public List <FilterItem> ReadFromFile(string fileName)
{
var anns = new AnnotationFormat().ReadFromFile(fileName);
var result = new List <FilterItem>();
foreach (var ann in anns)
{
var item = new FilterItem();
item.Chr = ann.Annotations["chr"] as string;
item.Start = ann.Annotations["start"] as string;
item.End = ann.Annotations["end"] as string;
item.MajorAllele = ann.Annotations["major_allele"] as string;
item.MinorAllele = ann.Annotations["minor_allele"] as string;
item.ReferenceAllele = ann.Annotations["ref_allele"] as string;
item.NormalMajorCount = int.Parse(ann.Annotations["normal_major_count"] as string);
item.NormalMinorCount = int.Parse(ann.Annotations["normal_minor_count"] as string);
item.TumorMajorCount = int.Parse(ann.Annotations["tumor_major_count"] as string);
item.TumorMinorCount = int.Parse(ann.Annotations["tumor_minor_count"] as string);
item.FisherGroup = double.Parse(ann.Annotations["fisher_group"] as string);
item.FisherNormal = ann.Annotations["fisher_normal"] as string;
item.BrglmConverged = ann.Annotations["brglm_converged"] as string;
item.BrglmGroup = double.Parse(ann.Annotations["brglm_group"] as string);
item.BrglmScore = ann.Annotations["brglm_score"] as string;
item.BrglmStrand = ann.Annotations["brglm_strand"] as string;
item.BrglmPosition = ann.Annotations["brglm_position"] as string;
item.BrglmGroupFdr = double.Parse(ann.Annotations["brglm_group_fdr"] as string);
item.Filter = ann.Annotations["filter"] as string;
item.Identity = ann.Annotations["Identity"] as string;
result.Add(item);
}
return(result);
}
public FindParticipantProtein(string clinicShipmentPortionFile)
{
var anns = new AnnotationFormat().ReadFromFile(clinicShipmentPortionFile);
maps = anns.ToDictionary(m => m.Annotations["bcr_shipment_portion_uuid"].ToString().ToLower(),
m => m.Annotations["bcr_sample_barcode"].ToString());
}
public FindParticipantProtein(string clinicShipmentPortionFile)
{
var anns = new AnnotationFormat().ReadFromFile(clinicShipmentPortionFile);
maps = anns.ToDictionary(m => m.Annotations["bcr_shipment_portion_uuid"].ToString().ToLower(),
m => m.Annotations["bcr_sample_barcode"].ToString());
}
private static void Format <T>(StringBuilder sb, IEnumerable <T> nags, AnnotationFormat annotationFormat)
where T : struct, Enum
{
foreach (var nag in nags)
{
Format(sb, nag, annotationFormat).Append(" ");
}
}
public string ToString(AnnotationFormat format)
{
StringBuilder sb = new StringBuilder();
Format(sb, MoveNAG, format);
Format(sb, PositionalNAGs, format);
Format(sb, TimeTroubleNAG, format);
Format(sb, NonStandardNAGs, format);
return(sb.ToString().Trim());
}
public void TestReadingAnnotationsForOneId()
{
using (var reader = new StreamReader(new MemoryStream(Encoding.UTF8.GetBytes(Examples.MouseUniprotAnnotations))))
{
var annotations = AnnotationFormat.ReadMappings(reader, 0, new [] { 1, 21 }).ToDictionary(annot => annot.fromId, annot => annot.toIds);
Assert.AreEqual("Htr2c", annotations["p34968"][1][0]);
Assert.AreEqual("Htr2c", annotations["b1atn5"][1][0]);
Assert.AreEqual("Abca7", annotations["q91v24"][1][0]);
Assert.AreEqual("Abhd10", annotations["q8k188"][1][0]);
}
}
public static List <SomaticItem> ParseAnnovarFile(string fileName, Regex reg, string scorePrefix, Func <string, bool> acceptChromosome)
{
var result = new List <SomaticItem>();
var bar = Path.GetFileName(fileName).StringBefore(".");
var annos = new AnnotationFormat("^#").ReadFromFile(fileName);
var headers = File.ReadAllLines(fileName).Where(l => !l.StartsWith("#")).First().Split('\t');
foreach (var ann in annos)
{
//Chr Start End Ref Alt Func.refGene Gene.refGene GeneDetail.refGene ExonicFunc.refGene AAChange.refGene snp138 cosmic70 FILTER INFO FORMAT H_LS-A7-A0D9-01A-31W-A071-09-1 H_LS-A7-A0D9-10A-01W-A071-09-1
var chr = ann.Annotations["Chr"].ToString();
if (!acceptChromosome(chr))
{
continue;
}
var m1 = reg.Match(ann.Annotations[headers[headers.Length - 2]].ToString());
