Step 1: Automation of the Partitioning Schemes
Automatic fragmentation of a large
molecule is desirable, because of the time-consuming and error-prone
aspects associated with manual fragmentation. This implementation
supports protein files in PDB or MOL2 format.
For MFCC, splitting the buried
residues of the protein results in two incomplete points that require
two groups of conjugated caps (‘caps’) to cap the severed valences
(Figure 3). For example, if the split incomplete fragments include three
amino acids, the estimated atoms in the subsystems and two caps are
always >60, even for small residues, constituting a
relatively large system for QM calculations. According to computational
efficiency and convenience, each fragment was generated by breaking the
peptide bond. Assuming that one protein contains N amino acid residues
(Figure 3; N = 3), N fragments and N − 1 caps will be generated after
fragmentation. First, all molecular coordinates are scanned to search
the split points, after which the protein is split into N incomplete
fragments; each includes one or two caps according to residue location
(terminus or buried area). Splitting buried residues results in two
incomplete points, which require two caps (e.g., Frag 2 in Figure 3).
For residues located at termini, there is only one split point,
requiring only one cap (e.g., Frags 1 and 3 in Figure 3). The caps are
not only used as buffered groups for residues, but also mimic the
protein’s real environment.