v2.10.0

ComplexMixtures v2.10.0

Diff since v2.9.0

New feature

The indexing and iteration of a ResidueContributions object was improved, such that:

1. When a ResidueContributions contain the contributions of more than one residue, the iteration occurs within residues.
2. When a ResidueContributions contain the contribution of a single residue, iteration occurs over the contributions of this residue.

This implies many facilities and greater consistency in iterations over these objects as, for instance, it is possible to extract (for example) the contribution at a given distance for all residues as:

julia> rc = ResidueContributions(R, select(ats, "protein"))

          Residue Contributions - 274 residues.
     3.51 █     █      █     █            █                                    
     3.27 █              █   █                                                 
     3.03 █     █    █       █            █       █       █                █   
     2.79 █    ██    █ █ █   █            █      ██          █        █    █   
 d   2.55 █ █  ██    █ █ █   █            ██     ██ █  █  █  █  █     █    █   
     2.31 █ █  ██    █ ███   █    ██      ██     ██ █  ██ █  ██ █     █    █   
     2.07 █ █   █  █ █████   █    ██      ██     ██ █  ██ █  █ ██    █     █   
     1.83 █   █ █  █ █████   █    ██      █      ██ █  ██    █ ██     █    █   
     1.59                                                                      
         A1      T33     T66     S98     S130    T162    A194    H226    G258    

julia> id = findfirst(>=(2.0), rc.d) # index of distance for 2.0 Angstroms
26

julia> c_at_2 = [ r[id] for r in rc ]  # collect contributions for all residues at index id
274-element Vector{Float64}:
 0.025795222201597752
 0.027637738073140444
 ⋮
 0.0
 0.03316528568776853

The above is possible because, for example, rc[1] returns a "single-residue" ResidueContributions object, and indexing this object returns the contributions of such residue at each distance.

Similarly, these features allow other Julia functions to be broadcasted over the residue contributions:

getindex.(rc, 26) # returns the contribution at index 26 for all residues (equivalent to the `[ r[id] for r in rc ]` above)
last.(rc) # returns the contribution of each residue at the largest distance
maximum.(rc) # returns the maximum contribution of each residue
findmax.(rc) # returns the maximum and index of the distance maximum of the contributions of each residue

Merged pull requests:

• New single residue contribution iterator and indexing (#50) (@lmiq)

v2.9.0

• Update docs to make Trajectory a minor menu.

v2.9.0

ComplexMixtures v2.9.0

Diff since v2.8.5

New features

• The calls to mddf and coordination_number now accept the name of the trajetory file and the solute and solvent selections, passing by the construction of the Trajectory object. This is now the preferred and documented interface:

  mddf(trajectory_file, solute, solvent, options)
  mddf(trajectory_file, solute_and_solvent, options) # for auto-correlations
  coordination_number(trajectory_file, solute, solvent, option)
  coordination_number(trajectory_file, solute_and_solvent, options) # for auto-correlations

  where the solute, solvent and solute_and_solvent objects are AtomSelection structures.

Other changes:

• documentation updates.

Merged pull requests:

• New call without trajectory (#49) (@lmiq)

v2.8.5

ComplexMixtures v2.8.5

Diff since v2.8.4

• Throw ArgumentError messages if n_random_samples <= 0 or if sum(frame_weights[considered_range]) <= 0.

Merged pull requests:

• Error on zeros (#47) (@lmiq)

v2.8.4

ComplexMixtures v2.8.4

Diff since v2.8.3

• Improve error message for ResidueContributions with wrong arguments.

v2.8.3

ComplexMixtures v2.8.3

Diff since v2.8.2

• Add compatibility to PDBTools.jl v2

Merged pull requests:

• Compatibility with PDBTools 2.0 (#45) (@lmiq)

v2.8.2

ComplexMixtures v2.8.2

Diff since v2.8.1

• Fix bug in the computation of contributions of subgroups of solute groups that are discontinous molecules in the structure file.
• Computing contributions is much faster.
• Remove the experimental _unsafe_types_from_indices option from contributions, which is not necessary anymore.

Merged pull requests:

• run GC after reading frames and show progress (#42) (@lmiq)
• fix contributions of non-sequential solutes, and improve performance of contributions function (#43) (@lmiq)

v2.8.1

ComplexMixtures v2.8.1

Diff since v2.8.0

• Add (experimental) option _unsafe_types_from_indices::Bool to contributions and ResidueContributions, to compute (much) faster the residue contributions. By default, it is set to false, and the previous safe search of types is used.

v2.8.0

ComplexMixtures v2.8.0

Diff since v2.7.8

New feature:

• save and load ResidueContributions objects.
• some improved error messages.

v2.7.8

ComplexMixtures v2.7.8

Diff since v2.7.7

• Silence warning given in precompilation.