var m1Major = int.Parse(m1.Groups[1].Value);
var m1Minor = int.Parse(m1.Groups[2].Value);
var m2 = reg.Match(ann.Annotations[headers[headers.Length - 1]].ToString());
var m2Major = int.Parse(m2.Groups[1].Value);
var m2Minor = int.Parse(m2.Groups[2].Value);
var isNormalFirst = (((double)m1Minor) / (m1Major + m1Minor)) < (((double)m2Minor) / (m2Major + m2Minor));
var info = ann.Annotations["INFO"].ToString();
var item = new SomaticItem()
{
Sample = bar,
Chrom = chr,
StartPosition = int.Parse(ann.Annotations["Start"].ToString()),
RefAllele = ann.Annotations["Ref"].ToString(),
AltAllele = ann.Annotations["Alt"].ToString(),
NormalMajorCount = isNormalFirst ? m1Major : m2Major,
NormalMinorCount = isNormalFirst ? m1Minor : m2Minor,
TumorMajorCount = isNormalFirst ? m2Major : m1Major,
TumorMinorCount = isNormalFirst ? m2Minor : m1Minor,
Score = double.Parse(info.StringAfter(scorePrefix)),
RefGeneFunc = GetDictionaryValue(ann.Annotations, "Func.refGene", string.Empty),
RefGeneName = GetDictionaryValue(ann.Annotations, "Gene.refGene", string.Empty),
RefGeneExonicFunc = GetDictionaryValue(ann.Annotations, "ExonicFunc.refGene", string.Empty)
};
result.Add(item);
}
return(result);
}
public static List<SomaticItem> ParseAnnovarFile(string fileName, Regex reg, string scorePrefix, Func<string, bool> acceptChromosome)
{
var result = new List<SomaticItem>();
var bar = Path.GetFileName(fileName).StringBefore(".");
var annos = new AnnotationFormat("^#").ReadFromFile(fileName);
var headers = File.ReadAllLines(fileName).Where(l => !l.StartsWith("#")).First().Split('\t');
foreach (var ann in annos)
{
//Chr Start End Ref Alt Func.refGene Gene.refGene GeneDetail.refGene ExonicFunc.refGene AAChange.refGene snp138 cosmic70 FILTER INFO FORMAT H_LS-A7-A0D9-01A-31W-A071-09-1 H_LS-A7-A0D9-10A-01W-A071-09-1
var chr = ann.Annotations["Chr"].ToString();
if (!acceptChromosome(chr))
{
continue;
}
var m1 = reg.Match(ann.Annotations[headers[headers.Length - 2]].ToString());
var m1Major = int.Parse(m1.Groups[1].Value);
var m1Minor = int.Parse(m1.Groups[2].Value);
var m2 = reg.Match(ann.Annotations[headers[headers.Length - 1]].ToString());
var m2Major = int.Parse(m2.Groups[1].Value);
var m2Minor = int.Parse(m2.Groups[2].Value);
var isNormalFirst = (((double)m1Minor) / (m1Major + m1Minor)) < (((double)m2Minor) / (m2Major + m2Minor));
var info = ann.Annotations["INFO"].ToString();
var item = new SomaticItem()
{
Sample = bar,
Chrom = chr,
StartPosition = int.Parse(ann.Annotations["Start"].ToString()),
RefAllele = ann.Annotations["Ref"].ToString(),
AltAllele = ann.Annotations["Alt"].ToString(),
NormalMajorCount = isNormalFirst ? m1Major : m2Major,
NormalMinorCount = isNormalFirst ? m1Minor : m2Minor,
TumorMajorCount = isNormalFirst ? m2Major : m1Major,
TumorMinorCount = isNormalFirst ? m2Minor : m1Minor,
Score = double.Parse(info.StringAfter(scorePrefix)),
RefGeneFunc = GetDictionaryValue(ann.Annotations, "Func.refGene", string.Empty),
RefGeneName = GetDictionaryValue(ann.Annotations, "Gene.refGene", string.Empty),
RefGeneExonicFunc = GetDictionaryValue(ann.Annotations, "ExonicFunc.refGene", string.Empty)
};
result.Add(item);
}
return result;
}
public List <IIdentifiedSpectrum> ExtractPeptides(Dictionary <string, MaxQuantModificationItem> shortModMap, string peptideFile)
{
var mods = MaxQuantModificationList.ReadFromFile(_option.MaxQuantModificationXml);
var sites = new AnnotationFormat().ReadFromFile(_option.MaxQuantSiteFile);
var msmsIds = new HashSet <string>(from s in sites
let msmsids = s.Annotations["MS/MS IDs"] as string
let msmsIdList = msmsids.Split(';')
from msmsId in msmsIdList
select msmsId);
var format = new AnnotationFormat();
var msmsList = format.ReadFromFile(_option.MaxQuantMSMSFile);
msmsList.RemoveAll(l => !msmsIds.Contains(l.Annotations["id"].ToString()));
using (var sw = new StreamWriter(peptideFile))
{
