Merge branch 'master' into bcif

Author: BradyAJohnston

NEWS.md

@@ -1,5 +1,11 @@
 # BioStructures.jl release notes
 
+## v4.5.0 - May 2025
+
+* `chiangle` and `chiangles` are added to calculate the sidechain dihedral angles in protein residues.
+* mmCIF files with secondary structure annotations will now, by default, parse and assign secondary structure. Setting `run_dssp` or `run_stride` to `true` will cause DSSP or STRIDE to be run and provide the secondary structure, as before.
+* A constructor for `MetaGraph` on a `Chain` is added that constructs a graph of atoms where edges are determined by the known bonds of standard amino acids in the chain.
+
 ## v4.4.2 - Mar 2025
 
 * A bug in storing indices for `Transformation` is fixed.

Project.toml

@@ -1,7 +1,7 @@
 name = "BioStructures"
 uuid = "de9282ab-8554-53be-b2d6-f6c222edabfc"
 authors = ["Joe G Greener <jgreener@hotmail.co.uk>"]
-version = "4.4.2"
+version = "4.5.0"
 
 [deps]
 BioGenerics = "47718e42-2ac5-11e9-14af-e5595289c2ea"

docs/src/documentation.md

@@ -527,7 +527,7 @@ The coordinate sets must be the same size and have the number of dimensions in t
 
