public void TestXml()
{
var items = new MappedItemGroupXmlFileFormat().ReadFromFile("../../../data/mappedgroup.xml");
Assert.AreEqual(2, items.Count);
var query = items.GetQueries();
Assert.AreEqual(1, query.Count);
var sam = query[0];
Assert.AreEqual(20, sam.Locations.Count);
Assert.AreEqual(2, sam.QueryCount);
Assert.AreEqual(2, items[0].Count);
var s1 = items[0][0];
var s2 = items[0][1];
Assert.AreEqual(1, s1.MappedRegions.Count);
Assert.AreEqual(2, s2.MappedRegions.Count);
Assert.AreEqual(2, items[0].QueryCount);
Assert.AreEqual(1.5, items[0].GetEstimatedCount());
Assert.AreEqual(2, items[1].QueryCount);
Assert.AreEqual(0.5, items[1].GetEstimatedCount());
}
public override IEnumerable<string> Process()
{
var format = new MappedItemGroupXmlFileFormat();
Progress.SetMessage("reading mapped reads from " + _options.CountFile + " ...");
var mapped = format.ReadFromFile(_options.CountFile);
var sequenceLocusSet = new HashSet<string>(from item in mapped
from mi in item
from mr in mi.MappedRegions
from al in mr.AlignedLocations
select string.Format("{0}:{1}:{2}", al.Parent.Sequence, al.Seqname, al.Start));
Progress.SetMessage("There are {0} unique sequence:locus", sequenceLocusSet.Count);
using (var sw = new StreamWriter(_options.OutputFile))
{
using (var sr = SAMFactory.GetReader(_options.BamFile, false))
{
sr.ReadHeaders().ForEach(m => sw.WriteLine(m));
int count = 0;
int accepted = 0;
string line;
while ((line = sr.ReadLine()) != null)
{
if (count % 1000 == 0)
{
if (Progress.IsCancellationPending())
{
throw new UserTerminatedException();
}
}
if (count % 100000 == 0 && count > 0)
{
Progress.SetMessage("{0} candidates from {1} reads", accepted, count);
}
count++;
var parts = line.Split('\t');
var locus = string.Format("{0}:{1}:{2}", parts[SAMFormatConst.SEQ_INDEX], parts[SAMFormatConst.RNAME_INDEX], parts[SAMFormatConst.POS_INDEX]);
if (!sequenceLocusSet.Contains(locus))
{
continue;
}
sw.WriteLine(line);
accepted++;
}
}
}
return new[] { _options.OutputFile };
}
private static void SaveItems(List<MappedItemGroup> items1, string outputFile, MappedItemGroupSequenceWriter writer,
MappedItemGroupXmlFileFormat format, List<string> result)
{
items1.RemoveAll(m => m.QueryCount == 0);
var xml1 = outputFile + ".xml";
if (items1.Any(m => m.Name.Contains(".tRNA")))
{
items1.SortTRna();
}
writer.WriteToFile(outputFile, items1);
format.WriteToFile(xml1, items1);
result.Add(outputFile);
result.Add(xml1);
}
public override IEnumerable<string> Process()
{
var result = new List<string>();
var samformat = _options.GetEngineFormat();
var format = new MappedItemGroupXmlFileFormat();
Progress.SetMessage("reading mapped reads from " + _options.InputFile1 + " ...");
var items1 = format.ReadFromFile(_options.InputFile1);
Progress.SetMessage("reading mapped reads from " + _options.InputFile2 + " ...");
var items2 = format.ReadFromFile(_options.InputFile2);
var reads1 = items1.GetQueries().ToDictionary(m => m.Qname);
var reads2 = items2.GetQueries().ToDictionary(m => m.Qname);
var qnames = reads1.Keys.Union(reads2.Keys).Distinct().ToList();
foreach (var qname in qnames)
{
if (!reads1.ContainsKey(qname) || !reads2.ContainsKey(qname))
continue;
var r1 = reads1[qname];
var r2 = reads2[qname];
var res = samformat.CompareScore(r1.AlignmentScore, r2.AlignmentScore);