sw.WriteLine("FileScan\tSequence\tMH+\tDiff(MH+)\tCharge\tScore\tReference\tModification\tRetentionTime");
foreach (var msms in msmsList)
{
string modification;
string modifiedSequence;
ParseModification(msms, shortModMap, mods, out modification, out modifiedSequence);
var mh = double.Parse(msms.Annotations["Mass"].ToString()) + Atom.H.MonoMass;
var diffStr = msms.Annotations["Mass Error [ppm]"].ToString();
var diffmh = diffStr.Equals("NaN") ? 0 : PrecursorUtils.ppm2mz(mh, double.Parse(diffStr));
sw.WriteLine("{0},{1}\t{2}\t{3}\t{4}\t{5}\t{6}\t{7}\t{8}\t{9}",
msms.Annotations["Raw file"],
msms.Annotations["Scan number"],
modifiedSequence,
mh,
diffmh,
msms.Annotations["Charge"],
msms.Annotations["Score"],
msms.Annotations["Proteins"].ToString().Replace(";", "/"),
modification,
msms.Annotations["Retention time"]);
}
}
return(new MascotPeptideTextFormat().ReadFromFile(peptideFile));
}
private static StringBuilder Format <T>(StringBuilder sb, T nag, AnnotationFormat annotationFormat)
where T : struct, Enum
{
var value = (int)EnumsNET.Enums.GetUnderlyingValue(nag);
if (value != 0)
{
var strNag = annotationFormat == AnnotationFormat.Symbolic
? nag.AsString(_symbolFormat)
: "$" + EnumsNET.Enums.GetUnderlyingValue(nag);
if (!string.IsNullOrWhiteSpace(strNag))
{
sb.Append(strNag).Append(" ");
}
}
return(sb);
}
public void TestReadingAnnotationsWithDifferentSelections([ValueSource(nameof(_selections))] int[] selection)
{
using (var reader = new StreamReader(new MemoryStream(Encoding.UTF8.GetBytes(Examples.MouseUniprotAnnotations))))
{
var annotations = AnnotationFormat.ReadMappings(reader, 0, selection).ToArray();
var expected = new[]
{
"p34968", "b1atn5", "q5wru6", "q91v24", "q9jl36", "q9dbm0", "q6pe15", "q3tmb4", "q8c3s8", "q8c724", "q8k188"
};
Assert.AreEqual(expected.Length, annotations.Length);
CollectionAssert.AreEquivalent(expected, annotations.Select(annot => annot.fromId));
foreach (var(fromId, toIds) in annotations)
{
Assert.AreEqual(fromId, fromId.ToLower());
Assert.AreEqual(toIds.Length, selection.Length);
}
}
}
public override IEnumerable <string> Process()
{
var format = new AnnotationFormat();
var items = format.ReadFromFile(_options.ClinicalFile);
format.Format.Headers = sampleBarcodeKey + "\t" + format.Format.Headers;
var itemMap = items.ToDictionary(m => m.BarCode());
using (StreamReader sr = new StreamReader(_options.DataFile))
{
var barcodes = sr.ReadLine().Split('\t').Where(m => m.StartsWith("TCGA")).ToList();
List <Annotation> found = new List <Annotation>();
foreach (var barcode in barcodes)
{
var patient = barcode.Substring(0, 12);
Annotation ann;
if (!itemMap.TryGetValue(patient, out ann))
{
if (_options.ThrowException)
{
throw new Exception("Cannot find patient information for " + patient);
}
Console.Error.WriteLine("Cannot find patient information for " + patient);
ann = new Annotation();
}
var curann = new Annotation();
curann.Annotations[sampleBarcodeKey] = barcode;
curann.Annotations[TCGAClinicalInformation.BcrPatientBarcode] = patient;
foreach (var e in ann.Annotations)
{
curann.Annotations[e.Key] = e.Value;
}
found.Add(curann);
}
format.WriteToFile(_options.OutputFile, found);
}
return(new[] { _options.OutputFile });
}
public override IEnumerable<string> Process()
{
var format = new AnnotationFormat();
var items = format.ReadFromFile(_options.ClinicalFile);
format.Format.Headers = sampleBarcodeKey + "\t" + format.Format.Headers;
var itemMap = items.ToDictionary(m => m.BarCode());
using (StreamReader sr = new StreamReader(_options.DataFile))
{
var barcodes = sr.ReadLine().Split('\t').Where(m => m.StartsWith("TCGA")).ToList();
List<Annotation> found = new List<Annotation>();
foreach (var barcode in barcodes)
{
var patient = barcode.Substring(0, 12);
Annotation ann;
if (!itemMap.TryGetValue(patient, out ann))
{
if (_options.ThrowException)
{
throw new Exception("Cannot find patient information for " + patient);
}
