ABS

 Computational Structural Biology (CSB) is the scientific domain
concerned with the development of algorithms and software to
understand and predict the function of biological macromolecules.  A
central issue in this realm, addressed using both experimental and
computational techniques, is the study of protein dynamics and
molecular motions.  Dynamics are indeed key to predict the structure
from the sequence, to understand the function of a protein whose
static structure is known, and also to unveil the structure of
molecular machines with tens/hundreds of subunits.  Alas, with motions
spanning $\sim$ 15 and $\sim$ 4 orders of magnitude in time and
amplitude respectively, dynamics are hard to study.

Using novel ideas and concepts from computational geometry,
optimization, machine learning, probabilistic algorithms, and
numerical probability applied to energy landscapes, the central goal
of ABS is to introduce a new generation of algorithms to make
accurate predictions on the dynamics of proteins.

High quality implementations of all algorithms will be integrated to
our library, the Structural Bioinformatics Library, which is the only
software platform providing the level of modularity required to
perform the integration of advanced complementary algorithms.

In collaboration with colleagues (biologist, biophysicists),
these  implementations will be used to get insights on systems of
crucial importance in biology and medicine.