if (res == 0)
{
items1.RemoveRead(qname);
items2.RemoveRead(qname);
}
else if (res < 0)
{
items2.RemoveRead(qname);
}
else
{
items1.RemoveRead(qname);
}
}
var writer = new MappedItemGroupSequenceWriter();
SaveItems(items1, _options.OutputFile1, writer, format, result);
SaveItems(items2, _options.OutputFile2, writer, format, result);
return result;
}
public override IEnumerable <string> Process()
{
var entries = (from line in File.ReadAllLines(options.InputFile)
let parts = line.Split('\t')
where parts.Length >= 3
let mirna = parts.Length == 3 ? string.Empty : parts[3]
select new { GroupName = parts[0], SampleName = parts[1], SmallRNAFile = parts[2], MiRNAFile = mirna }).ToList();
if (entries.All(m => !File.Exists(m.MiRNAFile)))
{
return(new SmallRNACategoryGroupPlusBuilder(options)
{
Progress = this.Progress
}.Process());
}
var groups = entries.GroupBy(m => m.GroupName).ToList();
var result = new List <string>();
foreach (var group in groups)
{
var catfile = Path.Combine(options.OutputDirectory, group.Key + ".catcount");
result.Add(catfile);
using (var sw = new StreamWriter(catfile))
{
sw.WriteLine("SampleName\tCategory\tLevel\tCount");
foreach (var entry in group)
{
Progress.SetMessage("Reading smallRNA mapped file " + entry.SmallRNAFile + " ...");
var others = new MappedItemGroupXmlFileFormat().ReadFromFile(entry.SmallRNAFile);
var otherQueries = (from g in others
from m in g
from mr in m.MappedRegions
from loc in mr.AlignedLocations
select new QueryRecord(loc.Parent.Qname,
m.Name.StringBefore(":"),
m.Name.StringAfter(":").StringBefore(":"),
m.Name.StringAfter(":").StringAfter(":"),
loc.Parent.QueryCount)).ToGroupDictionary(m => m.Query);
Progress.SetMessage("Reading smallRNA mapped file finished, {0} queries mapped.", otherQueries.Count);
//2570-KCV-01-19.bam.count.mapped.xml => 2570-KCV-01-19.bam.info
var infofile = Path.Combine(Path.GetDirectoryName(entry.SmallRNAFile), Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(entry.SmallRNAFile))) + ".info");
if (File.Exists(entry.MiRNAFile))
{
infofile = Path.Combine(Path.GetDirectoryName(entry.MiRNAFile), Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(entry.MiRNAFile))) + ".info");
Progress.SetMessage("Reading miRNA mapped file " + entry.MiRNAFile + " ...");
var mirnas = new MappedMirnaGroupXmlFileFormat().ReadFromFile(entry.MiRNAFile);
var mirnaQueries = (from g in mirnas
from m in g
from mr in m.MappedRegions
from mapped in mr.Mapped.Values
from loc in mapped.AlignedLocations
select new QueryRecord(loc.Parent.Qname.StringBefore(":CLIP_"),
"miRNA",
"miRNA",
m.Name,
loc.Parent.QueryCount)).ToGroupDictionary(m => m.Query);
Progress.SetMessage("Reading miRNA mapped file finished, {0} queries mapped.", mirnaQueries.Count);
foreach (var q in mirnaQueries)
{
List <QueryRecord> rec;
if (!otherQueries.TryGetValue(q.Key, out rec))
{
rec = q.Value;
otherQueries[q.Key] = q.Value;
}
else
{
rec.AddRange(q.Value);
}
}
Progress.SetMessage("Total {0} queries mapped.", otherQueries.Count);
}
var counts = new List <CategoryCount>();
FillCounts(counts, options.Categories, otherQueries);
var othercategories = (from v in otherQueries.Values
from item in v
select item.Biotype).Distinct().OrderBy(m => m).ToList();
FillCounts(counts, othercategories, otherQueries);
if (File.Exists(infofile))
{