Console.Error.WriteLine("Cannot find patient information for " + patient);
ann = new Annotation();
}
var curann = new Annotation();
curann.Annotations[sampleBarcodeKey] = barcode;
curann.Annotations[TCGAClinicalInformation.BcrPatientBarcode] = patient;
foreach (var e in ann.Annotations)
{
curann.Annotations[e.Key] = e.Value;
}
found.Add(curann);
}
format.WriteToFile(_options.OutputFile, found);
}
return new[] { _options.OutputFile };
}
public static List <SomaticItem> ParseGlmvcFile(string fileName, Func <string, bool> acceptChromosome)
{
var result = new List <SomaticItem>();
var bar = Path.GetFileName(fileName).StringBefore(".");
var annos = new AnnotationFormat().ReadFromFile(fileName);
foreach (var ann in annos)
{
var chr = ann.Annotations["chr"].ToString();
if (!acceptChromosome(chr))
{
continue;
}
var fdr = ann.Annotations["brglm_group_fdr"].ToString();
var item = new SomaticItem()
{
Sample = bar,
Chrom = chr,
StartPosition = int.Parse(ann.Annotations["start"].ToString()),
RefAllele = ann.Annotations["major_allele"].ToString(),
AltAllele = ann.Annotations["minor_allele"].ToString(),
NormalMajorCount = int.Parse(ann.Annotations["normal_major_count"].ToString()),
NormalMinorCount = int.Parse(ann.Annotations["normal_minor_count"].ToString()),
TumorMajorCount = int.Parse(ann.Annotations["tumor_major_count"].ToString()),
TumorMinorCount = int.Parse(ann.Annotations["tumor_minor_count"].ToString()),
LogisticScore = ann.Annotations["brglm_score"].ToString(),
LogisticStrand = ann.Annotations["brglm_strand"].ToString(),
LogisticPosition = ann.Annotations["brglm_position"].ToString(),
LogisticGroupFdr = fdr,
Score = -Math.Log(double.Parse(fdr)),
RefGeneFunc = GetDictionaryValue(ann.Annotations, "annovar_Func.refGene", string.Empty),
RefGeneName = GetDictionaryValue(ann.Annotations, "annovar_Gene.refGene", string.Empty),
RefGeneExonicFunc = GetDictionaryValue(ann.Annotations, "annovar_ExonicFunc.refGene", string.Empty),
RefGeneAAChange = GetDictionaryValue(ann.Annotations, "annovar_AAChange.refGene", string.Empty),
};
result.Add(item);
}
return(result);
}
public Dictionary <string, Dictionary <string, List <string> > > ReadDescriptionFromDirectory(string dir)
{
var sdrfFile = GetSdrfFiles(dir);
if (sdrfFile.Length == 0)
{
throw new ArgumentException("Cannot find sdrf file in directory " + dir);
}
var ann = new AnnotationFormat("^#").ReadFromFile(sdrfFile[0]);
Dictionary <string, Dictionary <string, List <string> > > result = new Dictionary <string, Dictionary <string, List <string> > >();
foreach (var a in ann)
{
string filename;
if (a.Annotations.ContainsKey("Array Data File"))
{
filename = a.Annotations["Array Data File"] as string;
}
else if (a.Annotations.ContainsKey("GEO asscession number"))
{
filename = a.Annotations["GEO asscession number"] as string;
}
else
{
throw new Exception("I don't know how to get filename from " + sdrfFile[0]);
}
filename = Path.GetFileNameWithoutExtension(filename);
var dic = new Dictionary <string, List <string> >();
foreach (var kv in a.Annotations)
{
dic[kv.Key] = new string[] { kv.Value as string }.ToList();
}
result[filename] = dic;
}
return(result);
}
public Dictionary<string, Dictionary<string, List<string>>> ReadDescriptionFromDirectory(string dir)
{
var sdrfFile = GetSdrfFiles(dir);
if (sdrfFile.Length == 0)
{
throw new ArgumentException("Cannot find sdrf file in directory " + dir);
}
var ann = new AnnotationFormat("^#").ReadFromFile(sdrfFile[0]);
Dictionary<string, Dictionary<string, List<string>>> result = new Dictionary<string, Dictionary<string, List<string>>>();
foreach (var a in ann)
{
string filename;
if (a.Annotations.ContainsKey("Array Data File"))
{
filename = a.Annotations["Array Data File"] as string;
}
else if (a.Annotations.ContainsKey("GEO asscession number"))
{
filename = a.Annotations["GEO asscession number"] as string;
}
else
{
throw new Exception("I don't know how to get filename from " + sdrfFile[0]);
