Restructuring code directories to conform to nimble standard, adding …

…nimble install option, renamed functions to conform with fasta/q migration

.gitignore

@@ -12,6 +12,7 @@
 !**.R
 !**.bsh
 !**.nim
+!**.nims
 !**.nimble
 !**.md
 !**.c

.gitmodules

@@ -1,3 +1,3 @@
 [submodule "threadpools"]
-	path = threadpools
+	path = src/threadpools
 	url = https://github.com/yglukhov/threadpools.git

Makefile

@@ -1,7 +1,8 @@
 NIM = nim c
 
 conduit:
-	cd poaV2 && make
-	$(NIM) -d:release --threads:on --passL:poaV2/liblpo.a --passL:poaV2/align_score.o conduit.nim
-	$(NIM) -d:release conduitUtils.nim
+	cd src/poaV2 && make
+	mkdir bin/
+	cd src/ && $(NIM) -d:release --threads:on --passL:poaV2/liblpo.a --passL:poaV2/align_score.o conduit.nim && mv src/conduit bin/
+	cd src/ && $(NIM) -d:release conduitUtils.nim && mv src/conduitUtils bin/
 

conduit.nimble

@@ -0,0 +1,24 @@
+# Package
+
+version       = "0.1.0"
+author        = "Nathan Roach"
+description   = "De novo transcriptome assembler"
+license       = "GPLv2"
+
+# Dependencies
+
+requires "hts >= 0.3.1", "nim >= 1.0.0"
+
+srcDir = "src/"
+
+before install:
+  echo "Building poaV2"
+  withDir "src/poaV2":
+    exec "make"
+
+bin = @["conduit", "conduitUtils", "conduit_clustering"]
+# skipDirs = @["tests"]
+# skipFiles = @["GT04008021.bam"]
+
+# task test, "run the tests":
+#   exec "nim c --lineDir:on --debuginfo -r tests/all"

conduit.nim → src/conduit.nim

@@ -12,6 +12,8 @@ import hts
 import sets
 import poaV2/header
 import poaV2/poa
+import fasta
+import fastq
 
 {.experimental.}
 
@@ -316,7 +318,7 @@ proc runPOAandCollapsePOGraph(intuple : (string,string,string,string,uint16,uint
       fasta_file = infilepath
   elif format == "fastq":
     fasta_file = &"{outdir}{trim}.tmp.fa"
-    convertFASTQtoFASTA(infilepath,fasta_file)
+    convertFASTQFiletoFASTAfile(infilepath,fasta_file)
   var split_num = 200
   var (num_fastas,_) = splitFASTA2(fasta_file,&"{outdir}{trim}.tmp",split_num = split_num)
   var total_fastas = num_fastas
@@ -414,7 +416,8 @@ proc runGraphBasedIlluminaCorrection(intuple : (string,string,string,uint64,uint
   var records = getFastaRecordsFromTrimmedPOGraph(addr trim_po, representative_paths, read_supports, trim)
   var outfile : File
   discard open(outfile,this_fasta_filepath,fmWrite)
-  writeCorrectedReads(records,outfile)
+  writeFASTArecordsToFile(outfile,records)
+  # writeCorrectedReads(records,outfile)
   outfile.close()
   result = sameFileContent(last_fasta_filepath,this_fasta_filepath)
   removeFile(last_fasta_filepath)
@@ -444,7 +447,8 @@ proc runLinearBasedIlluminaCorrection(intuple : (string,string,uint64,uint64,uin
       corrected.add(FastaRecord(read_id : read.read_id, sequence : getSequenceFromPath(trim_po,trim_po.reads[0].corrected_path)))
   var outfile : File
   discard open(outfile,this_fasta_filepath,fmWrite)
-  writeCorrectedReads(corrected,outfile)
+  writeFASTArecordsToFile(outfile,corrected)
+  # writeCorrectedReads(corrected,outfile)
   outfile.close()
 
 

src/conduit.nims

@@ -0,0 +1,3 @@
+switch("threads", "on")
+switch("passL","src/poaV2/liblpo.a")
+switch("passL","src/poaV2/align_score.o")

conduitUtils.nim → src/conduitUtils.nim

@@ -7,6 +7,8 @@ import tables
 import sets
 import poGraphUtils
 import algorithm
+import fasta
+import fastq
 
 type
   BLASTmatch* = object
@@ -692,9 +694,9 @@ proc compareExactTranslations*(reference_infilepath : string, translation_infile
   var r_infile, t_infile : File
   discard open(r_infile,reference_infilepath,fmRead)
   discard open(t_infile,translation_infilepath,fmRead)
-  let r_records = poGraphUtils.parseFasta(r_infile)
+  let r_records = parseFasta(r_infile)
   r_infile.close()
-  let t_records = poGraphUtils.parseFasta(t_infile)
+  let t_records = parseFasta(t_infile)
   t_infile.close()
   var r_proteins,t_proteins : HashSet[string]
   for record in r_records:
@@ -717,7 +719,7 @@ proc compareBLASTPTranslations*(reference_infilepath : string, blastp_infilepath
   var match_set : HashSet[string]
   var ref_infile : File
   discard open(ref_infile,reference_infilepath,fmRead)
-  let reference_records = poGraphUtils.parseFasta(ref_infile)
+  let reference_records = parseFasta(ref_infile)
   ref_infile.close()
   for record in reference_records:
     reference_id_set.incl(record.read_id)
@@ -1171,7 +1173,8 @@ proc getNovelLociFASTA*(infilepath,gffcompare_infilepath,outfilepath : string,fi
     if record.read_id in novel_loci:
       new_records.add(record)
   infile.close
-  writeCorrectedReads(new_records,outfile)
+  # writeCorrectedReads(new_records,outfile)
+  writeFASTArecordsToFile(outfile,new_records)
   outfile.close
 