var lines = File.ReadAllLines(infofile);
Progress.SetMessage("reading mapping information from " + infofile + " ...");
int totalReads = 0;
int mappedReads = 0;
foreach (var line in lines)
{
if (line.StartsWith("TotalReads"))
{
totalReads = int.Parse(line.StringAfter("\t"));
}
else if (line.StartsWith("MappedReads"))
{
mappedReads = int.Parse(line.StringAfter("\t"));
}
}
var smallRNAReads = counts.Sum(m => m.Count);
sw.WriteLine("{0}\tTotal Reads\t0\t{1}", entry.SampleName, totalReads);
sw.WriteLine("{0}\tMapped Reads\t0\t{1}", entry.SampleName, mappedReads);
sw.WriteLine("{0}\tsmall RNA\t0\t{1}", entry.SampleName, smallRNAReads);
sw.WriteLine("{0}\tUnmapped\t1\t{1}", entry.SampleName, totalReads - mappedReads);
sw.WriteLine("{0}\tOther Mapped\t1\t{1}", entry.SampleName, mappedReads - smallRNAReads);
sw.WriteLine("{0}\tsmall RNA\t1\t{1}", entry.SampleName, smallRNAReads);
}
foreach (var rec in counts)
{
sw.WriteLine("{0}\t{1}\t{2}\t{3}", entry.SampleName, rec.Biotype, 2, rec.Count);
}
}
}
var data = (from line in File.ReadAllLines(catfile).Skip(1)
where !string.IsNullOrWhiteSpace(line)
let parts = line.Split('\t')
let level = double.Parse(parts[2])
where !(parts[1].Equals("small RNA") && level == 1)
select new
{
SampleName = parts[0],
Category = parts[1],
Level = level,
Count = int.Parse(parts[3])
}).ToList();
var tablefile = catfile + ".tsv";
result.Add(tablefile);
using (var sw = new StreamWriter(tablefile))
{
var samples = (from d in data
select d.SampleName).Distinct().OrderBy(m => m).ToList();
sw.WriteLine("Category\t{0}", samples.Merge("\t"));
var categories = (from d in data
where d.Level == 2
select d.Category).Distinct().OrderBy(m => m).ToList();
categories.Insert(0, "small RNA");
categories.Insert(0, "Other Mapped");
categories.Insert(0, "Unmapped");
categories.Insert(0, "Mapped Reads");
categories.Insert(0, "Total Reads");
Console.WriteLine(categories.Merge("\n"));
var map = data.ToDoubleDictionary(m => m.SampleName, m => m.Category);
foreach (var cat in categories)
{
sw.WriteLine("{0}\t{1}", cat,
(from sample in samples
let dic = map[sample]
select dic.ContainsKey(cat) ? dic[cat].Count.ToString() : "").Merge("\t"));
}
}
var rfile = new FileInfo(FileUtils.GetTemplateDir() + "/smallrna_category_group.r").FullName;
if (File.Exists(rfile))
{
var targetrfile = catfile + ".r";
using (var sw = new StreamWriter(targetrfile))
{
sw.WriteLine("catfile<-\"{0}\"", catfile);
sw.WriteLine("outputdir<-\"{0}\"", options.OutputDirectory);
sw.WriteLine("ispdf<-{0}", options.PdfGraph ? "1" : "0");
string line = File.ReadAllText(rfile);
using (var sr = new StreamReader(rfile))
{
if (line.Contains("#predefine_end"))
{
while ((line = sr.ReadLine()) != null)
{
if (line.Contains("#predefine_end"))
{
break;
}
}
}
while ((line = sr.ReadLine()) != null)
{
sw.WriteLine(line);
}
}
}
SystemUtils.Execute("R", "--vanilla --slave -f \"" + targetrfile + "\"");
}
}
return(result);
}
public override IEnumerable<string> Process()
{
var format = new MappedItemGroupXmlFileFormat();
using (StreamWriter sw = new StreamWriter(options.OutputFile))
{
sw.WriteLine("File\tFeature\tStrand\tCount\tPosition\tPercentage");
foreach (var file in options.GetCountFiles())
{
var xmlfile = file.File.EndsWith(".xml") ? file.File : file.File + ".mapped.xml";