}
filename = Path.GetFileNameWithoutExtension(filename);
var dic = new Dictionary<string, List<string>>();
foreach (var kv in a.Annotations)
{
dic[kv.Key] = new string[] { kv.Value as string }.ToList();
}
result[filename] = dic;
}
return result;
}
public List <BreastCancerSampleItem> ParseDataset(string datasetDirectory)
{
var files = new HashSet <string>(from f in CelFile.GetCelFiles(datasetDirectory, false)
select Path.GetFileNameWithoutExtension(f));
var sdrfFile = Directory.GetFiles(datasetDirectory, "*.sdrf.txt");
if (sdrfFile.Length == 0)
{
throw new ArgumentException("Cannot find sdrf file in directory " + datasetDirectory);
}
var ann = new AnnotationFormat("^#").ReadFromFile(sdrfFile[0]);
return((from a in ann
let filename = Path.GetFileNameWithoutExtension(FindValue(a, ColumnName.Sample))
where files.Contains(filename)
select new BreastCancerSampleItem()
{
Dataset = Path.GetFileName(datasetDirectory),
Sample = filename,
Age = FindValue(a, ColumnName.Age),
ER = new StatusValue(FindValue(a, ColumnName.ER)).Value,
PR = new StatusValue(FindValue(a, ColumnName.PR)).Value,
HER2 = new StatusValue(FindValue(a, ColumnName.HER2)).Value,
Stage = FindValue(a, ColumnName.Stage),
TumorStatus = FindValue(a, ColumnName.TumorStage),
Grade = FindValue(a, ColumnName.Grade),
NodalStatus = FindValue(a, ColumnName.NodalStatus),
PCR = FindValue(a, ColumnName.PCR),
DFS = FindValue(a, ColumnName.DFS),
DFSTime = FindValue(a, ColumnName.DFSTime),
RFS = FindValue(a, ColumnName.RFS),
RFSTime = FindValue(a, ColumnName.RFSTime),
DMFS = FindValue(a, ColumnName.DMFS),
DMFSTime = FindValue(a, ColumnName.DMFSTime),
OverallSurvival = FindValue(a, ColumnName.OverallServive),
DeadOfDisease = FindValue(a, ColumnName.DeadOfDisease)
}).ToList());
}
public List<BreastCancerSampleItem> ParseDataset(string datasetDirectory)
{
var files = new HashSet<string>(from f in CelFile.GetCelFiles(datasetDirectory, false)
select Path.GetFileNameWithoutExtension(f));
var sdrfFile = Directory.GetFiles(datasetDirectory, "*.sdrf.txt");
if (sdrfFile.Length == 0)
{
throw new ArgumentException("Cannot find sdrf file in directory " + datasetDirectory);
}
var ann = new AnnotationFormat("^#").ReadFromFile(sdrfFile[0]);
return (from a in ann
let filename = Path.GetFileNameWithoutExtension(FindValue(a, ColumnName.Sample))
where files.Contains(filename)
select new BreastCancerSampleItem()
{
Dataset = Path.GetFileName(datasetDirectory),
Sample = filename,
Age = FindValue(a, ColumnName.Age),
ER = new StatusValue(FindValue(a, ColumnName.ER)).Value,
PR = new StatusValue(FindValue(a, ColumnName.PR)).Value,
HER2 = new StatusValue(FindValue(a, ColumnName.HER2)).Value,
Stage = FindValue(a, ColumnName.Stage),
TumorStatus = FindValue(a, ColumnName.TumorStage),
Grade = FindValue(a, ColumnName.Grade),
NodalStatus = FindValue(a, ColumnName.NodalStatus),
PCR = FindValue(a, ColumnName.PCR),
DFS = FindValue(a, ColumnName.DFS),
DFSTime = FindValue(a, ColumnName.DFSTime),
RFS = FindValue(a, ColumnName.RFS),
RFSTime = FindValue(a, ColumnName.RFSTime),
DMFS = FindValue(a, ColumnName.DMFS),
DMFSTime = FindValue(a, ColumnName.DMFSTime),
OverallSurvival = FindValue(a, ColumnName.OverallServive),
DeadOfDisease = FindValue(a, ColumnName.DeadOfDisease)
}).ToList();
}
public List <IIdentifiedSpectrum> ReadFromFile(string fileName)
{
var result = new List <IIdentifiedSpectrum>();
var anns = new AnnotationFormat().ReadFromFile(fileName);
foreach (var ann in anns)
{
var peptideId = ann.Annotations["PeptideId"] as string;
var sequence = peptideId.StringBefore("_");
var spec = new IdentifiedSpectrum();
var pep = new IdentifiedPeptide(spec);
pep.Sequence = sequence;
spec.Query.FileScan.Experimental = ann.Annotations["Sample"] as string;
spec.Query.FileScan.RetentionTime = double.Parse(ann.Annotations["PredictionRetentionTime"] as string);
spec.Query.FileScan.Charge = int.Parse(ann.Annotations["Charge"] as string);
