It is immediately possible to compute topological descriptors at
all the levels of definition from single residue (node) to the entire
protein (whole network) passing by cluster of nodes (structural
domains).
Thus, we can compute the degree of each node (how many
edges correspond to a residue) that is a local, microscopic feature of
the system or the “average shortest path” corresponding to the
average length of minimal paths connecting all the node pairs (this
is a wiring architecture global descriptor) or to group the residues
into maximally connected clusters (a mesoscale feature) [5, 6]. The
unique feature of graph formalization is that these different levels
views are strictly intermingled and cannot by no way be considered
as independent and/or simply corresponding to gross averages: a
residue endowed with a low average shortest path with other
residues (single node property) obtains its value thanks to the entire
network wiring architecture (top-down causation), while in the
same time influences the flux of information (energy) across the
entire molecule (bottom-up causation). This natural interaction of
scales allows to disentangle the different aspects of allostery from
the recognition of the aminoacid residues more involved in signal
transmission (microscopic layer) to the ‘paths’ linking sensor and
effector sites (mesoscopic layer) and the modifications of the entire
graph wiring architecture (macroscopic layer).
In the following, after a general introduction to complex network analysis and a thorough definition of the main topological
descriptors, we will present some practical examples of network
approaches to allostery.
2 Materials
2.1 Structural Data
The computational approach of protein contact networks relies on
the availability of structural information on proteins.
The reference database for protein structures is the Protein
Data Bank [7] (PDB, http://www.rcsb.org/), which also defines
the PDB format, a standard for recording atom files. All files
recorded on the Protein Data Bank repository follow the PDB
format. Information in PDB files is organized in lines, named
records. The PDB files include many types of records, recognizable
by the line header and arranged in a given fashion, to convey all
structural data through a standard format.
Atom coordinates are reported in the ATOM record, organized
as reported in Table 1.
The information of interest for the protein contact networks
constructions are the coordinates recorded in columns 31–54.
Notice that alpha carbons are distinguished by other carbons in
the residues and denoted by the character string “CA”.
Disclosing Allostery Through Protein Contact Networks
9
all the levels of definition from single residue (node) to the entire
protein (whole network) passing by cluster of nodes (structural
domains).
Thus, we can compute the degree of each node (how many
edges correspond to a residue) that is a local, microscopic feature of
the system or the “average shortest path” corresponding to the
average length of minimal paths connecting all the node pairs (this
is a wiring architecture global descriptor) or to group the residues
into maximally connected clusters (a mesoscale feature) [5, 6]. The
unique feature of graph formalization is that these different levels
views are strictly intermingled and cannot by no way be considered
as independent and/or simply corresponding to gross averages: a
residue endowed with a low average shortest path with other
residues (single node property) obtains its value thanks to the entire
network wiring architecture (top-down causation), while in the
same time influences the flux of information (energy) across the
entire molecule (bottom-up causation). This natural interaction of
scales allows to disentangle the different aspects of allostery from
the recognition of the aminoacid residues more involved in signal
transmission (microscopic layer) to the ‘paths’ linking sensor and
effector sites (mesoscopic layer) and the modifications of the entire
graph wiring architecture (macroscopic layer).
In the following, after a general introduction to complex network analysis and a thorough definition of the main topological
descriptors, we will present some practical examples of network
approaches to allostery.
2 Materials
2.1 Structural Data
The computational approach of protein contact networks relies on
the availability of structural information on proteins.
The reference database for protein structures is the Protein
Data Bank [7] (PDB, http://www.rcsb.org/), which also defines
the PDB format, a standard for recording atom files. All files
recorded on the Protein Data Bank repository follow the PDB
format. Information in PDB files is organized in lines, named
records. The PDB files include many types of records, recognizable
by the line header and arranged in a given fashion, to convey all
structural data through a standard format.
Atom coordinates are reported in the ATOM record, organized
as reported in Table 1.
The information of interest for the protein contact networks
constructions are the coordinates recorded in columns 31–54.
Notice that alpha carbons are distinguished by other carbons in
the residues and denoted by the character string “CA”.
Disclosing Allostery Through Protein Contact Networks
9