var count = format.ReadFromFile(xmlfile).OrderByDescending(m => m.GetEstimatedCount()).ToList();
foreach (var group in count)
{
var item = group[0];
Dictionary<long, double> positionCount = new Dictionary<long, double>();
foreach (var region in item.MappedRegions)
{
foreach (var loc in region.AlignedLocations)
{
for (long p = loc.Start; p <= loc.End; p++)
{
var offset = region.Region.Strand == '+' ? p - region.Region.Start + 1 : region.Region.End - p + 1;
double v;
if (!positionCount.TryGetValue(offset, out v))
{
v = 0;
}
positionCount[offset] = v + loc.Parent.GetEstimatedCount();
}
}
}
var allcount = item.GetEstimatedCount();
var keys = positionCount.Keys.ToList();
keys.Sort();
foreach (var key in keys)
{
sw.WriteLine("{0}\t{1}\t{2}\t{3:0.##}\t{4}\t{5:0.00}",
file.Name,
item.Name,
item.MappedRegions.First().Region.Strand,
item.GetEstimatedCount(),
key,
positionCount[key] / allcount);
}
}
}
}
var rfile = new FileInfo(FileUtils.GetTemplateDir() + "/smallrna_position.r").FullName;
if (File.Exists(rfile))
{
var targetr = Path.ChangeExtension(options.OutputFile, ".r").Replace("\\", "/");
var content = File.ReadAllText(rfile).Replace("$$workspace", Path.GetDirectoryName(Path.GetFullPath(options.OutputFile)).Replace("\\", "/"))
.Replace("$$positionfile", Path.GetFileName(options.OutputFile).Replace("\\", "/"));
File.WriteAllText(targetr, content);
if (File.Exists(targetr))
{
SystemUtils.Execute("R", "--vanilla -f " + targetr);
}
}
return new string[] { Path.GetFullPath(options.OutputFile) };
}
public override IEnumerable <string> Process()
{
var result = new List <string>();
//read regions
var featureLocations = options.GetSequenceRegions();
Progress.SetMessage("There are {0} coordinate entries", featureLocations.Count);
if (featureLocations.Count == 0)
{
throw new Exception(string.Format("No coordinate found in file {0}", options.CoordinateFile));
}
var fGroups = featureLocations.GroupBy(l => l.Category).OrderByDescending(l => l.Count()).ToList();
foreach (var fg in fGroups)
{
Console.WriteLine("{0} = {1}", fg.Key, fg.Count());
}
var featureChroms = new HashSet <string>(from feature in featureLocations
select feature.Seqname);
var resultFilename = options.OutputFile;
result.Add(resultFilename);
HashSet <string> cca = new HashSet <string>();
if (File.Exists(options.CCAFile))
{
cca = new HashSet <string>(File.ReadAllLines(options.CCAFile));
}
//parsing reads
List <QueryInfo> totalQueries;
var reads = ParseCandidates(options.InputFiles, resultFilename, out totalQueries);
if (reads.Count == 0)
{
throw new ArgumentException("No read found in file " + options.InputFiles.Merge(","));
}
HashSet <string> excludeQueries = new HashSet <string>();
if (!string.IsNullOrEmpty(options.ExcludeXml))
{
Progress.SetMessage("Excluding queries in {0} ...", options.ExcludeXml);
excludeQueries = new HashSet <string>(from q in MappedItemGroupXmlFileFormat.ReadQueries(options.ExcludeXml)
select q.StringBefore(SmallRNAConsts.NTA_TAG));
reads.RemoveAll(m => excludeQueries.Contains(m.Locations.First().Parent.Qname.StringBefore(SmallRNAConsts.NTA_TAG)));
Progress.SetMessage("Total candidate {0} for mapping ...", reads.Count);
}
var hasMicroRnaNTA = reads.Any(l => l.NTA.Length > 0);
var hasTrnaNTA = hasMicroRnaNTA || File.Exists(options.CCAFile);
if (!options.NoCategory)
{
//First of all, draw candidate mapping position graph