spec.IsPrecursorMonoisotopic = true;
spec.TheoreticalMH = PrecursorUtils.MzToMH(double.Parse(ann.Annotations["TheoreticalMz"] as string), spec.Query.FileScan.Charge, true);
result.Add(spec);
}
return(result);
}
public void ParseDataset(string datasetDirectory, Dictionary <string, BreastCancerSampleItem> sampleMap)
{
var files = new HashSet <string>(from f in CelFile.GetCelFiles(datasetDirectory, false)
select Path.GetFileNameWithoutExtension(f));
var sdrfFile = Directory.GetFiles(datasetDirectory, "*.sdrf.txt");
if (sdrfFile.Length == 0)
{
throw new ArgumentException("Cannot find sdrf file in directory " + datasetDirectory);
}
var ann = new AnnotationFormat("^#").ReadFromFile(sdrfFile[0]);
var dataset = Path.GetFileName(datasetDirectory);
foreach (var a in ann)
{
var filename = Path.GetFileNameWithoutExtension(FindValue(a, ColumnName.Sample));
if (files.Contains(filename))
{
if (!sampleMap.ContainsKey(filename))
{
sampleMap[filename] = new BreastCancerSampleItem();
sampleMap[filename].Dataset = dataset;
sampleMap[filename].Sample = filename;
}
var item = sampleMap[filename];
string value;
if (FindValue(a, ColumnName.Age, out value))
{
item.Age = value;
}
if (FindValue(a, ColumnName.ER, out value))
{
item.ER = StatusValue.TransferStatus(value);
}
if (FindValue(a, ColumnName.PR, out value))
{
item.PR = StatusValue.TransferStatus(value);
}
if (FindValue(a, ColumnName.HER2, out value))
{
item.HER2 = StatusValue.TransferStatus(value);
}
if (FindValue(a, ColumnName.Stage, out value))
{
item.Stage = value;
}
if (FindValue(a, ColumnName.TumorStage, out value))
{
item.TumorStatus = value;
}
if (FindValue(a, ColumnName.Grade, out value))
{
item.Grade = value;
}
if (FindValue(a, ColumnName.NodalStatus, out value))
{
item.NodalStatus = value;
}
if (FindValue(a, ColumnName.PCR, out value))
{
item.PCR = value;
}
if (FindValue(a, ColumnName.DFS, out value))
{
item.DFS = value;
}
if (FindValue(a, ColumnName.DFSTime, out value))
{
item.DFSTime = value;
}
if (FindValue(a, ColumnName.RFS, out value))
{
item.RFS = value;
}
if (FindValue(a, ColumnName.RFSTime, out value))
{
item.RFSTime = value;
}
if (FindValue(a, ColumnName.DMFS, out value))
{
item.DMFS = value;
}
if (FindValue(a, ColumnName.DMFSTime, out value))
{
item.DMFSTime = value;
}
if (FindValue(a, ColumnName.OverallServive, out value))
{
item.OverallSurvival = value;
}
if (FindValue(a, ColumnName.DeadOfDisease, out value))
{
item.DeadOfDisease = value;
}
}
}
}
public override IEnumerable <string> Process()
{
//Extract chromotagraph information
var chroOptions = new ChromatographProfileBuilderOptions();
options.CopyProperties(chroOptions);
chroOptions.InputFile = options.InputFile;
chroOptions.OutputFile = options.BoundaryOutputFile;
chroOptions.DrawImage = false;
var builder = new ChromatographProfileBuilder(chroOptions);
if (!File.Exists(options.BoundaryOutputFile) || options.Overwrite)
{
Progress.SetMessage("Finding envelope ...");
builder.Progress = this.Progress;
builder.Process();
}
//Calculate deuterium enrichment for peptide
if (!File.Exists(options.DeuteriumOutputFile) || options.Overwrite)
{
Progress.SetMessage("Calculating deuterium ...");
var deuteriumOptions = new RTemplateProcessorOptions()
{
InputFile = options.BoundaryOutputFile,
OutputFile = options.DeuteriumOutputFile,
RTemplate = DeuteriumR,
RExecute = SystemUtils.GetRExecuteLocation(),
CreateNoWindow = true
};
deuteriumOptions.Parameters.Add("outputImage<-" + (options.DrawImage ? "1" : "0"));
deuteriumOptions.Parameters.Add("excludeIsotopic0<-" + (options.ExcludeIsotopic0 ? "1" : "0"));
new RTemplateProcessor(deuteriumOptions)
{
Progress = this.Progress
}.Process();
}
var deuteriumMap = new AnnotationFormat().ReadFromFile(options.DeuteriumOutputFile).ToDictionary(m => m.Annotations["ChroFile"].ToString());
//Read old spectra information
var format = new MascotPeptideTextFormat();
