Structural properties

distances(simulation, indices1::AbstractVector{<:Integer}, indices2::AbstractVector{<:Integer})
distances(simulation, selection1::AbstractVector{<:PDBTools.Atom}, selection2::AbstractVector{<:PDBTools.Atom})

Function that calculates the distance between the centers of mass of two selections in a simulation.

The selections are defined by the indices1 and indices2 vectors, which are the indices of the atoms, or by the selection1 and selection2 vectors, which are vectors of PDBTools.Atom objects.

Use silent=true to suppress the progress bar.

Example

julia> using PDBTools

julia> using MolSimToolkit, MolSimToolkit.Testing

julia> sim = Simulation(Testing.namd_pdb, Testing.namd_traj);

julia> ats = atoms(sim);

julia> i1 = findall(sel"protein and residue 1", ats); # indices

julia> i2 = findall(sel"protein and residue 15", ats); # indices

julia> distances(sim, i1, i2; silent=true)
5-element Vector{Float64}:
 23.433267858947584
 30.13791365033211
 28.48617683945202
 27.92740141686934
 23.235012287435566

julia> distances(sim, 
           filter(sel"protein and residue 1", ats), # selection (PDBTools.Atom)
           filter(sel"protein and residue 15", ats); # selection (PDBTools.Atom) 
           silent=true
       )
5-element Vector{Float64}:
 23.433267858947584
 30.13791365033211
 28.48617683945202
 27.92740141686934
 23.235012287435566

center_of_mass(
    indices::AbstractVector{Int};
    simulation::Simulation,
    positions::FramePositions,
    iref::Union{Nothing,Int} = max(1, div(length(indices),2)),
)

Calculate the center of mass of a selection of atoms in a simulation given the positions. The selection is defined by the indices vector, which is the indices of the atoms.

The iref parameter is the index of the reference atom. The center of mass is calculated by first computing the minimum-image of all atoms relative to this atom. By default, it is the atom closest to the middle of the indices vector. If iref is nothing, the center of mass is calculated without wrapping the coordinates.

julia> using PDBTools 

julia> using MolSimToolkit, MolSimToolkit.Testing

julia> simulation = Simulation(Testing.namd_pdb, Testing.namd_traj);

julia> protein_indices = findall(sel"protein", atoms(simulation));

julia> first_frame!(simulation); # move simulation to the first frame

julia> coor = positions(current_frame(simulation));

julia> cm = center_of_mass(protein_indices, simulation, coor)
3-element Point3D{Float64} with indices SOneTo(3):
 -3.7290442807974906
 -1.5339226637687564
  1.960640754560446

most_representative_structure(simulation::Simulation; atoms = nothing)

Find the most representative structure in a simulation. The most representative structure is the one that minimizes the RMSD with respect to the average structure of the simulation. The average structure is defined iteratively, first by aligning all frames to the first frame, and then by averaging the aligned structures. The structure most similar to the average is then identified and used as the reference structure for the next iteration. The process is repeated until the structure most similar to the average is the same as the previous iteration.

Arguments

• simulation::Simulation: Simulation object.
• atoms: Atoms to consider in the calculation:
  • atoms is nothing: the function will consider all alpha-carbons in proteins ("protein and name CA").
  • atoms is an AbstractVector{<:PDBTools.Atom}: the function will consider the atoms in the vector.
  • atoms is an AbstractVector{<:Int}: the function will consider the atoms with the indices in the vector.
  • atoms is a String: the function will consider the atoms selected by the string.

Returns

• Tuple (Int, Float64), with:
  • Index of the most representative structure.
  • RMSD of the most representative structure with respect to the average structure.

Example

julia> using MolSimToolkit, MolSimToolkit.Testing, PDBTools

julia> simulation = Simulation(Testing.namd_pdb, Testing.namd_traj);

julia> most_representative_structure(simulation) # atoms == nothing (all alpha-carbons in proteins)
(4, 1.1681526249035976)

julia> most_representative_structure(simulation; atoms = "protein and name CA") # atoms is a String
(4, 1.1681526249035976)

julia> calphas = select(atoms(simulation), "name CA");

julia> most_representative_structure(simulation; atoms = calphas) # atoms is an Vector{PDBTools.Atom}
(4, 1.1681526249035976)

julia> ica = PDBTools.index.(calphas)

julia> most_representative_structure(simulation; atoms = ica) # atoms is an vector of indices
(4, 1.1681526249035976)