "./example.cif",
             StrictnessLevel::Medium
).unwrap();
let result = calculate_sasa(&pdb,None,None,SASALevel::Residue);

Full documentation can be found here

Using in Python 🐍

You can now utilize RustSasa within Python to speed up your scripts! Take a look at rust-sasa-python!

Installation:

pip install rust-sasa-python

Example:

from rust_sasa_python import calculate_sasa_at_residue_level
residue_sasa_values = calculate_sasa_at_residue_level("path_to_pdb_file.pdb") # Also supports mmCIF files!

See full docs here

Benchmarking

Benchmarks were performed on an M2 Apple Mac with 8GB of RAM and 8 Cores with the protein AF-A0A2K5XT84-F1 (AlphaFold).

• Biopython: ~150ms

• Freesasa: ~90ms

• RustSASA: ~40ms

Citations:

1: Shrake A, Rupley JA. Environment and exposure to solvent of protein atoms. Lysozyme and insulin. J Mol Biol. 1973 Sep 15;79(2):351-71. doi: 10.1016/0022-2836(73)90011-9. PMID: 4760134.