 The contacting elements in a molecular structure form a graph, and this can be retrieved using `MetaGraph`.
 This extends `MetaGraph` from [MetaGraphs.jl](https://github.com/JuliaGraphs/MetaGraphs.jl), allowing you to use all the graph analysis tools in [Graphs.jl](https://github.com/JuliaGraphs/Graphs.jl).
-For example:
+For example to find the graph from a contact map of Cβ atoms closer than 8 Å:
 
 ```julia-repl
 julia> using Graphs, MetaGraphs
@@ -552,6 +552,9 @@ See the [Graphs docs](https://juliagraphs.org/Graphs.jl/dev) for details on how
 Similar to [`ContactMap`](@ref), contacts are found between any element type passed in.
 So if you wanted the graph of chain contacts in a protein complex you could give a [`Model`](@ref) as the first argument.
 
+Calling `MetaGraph` on a [`Chain`](@ref) without a contact distance does something different: it constructs a graph of atoms where edges are determined by the known bonds of standard amino acids in the chain.
+Hydrogens should be present and atom names should match those in OpenMM.
+
 ## Assigning secondary structure
 
 Any secondary structure assignment at a residue is accessed via [`sscode`](@ref):

ext/BioStructuresGraphsExt.jl

@@ -8,7 +8,8 @@ using BioStructures: findatombyname
 """
     MetaGraph(element, contact_distance)
 
-Construct a graph of atoms where edges are contacts separated by less than `contact_distance`.
+Construct a graph of elements where edges are contacts separated by less
+than `contact_distance`.
 
 See Graphs.jl and MetaGraphs.jl for more on how to use graphs.
 """
@@ -30,22 +31,24 @@ function MetaGraphs.MetaGraph(el::StructuralElementOrList, contact_dist::Real)
 end
 
 """
-    MetaGraph(chain::Chain; strict::Bool = true)
+    MetaGraph(chain::Chain; strict::Bool=true)
 
-Construct a graph of atoms where edges are determined by the known bonds of residues in the chain.
+Construct a graph of atoms where edges are determined by the known bonds
+of residues in the chain.
 
 By default, the graph is constructed in `strict` mode, which means that:
 
 - residue and atom names must be standard
 - all hydrogens are present
 - HIS must be disambiguated as HIE, HID, or HIP
 
-These constraints can be relaxed by setting `strict = false`, at some risk to accuracy.
+These constraints can be relaxed by setting `strict = false`, at some
+risk to accuracy.
 
 See Graphs.jl and MetaGraphs.jl for more on how to use graphs.
 """
-function MetaGraphs.MetaGraph(chain::Chain; strict::Bool = true)
-    el_list = collectatoms(chain; expand_disordered = true)
+function MetaGraphs.MetaGraph(chain::Chain; strict::Bool=true)
+    el_list = collectatoms(chain; expand_disordered=true)
     mg = MetaGraph(length(el_list))
     for (i, el) in enumerate(el_list)
         set_prop!(mg, i, :element, el)

ext/BioStructuresMMTFExt.jl

@@ -197,7 +197,7 @@ function BioStructures.writemmtf(output::Union{AbstractString, IO},
                 atom_selectors::Function...;
                 expand_disordered::Bool=true,
                 gzip::Bool=false)
-    loop_el = isa(el, MolecularStructure) ? collectmodels(el) : el
+    loop_el = (isa(el, MolecularStructure) ? collectmodels(el) : el)
     ats = sort(sort(sort(collectatoms(loop_el, atom_selectors...;
                             expand_disordered=expand_disordered)),
                             by=residue), by=model)
@@ -251,7 +251,7 @@ function BioStructures.writemmtf(output::Union{AbstractString, IO},
                     "chainIndexList" => Any[length(d["chainIdList"]) - 1],
                     "description"    => "",
                     "sequence"       => "", # This is changed later
-                    "type"           => ishetero(res) ? "non-polymer" : "polymer",
+                    "type"           => (ishetero(res) ? "non-polymer" : "polymer"),
                 ))
             end
             if !ishetero(res)
@@ -286,7 +286,7 @@ function BioStructures.writemmtf(output::Union{AbstractString, IO},
                     "bondAtomList"     => Any[],
                     "elementList"      => Any[element(at) for at in ats_res],
                     # MMTF specifies missing charges as zero
-                    "formalChargeList" => Any[charge(at) == "" ? 0 : parse(Int64, charge(at)) for at in ats_res],
+                    "formalChargeList" => Any[(charge(at) == "" ? 0 : parse(Int64, charge(at))) for at in ats_res],
                     "singleLetterCode" => "",
                     "chemCompType"     => "",
                     "atomNameList"     => Any[at_names...],
@@ -295,17 +295,17 @@ function BioStructures.writemmtf(output::Union{AbstractString, IO},
             end
 
             push!(d["groupIdList"], resnumber(res))
