Merge pull request #16 from Mirmu/fix07

Fixes for julia 0.7, drop support of 0.5

.travis.yml

@@ -4,8 +4,8 @@ os:
   - linux
   - osx
 julia:
-  - 0.5
   - 0.6
+  - 0.7
   - nightly
 notifications:
   email: false

README.md

@@ -25,7 +25,7 @@ The code is written in [Julia][julia], and the functions are called
 from within Julia. However, a command-line interface is also provided for
 those unfamiliar with the language (see the documentation).
 
-The package is tested against Julia `0.5` and `0.6` on Linux, OS X, and Windows.
+The package is tested against Julia `0.6` and `0.7` on Linux, OS X, and Windows.
 
 ## Installation
 

REQUIRE

@@ -1,8 +1,9 @@
-julia 0.5
+julia 0.6
 FastaIO
-Iterators
+IterTools
 Distributions
 MathProgBase 0.6.0 0.8-
 ExtractMacro
 ArgParse
 GLPKMathProgInterface 0.3.0
+Compat 0.66

appveyor.yml

@@ -1,11 +1,11 @@
 environment:
   matrix:
-  - JULIAVERSION: "julialang/bin/winnt/x86/0.5/julia-0.5-latest-win32.exe"
-  - JULIAVERSION: "julialang/bin/winnt/x64/0.5/julia-0.5-latest-win64.exe"
-  - JULIAVERSION: "julialang/bin/winnt/x86/0.6/julia-0.6-latest-win32.exe"
-  - JULIAVERSION: "julialang/bin/winnt/x64/0.6/julia-0.6-latest-win64.exe"
-  - JULIAVERSION: "julianightlies/bin/winnt/x86/julia-latest-win32.exe"
-  - JULIAVERSION: "julianightlies/bin/winnt/x64/julia-latest-win64.exe"
+  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x86/0.6/julia-0.6-latest-win32.exe"
+  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x64/0.6/julia-0.6-latest-win64.exe"
+  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x86/0.7/julia-0.7-latest-win32.exe"
+  - JULIA_URL: "https://julialang-s3.julialang.org/bin/winnt/x64/0.7/julia-0.7-latest-win64.exe"
+  - JULIA_URL: "https://julialangnightlies-s3.julialang.org/bin/winnt/x86/julia-latest-win32.exe"
+  - JULIA_URL: "https://julialangnightlies-s3.julialang.org/bin/winnt/x64/julia-latest-win64.exe"
 
 branches:
   only:
@@ -19,9 +19,15 @@ notifications:
     on_build_status_changed: false
 
 install:
+  - ps: "[System.Net.ServicePointManager]::SecurityProtocol = [System.Net.SecurityProtocolType]::Tls12"
+# if there's a newer build queued for the same PR, cancel this one
+  - ps: if ($env:APPVEYOR_PULL_REQUEST_NUMBER -and $env:APPVEYOR_BUILD_NUMBER -ne ((Invoke-RestMethod `
+        https://ci.appveyor.com/api/projects/$env:APPVEYOR_ACCOUNT_NAME/$env:APPVEYOR_PROJECT_SLUG/history?recordsNumber=50).builds | `
+        Where-Object pullRequestId -eq $env:APPVEYOR_PULL_REQUEST_NUMBER)[0].buildNumber) { `
+        throw "There are newer queued builds for this pull request, failing early." }
 # Download most recent Julia Windows binary
   - ps: (new-object net.webclient).DownloadFile(
-        $("http://s3.amazonaws.com/"+$env:JULIAVERSION),
+        $env:JULIA_URL,
         "C:\projects\julia-binary.exe")
 # Run installer silently, output to C:\projects\julia
   - C:\projects\julia-binary.exe /S /D=C:\projects\julia

docs/src/index.md

@@ -22,7 +22,7 @@ The code is written in [Julia](http://julialang.org), and the functions are call
 from within Julia. However, a [command-line interface](@ref CLI) is also provided for
 those unfamiliar with the language.
 
-The current code was tested on Julia versions 0.4 and 0.5.
+The current code was tested on Julia versions 0.6 and 0.7.
 