var miRNAPositionFile = Path.ChangeExtension(options.OutputFile, SmallRNAConsts.miRNA + ".candidates.position");
if (!options.NotOverwrite || !File.Exists(miRNAPositionFile))
{
Progress.SetMessage("Drawing microRNA candidates position pictures...");
var notNTAreads = hasMicroRnaNTA ? reads.Where(m => m.NTA.Length == 0).ToList() : reads;
DrawPositionImage(notNTAreads, featureLocations.Where(m => m.Category.Equals(SmallRNAConsts.miRNA)).ToList(), SmallRNABiotype.miRNA.ToString(), miRNAPositionFile);
}
}
var featureGroups = new List <FeatureItemGroup>();
var mappedfile = resultFilename + ".mapped.xml";
if (File.Exists(mappedfile) && options.NotOverwrite)
{
Progress.SetMessage("Reading mapped feature items...");
featureGroups = new FeatureItemGroupXmlFormat().ReadFromFile(mappedfile);
}
else
{
Progress.SetMessage("Mapping feature items...");
//mapping reads to features based on miRNA, tRNA, mt_tRNA and other smallRNA priority
MapReadToSequenceRegion(featureLocations, reads, cca, hasMicroRnaNTA, hasTrnaNTA);
var featureMapped = featureLocations.GroupByName();
featureMapped.RemoveAll(m => m.GetEstimatedCount() == 0);
featureMapped.ForEach(m => m.CombineLocations());
if (options.NoCategory)
{
featureGroups = featureMapped.GroupByIdenticalQuery();
}
else
{
var mirnaGroups = featureMapped.Where(m => m.Name.StartsWith(SmallRNAConsts.miRNA)).GroupBySequence();
if (mirnaGroups.Count > 0)
{
OrderFeatureItemGroup(mirnaGroups);
Progress.SetMessage("writing miRNA count ...");
var mirnaCountFile = Path.ChangeExtension(resultFilename, "." + SmallRNAConsts.miRNA + ".count");
new SmallRNACountMicroRNAWriter(options.Offsets).WriteToFile(mirnaCountFile, mirnaGroups);
result.Add(mirnaCountFile);
featureGroups.AddRange(mirnaGroups);
var positionFile = Path.ChangeExtension(options.OutputFile, SmallRNAConsts.miRNA + ".position");
SmallRNAMappedPositionBuilder.Build(mirnaGroups, Path.GetFileNameWithoutExtension(options.OutputFile), positionFile, m => m[0].Name.StringAfter(":"));
}
mirnaGroups.Clear();
var trnaCodeGroups = featureMapped.Where(m => m.Name.StartsWith(SmallRNAConsts.tRNA)).GroupByFunction(SmallRNAUtils.GetTrnaAnticodon, false);
if (trnaCodeGroups.Count > 0)
{
OrderFeatureItemGroup(trnaCodeGroups);
Progress.SetMessage("writing tRNA code count ...");
var trnaCodeCountFile = Path.ChangeExtension(resultFilename, "." + SmallRNAConsts.tRNA + ".count");
new FeatureItemGroupCountWriter(m => m.DisplayNameWithoutCategory).WriteToFile(trnaCodeCountFile, trnaCodeGroups);
result.Add(trnaCodeCountFile);
featureGroups.AddRange(trnaCodeGroups);
var positionFile = Path.ChangeExtension(options.OutputFile, SmallRNAConsts.tRNA + ".position");
SmallRNAMappedPositionBuilder.Build(trnaCodeGroups, Path.GetFileName(options.OutputFile), positionFile, m => m[0].Name.StringAfter(":"));
}
trnaCodeGroups.Clear();
var otherFeatures = featureMapped.Where(m => !m.Name.StartsWith(SmallRNAConsts.miRNA) && !m.Name.StartsWith(SmallRNAConsts.tRNA)).ToList();
var exportBiotypes = SmallRNAUtils.GetOutputBiotypes(options);
foreach (var biotype in exportBiotypes)
{
WriteGroups(result, resultFilename, featureGroups, otherFeatures, biotype);
}
var leftFeatures = otherFeatures.Where(l => !exportBiotypes.Any(b => l.Name.StartsWith(b))).ToList();
