public override IEnumerable<string> Process(string useless)
{
var result = new MappedMirnaGroupXmlFileFormat().ReadFromFile(options.InputFile);
using (StreamWriter sw = new StreamWriter(options.OutputFile))
{
sw.WriteLine("miRNA\tLocation\tTotalCount\tPerfectMatch\tMiss5_2\tMiss3_3\tMissInternal");
foreach (var res in result)
{
var items = res.GetAlignedLocations();
if(res.DisplayName.Equals("hsa-mir-486-5p:TCCTGTACTGAGCTGCCCCGAG")){
items.ForEach(m => Console.WriteLine(m.Parent.Qname + "\t" + m.Strand + "\t" + m.MismatchPositions));
}
var pmcount = items.Count(m => m.NumberOfMismatch == 0);
var mis5 = items.Count(m =>
{
SAMAlignedLocation loc = m;
if (loc.NumberOfMismatch == 0)
{
return false;
}
var mp = loc.MismatchPositions;
if (loc.Strand == '-')
{
mp = new string(mp.Reverse().ToArray());
}
return reg5.Match(mp).Success;
});
var mis3 = items.Count(m =>
{
var loc = m;
if (loc.NumberOfMismatch == 0)
{
return false;
}
var mp = loc.MismatchPositions;
if (loc.Strand == '+')
{
mp = new string(mp.Reverse().ToArray());
}
return reg3.Match(mp).Success;
});
sw.WriteLine("{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}", res.DisplayName, res.DisplayLocation, items.Count, pmcount, mis5, mis3, items.Count - pmcount - mis5 - mis3);
}
}
return new string[] { options.OutputFile };
}
public override IEnumerable<string> Process()
{
var countfiles = options.GetCountFiles();
var dic = new Dictionary<string, Dictionary<string, MappedMirnaGroup>>();
foreach (var file in countfiles)
{
Progress.SetMessage("Reading miRNA mapped file " + file.File + " ...");
var mirnas = new MappedMirnaGroupXmlFileFormat().ReadFromFile(file.File);
dic[file.Name] = mirnas.ToDictionary(m => m.DisplayName);
}
var features = (from c in dic.Values
from k in c.Keys
select k).Distinct().OrderBy(m => m).ToList();
var names = dic.Keys.OrderBy(m => m).ToList();
using (StreamWriter sw = new StreamWriter(options.OutputFile))
using (StreamWriter swNTA = new StreamWriter(options.NTAFile))
using (StreamWriter swIso = new StreamWriter(options.IsomirFile))
using (StreamWriter swIsoNTA = new StreamWriter(options.IsomirNTAFile))
{
sw.WriteLine("Feature\tLocation\tSequence\t{0}", names.Merge("\t"));
swNTA.WriteLine("Feature\tLocation\tSequence\t{0}", names.Merge("\t"));
swIso.WriteLine("Feature\tLocation\tSequence\t{0}", names.Merge("\t"));
swIsoNTA.WriteLine("Feature\tLocation\tSequence\t{0}", names.Merge("\t"));
foreach (var feature in features)
{
OutputCount(sw, dic, feature, names, MirnaConsts.NO_OFFSET, false, "");
OutputCount(swNTA, dic, feature, names, MirnaConsts.NO_OFFSET, true, "");
OutputCount(swIso, dic, feature, names, 0, false, "_+_0");
OutputCount(swIso, dic, feature, names, 1, false, "_+_1");
OutputCount(swIso, dic, feature, names, 2, false, "_+_2");
OutputCount(swIsoNTA, dic, feature, names, 0, true, "_+_0");
OutputCount(swIsoNTA, dic, feature, names, 1, true, "_+_1");
OutputCount(swIsoNTA, dic, feature, names, 2, true, "_+_2");
}
}
var result = new[] { options.OutputFile, options.IsomirFile, options.NTAFile, options.IsomirNTAFile }.ToList();
return result;
}
public List<MappedMirnaGroup> Build(string countXmlFile)
{
var result = new MappedMirnaGroupXmlFileFormat().ReadFromFile(countXmlFile);
foreach (var group in result)
{
foreach (var mirna in group)
{
