More modification and support for regular expression

README.md

@@ -13,6 +13,8 @@ julia> using Pkg; Pkg.add("https://github.com/yufongpeng/PeptideSeq.jl")
 
 ## Example
 ```julia
+julia> using PeptideSeq
+
 julia> p = Protein("DPCHKPKRRKP")
 Protein:      DPCHKPKRRKP
 Modification: 

config_example/MODIFICATION.tsv

@@ -1,2 +1,9 @@
-Modification	Accurate Mass	Avearge Mass	Sites		
-3NPH	135.043262	135.12472	D	E	$
+Modification	Accurate Mass	Avearge Mass	Sites			
+3NPH	135.043262	135.12472	D	E	$	
+N-acetylation	42.010565	42.03672	K	^		
+methylation	14.01565	14.02705	K	R		
+O-Hex	162.052824	162.14072	S	T		
+O-HexNAc	203.079374	162.14072	S	T		
+hydroxylation	15.994915	15.999	K	P		
+phosphorylation	79.966333	79.97872	S	T	Y	H
+N-GlcNAc	203.079374	203.19472	"r""K[^P][S, T]"""			

src/PeptideSeq.jl

@@ -80,14 +80,14 @@ end
     add_modification!(source)
 
 Add custom modification. The first row is the header (can be empty), the first column is the name of modification, the second and third columns are addtional monoisotopic mass and average mass, respectively, and the other columns are the modification site. 
-^ repressents the N-terminal and \$ repressents the C-terminal.
+Modification sites can be string without quotation or regular expression like r"...". ^ repressents the N-terminal and \$ repressents the C-terminal. 
 See the example file "config_example/MODIFICATION.tsv".
 """
 function add_modification!(source)
     for m in CSV.Rows(source, delim = "\t")
         push!(MODIFICATION_MS[1], m[1] => parse(Float64, m[2]))
         push!(MODIFICATION_MS[2], m[1] => parse(Float64, m[3]))
-        push!(MODIFICATION_SITE, m[1] => [loc for loc in getindex.(Ref(m), 4:length(m)) if !ismissing(loc)])
+        push!(MODIFICATION_SITE, m[1] => [startwith(loc, "r") ? eval(Meta.parse(loc)) : loc for loc in getindex.(Ref(m), 4:length(m)) if !ismissing(loc)])
     end
 end
 

src/config.jl

@@ -66,11 +66,38 @@ const ADD_MS = (
     )  
 )
 
-const MODIFICATION_SITE = Dict{String, Vector{String}}("3NPH" => ["D", "E", "\$"])
+const MODIFICATION_SITE = Dict{String, Vector{Any}}(
+    "3NPH" => ["D", "E", "\$"],
+    "N-acetylation" => ["K", "^"],
+    "methylation" => ["K", "R"],
+    "O-Hex" => ["S", "T"],
+    "O-HexNAc" => ["S", "T"],
+    "hydroxylation" => ["K", "P"],
+    "phosphorylation" => ["S", "T", "Y", "H"],
+    "N-GlcNAc" => [r"K[^P][S, T]"]
+    )
 
 const MODIFICATION_MS = (
-    Dict{String, Float64}("3NPH" => 135.043262),
-    Dict{String, Float64}("3NPH" => 135.12472)
+    Dict{String, Float64}(
+        "3NPH" => 135.043262,
+        "N-acetylation" => 42.010565,
+        "methylation" => 14.015650,
+        "O-Hex" => 162.052824,
+        "O-HexNAc" => 203.079374,
+        "hydroxylation" => 15.994915,
+        "phosphorylation" => 79.966333,
+        "N-GlcNAc" => 203.079374
+    ),
+    Dict{String, Float64}(
+        "3NPH" => 135.12472,
+        "N-acetylation" => 42.03672,
+        "methylation" => 14.02705,
+        "O-Hex" => 162.14072,
+        "O-HexNAc" => 203.19472,
+        "hydroxylation" => 15.999,
+        "phosphorylation" => 79.97872,
+        "N-GlcNAc" => 203.19472
+    )
 )
 
 const ENZYME = Dict(

src/preparation.jl

@@ -9,19 +9,13 @@ See object `MODIFICATION_SITE` for available modifications.
 Currently, "3NPH" is supported.
 """
 function modify!(protein::Protein, modification::String...)
-    # If digestion had been done, add mass to each peptides
-    if CONFIG["ACCURATE"]
-        modification_ms = first(MODIFICATION_MS)
-    else
-        modification_ms = last(MODIFICATION_MS)
-    end
-    isempty(protein.peptides) || return _modify_mass!(protein, modification...)
-
     for k in modification
-        haskey(modification_ms, k) || continue
+        haskey(MODIFICATION_SITE, k) || continue
         locs = Int[]
-        for loc in modification_ms[k]
-            if loc == "^"
+        for loc in MODIFICATION_SITE[k]
+            if isa(loc, Regex)
+                append!(locs, locc.offset for locc in eachmatch(loc, protein.origin))
+            elseif loc == "^"
                 push!(locs, 1)
             elseif loc == "\$"
                 push!(locs, length(protein.origin))
@@ -31,7 +25,8 @@ function modify!(protein::Protein, modification::String...)
         end
         protein.modification[k] = locs
     end
-    protein
+    # If digestion had been done, add mass to each peptides
+    return isempty(protein.peptides) ? protein : _modify_mass!(protein, modification...)
 end
 
 function modify!(protein::Protein, modification::Dict{String, Vector{Int}})
@@ -52,20 +47,6 @@ function _modify_mass!(protein::Protein, modification::String...)
         modification_ms = last(MODIFICATION_MS)
     end
 
-    for k in modification
-        haskey(MODIFICATION_SITE, k) || continue
-        locs = Int[]
-        for loc in MODIFICATION_SITE[k]
-            append!(locs, findall(==(first(loc)), protein.origin))
-            if loc == "^"
-                push!(locs, 0)
-            elseif loc == "\$"
-                push!(locs, -1)
-            end
-        end
-        protein.modification[k] = locs
-    end
-
     for (k, v) in protein.modification
         for (i, pep) in enumerate(protein.peptides)
             id = filter(in(v), pep.position)
@@ -125,7 +106,7 @@ function digest!(protein::Protein, n_miss::Int, enzyme::String = "")
     end
 
     if enzyme == ""
-        protein.enzyme != "" && throw(ArgumentError("Please provide enzyme!"))
+        protein.enzyme == "" && throw(ArgumentError("Please provide enzyme!"))
     else
         protein.enzyme = enzyme
     end