WriteGroups(result, resultFilename, featureGroups, leftFeatures, null);
}
Progress.SetMessage("writing all smallRNA count ...");
new FeatureItemGroupCountWriter().WriteToFile(resultFilename, featureGroups);
result.Add(resultFilename);
Progress.SetMessage("writing mapping details...");
new FeatureItemGroupXmlFormatHand().WriteToFile(mappedfile, featureGroups);
}
var readSummary = GetReadSummary(featureGroups, excludeQueries, reads, totalQueries);
WriteInfoFile(result, resultFilename, readSummary, featureGroups);
result.Add(mappedfile);
Progress.End();
return(result);
}
public override IEnumerable<string> Process()
{
var entries = (from line in File.ReadAllLines(options.InputFile)
let parts = line.Split('\t')
where parts.Length >= 3
let mirna = parts.Length == 3 ? string.Empty : parts[3]
select new { GroupName = parts[0], SampleName = parts[1], SmallRNAFile = parts[2], MiRNAFile = mirna }).ToList();
if (entries.All(m => !File.Exists(m.MiRNAFile)))
{
return new SmallRNACategoryGroupPlusBuilder(options)
{
Progress = this.Progress
}.Process();
}
var groups = entries.GroupBy(m => m.GroupName).ToList();
var result = new List<string>();
foreach (var group in groups)
{
var catfile = Path.Combine(options.OutputDirectory, group.Key + ".catcount");
result.Add(catfile);
using (var sw = new StreamWriter(catfile))
{
sw.WriteLine("SampleName\tCategory\tLevel\tCount");
foreach (var entry in group)
{
Progress.SetMessage("Reading smallRNA mapped file " + entry.SmallRNAFile + " ...");
var others = new MappedItemGroupXmlFileFormat().ReadFromFile(entry.SmallRNAFile);
var otherQueries = (from g in others
from m in g
from mr in m.MappedRegions
from loc in mr.AlignedLocations
select new QueryRecord(loc.Parent.Qname,
m.Name.StringBefore(":"),
m.Name.StringAfter(":").StringBefore(":"),
m.Name.StringAfter(":").StringAfter(":"),
loc.Parent.QueryCount)).ToGroupDictionary(m => m.Query);
Progress.SetMessage("Reading smallRNA mapped file finished, {0} queries mapped.", otherQueries.Count);
//2570-KCV-01-19.bam.count.mapped.xml => 2570-KCV-01-19.bam.info
var infofile = Path.Combine(Path.GetDirectoryName(entry.SmallRNAFile), Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(entry.SmallRNAFile))) + ".info");
if (File.Exists(entry.MiRNAFile))
{
infofile = Path.Combine(Path.GetDirectoryName(entry.MiRNAFile), Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(Path.GetFileNameWithoutExtension(entry.MiRNAFile))) + ".info");
Progress.SetMessage("Reading miRNA mapped file " + entry.MiRNAFile + " ...");
var mirnas = new MappedMirnaGroupXmlFileFormat().ReadFromFile(entry.MiRNAFile);
var mirnaQueries = (from g in mirnas
from m in g
from mr in m.MappedRegions
from mapped in mr.Mapped.Values
from loc in mapped.AlignedLocations
select new QueryRecord(loc.Parent.Qname.StringBefore(":CLIP_"),
"miRNA",
"miRNA",
m.Name,
loc.Parent.QueryCount)).ToGroupDictionary(m => m.Query);
Progress.SetMessage("Reading miRNA mapped file finished, {0} queries mapped.", mirnaQueries.Count);
foreach (var q in mirnaQueries)
{
List<QueryRecord> rec;
if (!otherQueries.TryGetValue(q.Key, out rec))
{
rec = q.Value;
otherQueries[q.Key] = q.Value;
}
else
{
rec.AddRange(q.Value);
}
}
Progress.SetMessage("Total {0} queries mapped.", otherQueries.Count);
}
var counts = new List<CategoryCount>();
FillCounts(counts, options.Categories, otherQueries);