foreach (var region in mirna.MappedRegions)
{
var positions = region.Mapped.Keys.ToList();
foreach (var position in positions)
{
var mapped = region.Mapped[position];
mapped.AlignedLocations.RemoveAll(q =>
{
var snp = q.GetNotGsnapMismatch(q.Parent.Sequence);
if (null == snp)
{
return true;
}
return !snp.IsMutation('T', 'C');
});
if (mapped.AlignedLocations.Count == 0)
{
region.Mapped.Remove(position);
}
}
}
mirna.MappedRegions.RemoveAll(l => l.Mapped.Count == 0);
}
group.RemoveAll(n => n.MappedRegions.Count == 0);
}
result.RemoveAll(m => m.Count == 0);
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 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 MappedMirnaGroupXmlFileFormat();
Progress.SetMessage("reading mapped reads from " + options.ReferenceFile + " ...");
var refitems = format.ReadFromFile(options.ReferenceFile);
var refSpecies = refitems[0][0].Name.StringBefore("-");
Progress.SetMessage("reading mapped reads from " + options.SampleFile + " ...");
var samitems = format.ReadFromFile(options.SampleFile);
var samSpecies = samitems[0][0].Name.StringBefore("-");
var paired = GetPairedMiRNA(refitems, samitems);
//using (StreamWriter sw = new StreamWriter(targetFile))
//{
// sw.WriteLine("microRNA\t{0}_sequence\t{1}_sequence\tbp_difference\tquery_sequence\t{0}_count\t{0}_estimate_count\t{1}_count\t{1}_estimate_count", refName, samName);
// var keys = refitems.Keys.Union(samitems.Keys).Distinct().OrderBy(m => m).ToList();
// foreach (var key in keys)
// {
// sw.Write(key);
// string refseq = string.Empty;
// string samseq = string.Empty;
// MappedMiRNA refmirna = null;
// MappedMiRNA sammirna = null;
// Dictionary<string, List<QueryMapped>> refmirnamap = new Dictionary<string, List<QueryMapped>>();
// Dictionary<string, List<QueryMapped>> sammirnamap = new Dictionary<string, List<QueryMapped>>();
// if (refitems.ContainsKey(key))
// {
// refmirna = refitems[key];
// refseq = refmirna.Name.Contains(":") ? refmirna.Name.StringAfter(":") : string.Empty;
// refmirnamap = ConvertToMap(refmirna);
// }
// if (samitems.ContainsKey(key))
// {
// sammirna = samitems[key];
// samseq = sammirna.Name.Contains(":") ? sammirna.Name.StringAfter(":") : string.Empty;
// sammirnamap = ConvertToMap(sammirna);
// }
// var seqs = refmirnamap.Keys.Union(sammirnamap.Keys).OrderBy(m => m).ToList();
// CombinedSequence cs = null;
// if (!string.IsNullOrEmpty(refseq) && !string.IsNullOrEmpty(samseq))
// {
// cs = MirnaUtils.GetCombinedSequence(refseq, samseq);
// sw.Write("\t{0}\t{1}\t{2}", cs.GetAnnotatedSequence1(), cs.GetAnnotatedSequence2(), cs.MismatchPositions.Length);
// }
// else
// {
// sw.Write("\t{0}\t{1}\t-", refseq, samseq);
// }
// sw.WriteLine();
// foreach (var seq in seqs)
// {
// sw.Write("\t\t\t\t{0}", seq);
// if (refmirnamap.ContainsKey(seq))
// {
// sw.Write("\t{0:0.00}", refmirnamap[seq].Sum(m => m.QueryCount));
// sw.Write("\t{0:0.00}", refmirnamap[seq].Sum(m => m.EsminatedCount));
// }
// else
// {
// sw.Write("\t\t");
// }
// if (sammirnamap.ContainsKey(seq))
// {
// sw.Write("\t{0:0.00}", sammirnamap[seq].Sum(m => m.QueryCount));
// sw.Write("\t{0:0.00}", sammirnamap[seq].Sum(m => m.EsminatedCount));
// }
// else
// {
// sw.Write("\t\t");
// }
// sw.WriteLine();
// }
// }
//}
return new string[] { options.OutputFile };
}