-            push!(d["groupTypeList"], group_i == 0 ? length(d["groupList"]) - 1 : group_i - 1)
-            push!(d["insCodeList"], inscode(res) == ' ' ? '\0' : inscode(res))
+            push!(d["groupTypeList"], (group_i == 0 ? length(d["groupList"]) - 1 : group_i - 1))
+            push!(d["insCodeList"], (inscode(res) == ' ' ? '\0' : inscode(res)))
             push!(d["secStructList"], -1)
-            push!(d["sequenceIndexList"], ishetero(res) ? -1 : length(sequence) - 1)
+            push!(d["sequenceIndexList"], (ishetero(res) ? -1 : length(sequence) - 1))
 
             prev_resname = resname(res)
             prev_het = ishetero(res)
             last_res = res
         end
 
-        push!(d["altLocList"], altlocid(at) == ' ' ? '\0' : altlocid(at))
+        push!(d["altLocList"], (altlocid(at) == ' ' ? '\0' : altlocid(at)))
         push!(d["atomIdList"], serial(at))
         push!(d["bFactorList"], tempfactor(at))
         push!(d["occupancyList"], occupancy(at))

extractdata/bonding.jl

@@ -38,7 +38,6 @@ function parsexmlline(f, line, tag, keyname)
     return key => (; vals...)
 end
 
-
 atomtypes, residues = open("protein.ff14SB.xml", "r") do io
     line = readline(io)
     @assert line == "<ForceField>"

src/mmcif.jl

@@ -573,7 +573,7 @@ function writemmcif(output::IO,
                 atom_selectors::Function...;
                 expand_disordered::Bool=true, gzip::Bool=false)
     # Ensure multiple models get written out correctly
-    loop_el = isa(el, MolecularStructure) ? collectmodels(el) : el
+    loop_el = (isa(el, MolecularStructure) ? collectmodels(el) : el)
     ats = collectatoms(loop_el, atom_selectors...;
                                         expand_disordered=expand_disordered)
     if length(ats) > 0
@@ -590,7 +590,7 @@ function writemmcif(output::IO,
         appendatom!(atom_dict, at, string(modelnumber(at)),
             strip(chainid(at)) == "" ? "." : chainid(at),
             string(resnumber(at)), resname(at),
-            ishetero(at) ? "HETATM" : "ATOM")
+            (ishetero(at) ? "HETATM" : "ATOM"))
     end
 
     # Now the MMCIFDict has been generated, write it out to the file

src/model.jl

@@ -153,7 +153,7 @@ function Residue(r::Residue, ch::StructuralElement)
     rnew = Residue(r.name, r.number, r.ins_code, r.het_res, [name for name in r.atom_list],
         atom_dict, ch, r.ss_code)
     for (name, atom) in r.atoms
-        atom_dict[name] = isa(atom, Atom) ? Atom(atom, rnew) : DisorderedAtom(atom, rnew)
+        atom_dict[name] = (isa(atom, Atom) ? Atom(atom, rnew) : DisorderedAtom(atom, rnew))
     end
     return rnew
 end
@@ -183,7 +183,7 @@ function Chain(c::Chain, mo::StructuralElement)
     res_dict = Dict{String,AbstractResidue}()
     cnew = Chain(c.id, [id for id in c.res_list], res_dict, mo)
     for (id, res) in c.residues
-        res_dict[id] = isa(res, Residue) ? Residue(res, cnew) : DisorderedResidue(res, cnew)
+        res_dict[id] = (isa(res, Residue) ? Residue(res, cnew) : DisorderedResidue(res, cnew))
     end
     return cnew
 end
@@ -1412,7 +1412,7 @@ function collectresidues(el::Union{Chain,Vector{<:AbstractResidue}};
         end
         return res_list
     else
-        return isa(el, Chain) ? collect(el) : el
+        return (isa(el, Chain) ? collect(el) : el)
     end
 end
 
@@ -1506,7 +1506,7 @@ function collectatoms(el::Union{Residue,Vector{<:AbstractAtom}};
         end
         return at_list
     else
-        return isa(el, Residue) ? collect(el) : el
+        return (isa(el, Residue) ? collect(el) : el)
     end
 end
 

src/pdb.jl

@@ -201,12 +201,12 @@ end
 
 function parseoccupancy(line::String)
     ret = tryparse(Float64, line[55:60])
-    return ret === nothing ? 1.0 : ret
+    return (ret === nothing ? 1.0 : ret)
 end
 
 function parsetempfac(line::String)
     ret = tryparse(Float64, line[61:66])
-    return ret === nothing ? 0.0 : ret
+    return (ret === nothing ? 0.0 : ret)
 end
 
 function parseelement(line::String)

test/runtests.jl

@@ -2426,7 +2426,7 @@ end
     ]
     # Read from file - write a small multicif manually and read it back
     for gzip in (false, true)
-        transcoder = gzip ? (GzipCompressorStream,) : ()
+        transcoder = (gzip ? (GzipCompressorStream,) : ())
         open(transcoder..., temp_filename, "w") do f_out
             open(testfilepath("mmCIF", "1AKE.cif")) do f_in
                 write(f_out, f_in)
@@ -3192,7 +3192,7 @@ end
     @test isapprox(omegas[10], omegaangle(struc_1AKE['A'], 10), atol=1e-5)
 
     # Test that the entries in `chitables` are bonded
-    sortt((a, b)) = a < b ? (a, b) : (b, a)
+    sortt((a, b)) = (a < b ? (a, b) : (b, a))
     rd = BioStructures.residuedata
     for ct in BioStructures.chitables
         for (rname, alist) in ct