var spectra = format.ReadFromFile(options.InputFile);
foreach (var spec in spectra)
{
spec.Annotations.Remove("RetentionTime");
spec.Annotations.Remove("TheoreticalDeuterium");
spec.Annotations.Remove("ObservedDeuterium");
spec.Annotations.Remove("NumDeuteriumIncorporated");
spec.Annotations.Remove("NumExchangableHydrogen");
spec.Annotations.Remove("DeuteriumEnrichmentPercent");
}
var calcSpectra = new List <IIdentifiedSpectrum>();
var aas = new Aminoacids();
foreach (var pep in spectra)
{
var filename = Path.GetFileNameWithoutExtension(builder.GetTargetFile(pep));
if (deuteriumMap.ContainsKey(filename))
{
var numExchangeableHydrogens = aas.ExchangableHAtom(pep.Peptide.PureSequence);
var numDeuteriumIncorporated = double.Parse(deuteriumMap[filename].Annotations["NumDeuteriumIncorporated"] as string);
pep.Annotations["PeakRetentionTime"] = deuteriumMap[filename].Annotations["RetentionTime"];
pep.Annotations["TheoreticalDeuterium"] = deuteriumMap[filename].Annotations["TheoreticalDeuterium"];
pep.Annotations["ObservedDeuterium"] = deuteriumMap[filename].Annotations["ObservedDeuterium"];
pep.Annotations["NumDeuteriumIncorporated"] = deuteriumMap[filename].Annotations["NumDeuteriumIncorporated"];
pep.Annotations["NumExchangableHydrogen"] = numExchangeableHydrogens;
pep.Annotations["DeuteriumEnrichmentPercent"] = numDeuteriumIncorporated / numExchangeableHydrogens;
calcSpectra.Add(pep);
}
}
format.PeptideFormat.Headers = format.PeptideFormat.Headers + "\tPeakRetentionTime\tTheoreticalDeuterium\tObservedDeuterium\tNumDeuteriumIncorporated\tNumExchangableHydrogen\tDeuteriumEnrichmentPercent";
format.NotExportSummary = true;
format.WriteToFile(GetPeptideDeteriumFile(), calcSpectra);
var specGroup = calcSpectra.GroupBy(m => m.Peptide.PureSequence).OrderBy(l => l.Key).ToList();
var times = options.ExperimentalTimeMap.Values.Distinct().OrderBy(m => m).ToArray();
using (var sw = new StreamWriter(options.OutputFile))
{
sw.WriteLine("Peptide\t{0}", (from t in times select t.ToString()).Merge("\t"));
foreach (var peptide in specGroup)
{
var curSpectra = peptide.GroupBy(m => options.ExperimentalTimeMap[m.Query.FileScan.Experimental]).ToDictionary(l => l.Key, l => l.ToArray());
if (options.PeptideInAllTimePointOnly && times.Any(l => !curSpectra.ContainsKey(l)))
{
continue;
}
sw.Write(peptide.Key);
foreach (var time in times)
{
if (curSpectra.ContainsKey(time))
{
var deps = (from spec in curSpectra[time] select double.Parse(spec.Annotations["DeuteriumEnrichmentPercent"].ToString())).ToArray();
var depMedian = Statistics.Median(deps);
sw.Write("\t{0:0.######}", depMedian);
}
else
{
sw.Write("\tNA");
}
}
sw.WriteLine();
}
}
Progress.SetMessage("Peptide deuterium enrichment calculation finished ...");
return(new string[] { options.OutputFile });
}
public override IEnumerable <string> Process()
{
var annos = new AnnotationFormat().ReadFromFile(boundaryFile);
var aas = new Aminoacids();
var pbuilder = new EmassProfileBuilder();
var chroFormat = new ChromatographProfileTextWriter();
foreach (var ann in annos)
{
var chroFile = Path.Combine(ann.Annotations["ChroDirectory"] as string, ann.Annotations["ChroFile"] as string) + ".tsv";
var peptide = (ann.Annotations["PeptideId"] as string).StringBefore("_");
var start = double.Parse(ann.Annotations["ChroLeft"] as string);
var end = double.Parse(ann.Annotations["ChroRight"] as string);
var chro = chroFormat.ReadFromFile(chroFile);
var chroFiltered = chro.Profiles.Where(l => l.RetentionTime >= start && l.RetentionTime <= end).ToArray();
//all observed isotopimers, start from 1, transfer to zero-based
var allIsotopics = (from f in chroFiltered
from s in f
select s.Isotopic - 1).Distinct().OrderBy(l => l).ToArray();
//get minimum intensity of each isotopimer
var minIntensities = (from iso in allIsotopics
select(from f in chroFiltered
where f.Count > iso
select f[iso].Intensity).Min()).ToArray();
//how many isotopic should I trust?
