The
capabilities of
current
computer simulations provide a unique opportunity to model small-angle s
cattering (SAS) data at the atomisti
c level, and to in
clude other stru
ctural
constraints ranging from mole
cular and atomisti
c energeti
cs to
crystallography, ele
ctron mi
cros
copy and NMR. This extends the
capabilities of solution s
cattering and provides deeper insights into the physi
cs and
chemistry of the systems studied. Realizing this potential, however, requires integrating the experimental data with a new generation of modelling software. To a
chieve this, the CCP-SAS
collaboration (
ccpsas.org/" rel="references:http://www.ccpsas.org/">http://www.ccpsas.org/) is developing open-sour
ce, high-throughput and user-friendly software for the atomisti
c and
coarse-grained mole
cular modelling of s
cattering data. Robust state-of-the-art mole
cular simulation engines and mole
cular dynami
cs and Monte Carlo for
ce fields provide
constraints to the solution stru
cture inferred from the small-angle s
cattering data, whi
ch in
corporates the known physi
cal
chemistry of the system. The implementation of this software suite involves a tiered approa
ch in whi
ch
GenApp provides the deployment infrastru
cture for running appli
cations on both standard and high-performan
ce
computing hardware, and
SASSIE provides a workflow framework into whi
ch modules
can be plugged to prepare stru
ctures,
carry out simulations,
cal
culate theoreti
cal s
cattering data and
compare results with experimental data.
GenApp produ
ces the a
ccessible web-based front end termed
SASSIE-web, and
GenApp and
SASSIE also make
community SAS
codes available. Appli
cations are illustrated by
case studies: (i) inter-domain flexibility in two- to six-domain proteins as exemplified by HIV-1 Gag, MASP and ubiquitin; (ii) the hinge
conformation in human IgG2 and IgA1 antibodies; (iii) the
complex formed between a hexameri
c protein Hfq and mRNA; and (iv) syntheti
c `bottlebrush' polymers.