FreeSASA  1.1
Open source SASA calculations
View on GitHub

![DOI]( ![Build Status]( ![Coverage Status](

C-library for calculating Solvent Accessible Surface Areas.

License: MIT (see file LICENSE). Copyright: Simon Mitternacht 2013-2016.

FreeSASA is a C library and command line tool for calculating Solvent Accessible Surface Area (SASA) of biomolecules. It is designed to be very simple to use with defaults, but allows customization of all parameters of the calculation and provides a few different tools to analyze the results as well. Python bindings are also included in the repository.

The library includes both the algorithm by Lee & Richards and that by Shrake & Rupley. Verification has been done by comparing the results of the two calculations and by visual inspection of the surfaces found by them (and comparing with analytic results in the two-atom case). Both can be parameterized to arbitrary precision, and for high resolution versions of the algorithms, the calculations give identical results.

FreeSASA assigns a radius and a class to each atom. The atomic radii are by default those defined by Tsai et al. (JMB 1999, 290: 253) for standard amino acids and nucleic acids, and the van der Waals radius of the element for other atoms. Each atom is also assigned to a class. The default classes are Apolar (carbon) and Polar (all other elements). The program outputs the total SASA and the area of these two classes, if a protein has more than one chain the contribution of each chain is also included in the output.

Users can provide their own atomic radii and classes, either via configuration files or via the API. The input format for configuration files is described in the online documentation, and the share/ directory contains some sample configurations, including one for the NACCESS parameters (Hubbard & Thornton 1993).

Compiling and installing

FreeSASA can be compiled and installed using the following

make && make install

NB: If the source is downloaded from the git repository the configure-script needs to be set up first using autoreconf -i. Users who don't have autotools installed, can download a tarball that includes the autogenerated scripts from or from the latest GitHub-release.

The above commands build and install the command line tool freesasa (built in src/), the command

freesasa -h

gives an overview of options. To run a calculation from PDB-file input using the defaults, simply type

freesasa <pdb-file>

make install also installs the header freesasa.h and the library libfreesasa. If the configure script is called with the option --enable-python-bindings, the Python module is also built and installed.

Configuring with

./configure CFLAGS="-Ofast" 

increases the speed of the Shrake & Rupley algorithm significantly (10-20%), as compared to the standard "-O2". There seems to be no measurable effect on Lee & Richards.

The configuration script can be customized with general options:

And some options for developers:


Enabling Doxygen builds a full reference manual, documenting both CLI and API, also available on the web at

After building the package, calling

freesasa -h

explains how the commandline tool can be used.

Compatibility and dependencies

The program has been tested successfully with several versions of GNU C Compiler and Clang/LLVM. Building the library only requires standard C and GNU libraries. Developers who want to do testing need to install the Check unit testing framework. Building the full reference manual requires Doxygen (version > 1.8.8). Building the Python bindings requires Cython. Changing the selection parser and lexer requires Flex and Bison. These build options, which add extra dependencies, are disabled by default to simplify installation for users only interested in the command line tool and and/or C Library.

Citing FreeSASA

FreeSASA can be cited using the following publication

The DOI numbers from Zenodo can be used to cite a specific version of FreeSASA.