var iso0 = (from f in chroFiltered select f.First().Intensity).ToArray();
int maxIso = 1;
for (int i = 2; i < allIsotopics.Length; i++)
{
var isoi = (from f in chroFiltered select f.Count > i ? f[i].Intensity : minIntensities[i]).ToArray();
var corr = MathNet.Numerics.Statistics.Correlation.Pearson(iso0, isoi);
if (corr < minPearsonCorrelation)
{
maxIso = i;
break;
}
}
var maxIsoArray = allIsotopics.Where(l => l <= maxIso).ToArray();
//get observed profile in each scan
var observedIons = (from f in chroFiltered
select(from iso in maxIsoArray select f.Count > iso ? f[iso].Intensity : minIntensities[iso]).ToArray()).ToArray();
var atomComposition = aas.GetPeptideAtomComposition(peptide);
var hatom = (int)(Math.Round(aas.ExchangableHAtom(peptide)));
for (int h2 = 0; h2 < hatom; h2++)
{
}
var profiles = pbuilder.GetProfile(atomComposition, 1, 3);
var isotopicIons = (from peak in profiles
select new IsotopicIon()
{
Mz = peak.Mz,
Intensity = peak.Intensity,
}).ToArray();
}
return(null);
}
public static List<SomaticItem> ParseGlmvcFile(string fileName, Func<string, bool> acceptChromosome)
{
var result = new List<SomaticItem>();
var bar = Path.GetFileName(fileName).StringBefore(".");
var annos = new AnnotationFormat().ReadFromFile(fileName);
foreach (var ann in annos)
{
var chr = ann.Annotations["chr"].ToString();
if (!acceptChromosome(chr))
{
continue;
}
var fdr = ann.Annotations["brglm_group_fdr"].ToString();
var item = new SomaticItem()
{
Sample = bar,
Chrom = chr,
StartPosition = int.Parse(ann.Annotations["start"].ToString()),
RefAllele = ann.Annotations["major_allele"].ToString(),
AltAllele = ann.Annotations["minor_allele"].ToString(),
NormalMajorCount = int.Parse(ann.Annotations["normal_major_count"].ToString()),
NormalMinorCount = int.Parse(ann.Annotations["normal_minor_count"].ToString()),
TumorMajorCount = int.Parse(ann.Annotations["tumor_major_count"].ToString()),
TumorMinorCount = int.Parse(ann.Annotations["tumor_minor_count"].ToString()),
LogisticScore = ann.Annotations["brglm_score"].ToString(),
LogisticStrand = ann.Annotations["brglm_strand"].ToString(),
LogisticPosition = ann.Annotations["brglm_position"].ToString(),
LogisticGroupFdr = fdr,
Score = -Math.Log(double.Parse(fdr)),
RefGeneFunc = GetDictionaryValue(ann.Annotations, "annovar_Func.refGene", string.Empty),
RefGeneName = GetDictionaryValue(ann.Annotations, "annovar_Gene.refGene", string.Empty),
RefGeneExonicFunc = GetDictionaryValue(ann.Annotations, "annovar_ExonicFunc.refGene", string.Empty),
RefGeneAAChange = GetDictionaryValue(ann.Annotations, "annovar_AAChange.refGene", string.Empty),
};
result.Add(item);
}
return result;
}
public void ParseDataset(string datasetDirectory, Dictionary<string, BreastCancerSampleItem> sampleMap)
{
var files = new HashSet<string>(from f in CelFile.GetCelFiles(datasetDirectory, false)
select Path.GetFileNameWithoutExtension(f));
var sdrfFile = Directory.GetFiles(datasetDirectory, "*.sdrf.txt");
if (sdrfFile.Length == 0)
{
throw new ArgumentException("Cannot find sdrf file in directory " + datasetDirectory);
}
var ann = new AnnotationFormat("^#").ReadFromFile(sdrfFile[0]);
var dataset = Path.GetFileName(datasetDirectory);
foreach (var a in ann)
{
var filename = Path.GetFileNameWithoutExtension(FindValue(a, ColumnName.Sample));
if (files.Contains(filename))
{
if (!sampleMap.ContainsKey(filename))
{
sampleMap[filename] = new BreastCancerSampleItem();
sampleMap[filename].Dataset = dataset;
sampleMap[filename].Sample = filename;
}
var item = sampleMap[filename];
string value;
if (FindValue(a, ColumnName.Age, out value))
{
item.Age = value;
}
if (FindValue(a, ColumnName.ER, out value))
{
item.ER = StatusValue.TransferStatus(value);
}
if (FindValue(a, ColumnName.PR, out value))
{
item.PR = StatusValue.TransferStatus(value);
}
if (FindValue(a, ColumnName.HER2, out value))
{
item.HER2 = StatusValue.TransferStatus(value);
}
if (FindValue(a, ColumnName.Stage, out value))
{
item.Stage = value;
}
if (FindValue(a, ColumnName.TumorStage, out value))
{
item.TumorStatus = value;
}
if (FindValue(a, ColumnName.Grade, out value))
{
item.Grade = value;
}
if (FindValue(a, ColumnName.NodalStatus, out value))
{
item.NodalStatus = value;
}
if (FindValue(a, ColumnName.PCR, out value))
{
item.PCR = value;
}
if (FindValue(a, ColumnName.DFS, out value))
{
item.DFS = value;
}
if (FindValue(a, ColumnName.DFSTime, out value))
{
item.DFSTime = value;
}
if (FindValue(a, ColumnName.RFS, out value))
{
item.RFS = value;
}
if (FindValue(a, ColumnName.RFSTime, out value))
{
item.RFSTime = value;
}
if (FindValue(a, ColumnName.DMFS, out value))
{
item.DMFS = value;
}
if (FindValue(a, ColumnName.DMFSTime, out value))
{
item.DMFSTime = value;
}
if (FindValue(a, ColumnName.OverallServive, out value))
{
item.OverallSurvival = value;
}
if (FindValue(a, ColumnName.DeadOfDisease, out value))
{
item.DeadOfDisease = value;
}
}
}
}