 ### Installation
 

src/ParalogMatching.jl

@@ -2,7 +2,9 @@
 
 module ParalogMatching
 
-using FastaIO, ExtractMacro, Iterators
+using Compat
+using Compat.LinearAlgebra, Compat.Random, Compat.SparseArrays
+using FastaIO, ExtractMacro
 using Distributions
 using MathProgBase
 using GLPKMathProgInterface

src/matching_fasta.jl

@@ -28,7 +28,7 @@ function start_matching(X12::HarmonizedAlignments)
     @extract X12 : X1 X2
     @extract X1  : spec_id1=spec_id
     @extract X2  : spec_id2=spec_id
-    match = zeros(spec_id1)
+    match = fill!(similar(spec_id1), 0)
 
     # Finds the indices of the species with one single sequence
     ind1 = index_of_unique(spec_id1)
@@ -37,8 +37,8 @@ function start_matching(X12::HarmonizedAlignments)
     candi = intersect(spec_id1[ind1], spec_id2[ind2])
 
     for el in candi
-        a1 = findfirst(spec_id1, el)
-        a2 = findfirst(spec_id2, el)
+        a1 = findfirst(isequal(el), spec_id1)::Int
+        a2 = findfirst(isequal(el), spec_id2)::Int
         match[a1] = a2
     end
     return match
@@ -58,13 +58,13 @@ function initialize_matching(X12::HarmonizedAlignments, pseudo_count::Float64)
     corr = FastC(freq)
 
     # First compute corr from single matched families
-    single = X1.spec_id[find(match)]
+    single = X1.spec_id[findall(match .≠ 0)]
 
     # Computes the freq matrix for the given matched species "single"
     unitFC!(X1, X2, match, single, freq)
 
     # Finally compute the inverse of the corr matrix
-    if isempty(find(match))
+    if isempty(findall(match .≠ 0))
 	println("WARNING ! 0 sequence matched by uniqueness. No covariation strategy possible.")
 	invC = zeros(size(corr.Cij))
     else
@@ -115,8 +115,8 @@ function apply_matching!(X1, X2, match, lspec, lmatch)
     length(lspec) == length(lmatch) || error("data non compatible")
 
     for (i,el) in enumerate(lspec)
-        ind1 = find(spec_id1 .== el)
-        ind2 = find(spec_id2 .== el)
+        ind1 = findall(spec_id1 .== el)
+        ind2 = findall(spec_id2 .== el)
         match[ind1[lmatch[i][1]]] = ind2[lmatch[i][2]]
     end
     return nothing
@@ -184,13 +184,13 @@ function run_matching(X12::HarmonizedAlignments;
         # Performs the matching for each species of the batch
         res = par_corr(X1, X2, freq, invC, el, strategy, lpsolver)
         println("batch of species")
-        
+
         for (i,spids) in enumerate(el)
             println(X1.spec_name[spids]," ")
             println(res[i])
         end
         println()
-        
+
         # Applies the matching to the global matching vector
         apply_matching!(X1, X2, match, el, res)
 

src/plotting.jl

@@ -169,7 +169,7 @@ function trim_cov_match(nameX1, nameX2, namematch, prop, nameoutput)
     propo = round(Int, prop * len)
     edges1 = Int[a[1][1] for a in retsor[1:propo]]
     edges2 = Int[a[1][2] for a in retsor[1:propo]]
-    nuovomatch = zeros(X1.M)
+    nuovomatch = fill(0.0, X1.M)
     nuovomatch[edges1] = edges2
 
     write_fasta_match(X1, X2, nuovomatch, nameoutput)

src/readdata.jl

@@ -37,7 +37,7 @@ then you can use a regex like this one: `header_regex=r"^(?<species>[^/]+)/([^/]
 i.e. line start, anything except a slash (capture as the species name), followed by a slash, then anything
 except a slash (captured but ignored), followed by a slash, then the rest of the line (captured as the ID).
 """
-function read_fasta_alignment(filename::AbstractString, max_gap_fraction::Float64 = 1.0; header_regex::Union{Void,Regex} = nothing)
+function read_fasta_alignment(filename::AbstractString, max_gap_fraction::Float64 = 1.0; header_regex::Union{Nothing,Regex} = nothing)
     f = FastaReader(filename)
 