 proc parseOptions() : UtilOptions = 
@@ -1392,7 +1395,7 @@ proc main() =
           infile.close()
           translateTranscripts(records,opt.outfilepath,threshold = int(opt.min_length),stranded = opt.stranded)
         else:
-          let records = poGraphUtils.parseFasta(infile)
+          let records = parseFasta(infile)
           infile.close()
           translateTranscripts(records,opt.outfilepath,threshold = int(opt.min_length),stranded = opt.stranded)
       of "strandTranscripts":
@@ -1403,7 +1406,7 @@ proc main() =
           infile.close()
           strandTranscripts(records,opt.outfilepath)
         else:
-          let records = poGraphUtils.parseFasta(infile)
+          let records = parseFasta(infile)
           infile.close()
           strandTranscripts(records,opt.outfilepath)
       of "bed2gtf":

conduit_clustering.nim → src/conduit_clustering.nim

@@ -12,6 +12,7 @@ import sets
 import genomeKDE
 import algorithm
 import poGraphUtils
+import conduitUtils
 import fasta
 import fastq
 
@@ -24,6 +25,7 @@ type
     output_dir : string
     prefix : string
     out_type : string
+    reference : string
 
   SpliceSiteGraph = object
     adjacencies : seq[seq[uint32]]
@@ -91,11 +93,10 @@ proc writeClusterHelp() =
   echo "        Output reads in FASTA format"
 
 proc summaryFromBamRecord( record : Record,
-                           stranded : bool = false)
-                          :
-                          (char,
-                           (uint64, uint64),
-                           seq[(uint64, uint64)]) = 
+                           stranded : bool = false) :
+                            ( char,
+                              (uint64, uint64),
+                              seq[(uint64, uint64)] ) = 
   if record.flag.unmapped:
       return
   let alignment_start = uint64(record.start)
@@ -197,6 +198,9 @@ proc parseOptions() : ClusteringOptions =
   var prefix = &"cluster_"
   var prefix_flag = false
 
+  var reference_filepath = ""
+  var reference_filepath_flag = false
+
   var run_flag = true
   var help_flag = false
   var version_flag = false
@@ -262,6 +266,17 @@ proc parseOptions() : ClusteringOptions =
             else:
               output_format_flag = true
               output_format = "fasta"
+          of "r", "reference":
+            if not reference_filepath_flag:
+              reference_filepath_flag = true
+              if val != "":
+                reference_filepath = val
+              else:
+                last = "reference"
+            else:
+              echo "ERROR - two references provided"
+              help_flag = true
+              break
           of "o", "output-dir":
             if not output_dir_flag:
               output_dir_flag = true
@@ -304,6 +319,8 @@ proc parseOptions() : ClusteringOptions =
             output_dir = key
           of "prefix":
             prefix = key
+          of "reference":
+            reference_filepath = key
           of "":
             if not file_flag:
               file = key
@@ -327,7 +344,8 @@ proc parseOptions() : ClusteringOptions =
                            cluster_zero_introns : single_exon,
                            output_dir : output_dir,
                            prefix : prefix,
-                           out_type : output_format)
+                           out_type : output_format,
+                           reference : reference_filepath)
 
 
 proc bfs(ssgraph : ptr SpliceSiteGraph,
@@ -382,6 +400,7 @@ proc getWeightedSpliceJunctionLocations(
     result[1].add((ss,weight))
   echo result
 
+
 proc writeFASTAsFromBAM(bam : Bam,
                         read_id_to_cluster : ptr Table[string,int],
                         cluster_sizes : ptr CountTable[int],
@@ -406,10 +425,11 @@ proc writeFASTAsFromBAM(bam : Bam,
       result += 1
       # echo &"Opening cluster file {cluster_id + starting_count}"
       open_files[cluster_id] = file
-      file.writeFASTArecordToFile(record)
+      # file.writeFASTArecordToFile(record)
+      file.writeBamRecordToFASTAfile(record)
     else:
       # echo &"Appending to cluster file {cluster_id + starting_count}"
-      open_files[cluster_id].writeFASTArecordToFile(record)
+      open_files[cluster_id].writeBamRecordToFASTAfile(record)
     written_reads.inc(cluster_id)
     echo written_reads[cluster_id], "\t", cluster_sizes[][cluster_id]
     if written_reads[cluster_id] == cluster_sizes[][cluster_id]:
@@ -424,13 +444,13 @@ proc writeFASTAsFromBAM(bam : Bam,
 proc correctBamRecordWithGenome(record : Record, fai : Fai) : FastaRecord = 
   let summary = summaryFromBamRecord(record, true)
   var nt_sequence : string
-  var start_base = summary[1][0]
+  var start_base = int(summary[1][0])
   var end_base = -1
-  for donor,acceptor in summary[2]:
-    end_base = donor
+  for (donor,acceptor) in summary[2]:
+    end_base = int(donor)
     nt_sequence.add(fai.get(record.chrom,start_base,end_base))
-    start_base = acceptor
-  end_base = summary[1][1]
+    start_base = int(acceptor)
+  end_base = int(summary[1][1])
   nt_sequence.add(fai.get(record.chrom,start_base,end_base))
   if summary[0] == '-':
     nt_sequence = nt_sequence.revComp

src/conduit_clustering.nims

@@ -0,0 +1 @@
+switch("threads", "on")