var othercategories = (from v in otherQueries.Values
from item in v
select item.Biotype).Distinct().OrderBy(m => m).ToList();
FillCounts(counts, othercategories, otherQueries);
if (File.Exists(infofile))
{
var lines = File.ReadAllLines(infofile);
Progress.SetMessage("reading mapping information from " + infofile + " ...");
int totalReads = 0;
int mappedReads = 0;
foreach (var line in lines)
{
if (line.StartsWith("TotalReads"))
{
totalReads = int.Parse(line.StringAfter("\t"));
}
else if (line.StartsWith("MappedReads"))
{
mappedReads = int.Parse(line.StringAfter("\t"));
}
}
var smallRNAReads = counts.Sum(m => m.Count);
sw.WriteLine("{0}\tTotal Reads\t0\t{1}", entry.SampleName, totalReads);
sw.WriteLine("{0}\tMapped Reads\t0\t{1}", entry.SampleName, mappedReads);
sw.WriteLine("{0}\tsmall RNA\t0\t{1}", entry.SampleName, smallRNAReads);
sw.WriteLine("{0}\tUnmapped\t1\t{1}", entry.SampleName, totalReads - mappedReads);
sw.WriteLine("{0}\tOther Mapped\t1\t{1}", entry.SampleName, mappedReads - smallRNAReads);
sw.WriteLine("{0}\tsmall RNA\t1\t{1}", entry.SampleName, smallRNAReads);
}
foreach (var rec in counts)
{
sw.WriteLine("{0}\t{1}\t{2}\t{3}", entry.SampleName, rec.Biotype, 2, rec.Count);
}
}
}
var data = (from line in File.ReadAllLines(catfile).Skip(1)
where !string.IsNullOrWhiteSpace(line)
let parts = line.Split('\t')
let level = double.Parse(parts[2])
where !(parts[1].Equals("small RNA") && level == 1)
select new
{
SampleName = parts[0],
Category = parts[1],
Level = level,
Count = int.Parse(parts[3])
}).ToList();
var tablefile = catfile + ".tsv";
result.Add(tablefile);
using (var sw = new StreamWriter(tablefile))
{
var samples = (from d in data
select d.SampleName).Distinct().OrderBy(m => m).ToList();
sw.WriteLine("Category\t{0}", samples.Merge("\t"));
var categories = (from d in data
where d.Level == 2
select d.Category).Distinct().OrderBy(m => m).ToList();
categories.Insert(0, "small RNA");
categories.Insert(0, "Other Mapped");
categories.Insert(0, "Unmapped");
categories.Insert(0, "Mapped Reads");
categories.Insert(0, "Total Reads");
Console.WriteLine(categories.Merge("\n"));
var map = data.ToDoubleDictionary(m => m.SampleName, m => m.Category);
foreach (var cat in categories)
{
sw.WriteLine("{0}\t{1}", cat,
(from sample in samples
let dic = map[sample]
select dic.ContainsKey(cat) ? dic[cat].Count.ToString() : "").Merge("\t"));
}
}
var rfile = new FileInfo(FileUtils.GetTemplateDir() + "/smallrna_category_group.r").FullName;
if (File.Exists(rfile))
{
var targetrfile = catfile + ".r";
using (var sw = new StreamWriter(targetrfile))
{
sw.WriteLine("catfile<-\"{0}\"", catfile);
sw.WriteLine("outputdir<-\"{0}\"", options.OutputDirectory);
sw.WriteLine("ispdf<-{0}", options.PdfGraph ? "1" : "0");
string line = File.ReadAllText(rfile);
using (var sr = new StreamReader(rfile))
{
if (line.Contains("#predefine_end"))
{
while ((line = sr.ReadLine()) != null)
{
if (line.Contains("#predefine_end"))
{
break;
}
}
}
while ((line = sr.ReadLine()) != null)
{
sw.WriteLine(line);
}
}
}
SystemUtils.Execute("R", "--vanilla --slave -f \"" + targetrfile + "\"");
}
}
return result;
}
public void Test()
{
var items = new MappedItemGroupXmlFileFormat().ReadFromFile("../../../data/2570-KCV-01-19.bam.count.mapped.trna.xml");
Assert.AreEqual(310, items.Count);
}