     # pass 1
@@ -80,9 +80,9 @@ function read_fasta_alignment(filename::AbstractString, max_gap_fraction::Float6
 
     # pass 2
 
-    Z = Array{Int8}(fseqlen, length(seqs))
-    header =  Array{String}(length(seqs));
-    sequence =  Array{String}(length(seqs));
+    Z = Array{Int8}(undef, fseqlen, length(seqs))
+    header =  Array{String}(undef, length(seqs));
+    sequence =  Array{String}(undef, length(seqs));
     seqid = 1
     for (name, seq) in f
         header[seqid] = name
@@ -104,10 +104,10 @@ function read_fasta_alignment(filename::AbstractString, max_gap_fraction::Float6
     return Alignment(size(Z, 1), size(Z, 2), Int(maximum(Z)), Z', sequence, header, spec_name, spec_id, uniprot_id)
 end
 
-function specname(s::String, header_regex::Union{Void,Regex}, captureinds::NTuple{2,Integer})
+function specname(s::String, header_regex::Union{Nothing,Regex}, captureinds::NTuple{2,Integer})
     if header_regex ≢ nothing
         # user-defined format
-        if ismatch(header_regex, s)
+        if occursin(header_regex, s)
             captures = match(header_regex, s).captures
             length(captures) ≥ 2 ||
                 error("invalid header regex: should always return at least 2 captured groups if it matches; has returned: $(length(captures))")
@@ -124,14 +124,14 @@ function specname(s::String, header_regex::Union{Void,Regex}, captureinds::NTupl
         regex_joined = r"^(.*?)::(.*?)/(.*)$"
 
         # standard format
-        if ismatch(regex_uniprot, s)
+        if occursin(regex_uniprot, s)
             uniprot_id, spec_name = match(regex_uniprot, s).captures
 
         # custom internal formats
-        elseif ismatch(regex_oldskrr, s)
+        elseif occursin(regex_oldskrr, s)
             spec_name = match(regex_oldskrr, s).captures[1]
             uniprot_id = "000000"
-        elseif ismatch(regex_joined, s)
+        elseif occursin(regex_joined, s)
             spec_name = match(regex_joined, s).captures[3]
             uniprot_id = "000000"
         else
@@ -141,11 +141,11 @@ function specname(s::String, header_regex::Union{Void,Regex}, captureinds::NTupl
     return convert(String, uniprot_id), convert(String, spec_name)
 end
 
-function compute_spec(header::Vector{String}, header_regex::Union{Void,Regex} = nothing)
+function compute_spec(header::Vector{String}, header_regex::Union{Nothing,Regex} = nothing)
     M = length(header)
 
-    spec_name = Array{String}(M)
-    uniprot_id  = Array{String}(M)
+    spec_name = Array{String}(undef, M)
+    uniprot_id  = Array{String}(undef, M)
 
     captureinds = (1,2)
     if header_regex ≢ nothing

src/types.jl

@@ -2,7 +2,7 @@
 
 ########################### BLOCK FOR TYPES DEFINITION AND ALLOCATION #####################
 
-immutable Alignment
+struct Alignment
     N::Int
     M::Int
     q::Int
@@ -16,7 +16,7 @@ end
 
 Base.:(==)(X1::Alignment, X2::Alignment) = all(fn->getfield(X1, fn) == getfield(X2, fn), fieldnames(Alignment))
 
-immutable HarmonizedAlignments
+struct HarmonizedAlignments
     X1::Alignment
     X2::Alignment
 end
@@ -30,7 +30,7 @@ Base.:(==)(H1::HarmonizedAlignments, H2::HarmonizedAlignments) =
 # M contains the number of sequences and their pseudo count
 # matching contains the current matching between the two alignments used to compute FreqC
 # q is the number of different amino acids
-type FreqC
+mutable struct FreqC
     Pij::Matrix{Float64}
     Pi::Vector{Float64}
     specs::Vector{Int}
@@ -49,7 +49,7 @@ end
 # FastC is the type of the correlation matrix
 # specs is the specs presently matched and used for the computation
 # M contains the number of sequences and their pseudo count
-type FastC
+mutable struct FastC
     Cij::Matrix{Float64}
     specs::Vector{Int}
     M::Vector{Int}