protected virtual void ReadCountFile(string countFile)
{
Dictionary<string, string> counts = new MapReader(0, 1).ReadFromFile(countFile);
foreach (var c in counts)
{
Counts[c.Key] = int.Parse(c.Value);
}
}
public void TestRead()
{
var map = new MapReader("Derived Array Data File", "Comment [Aliquot Barcode]").ReadFromFile("../../../data/broad.mit.edu_BRCA.Genome_Wide_SNP_6.sdrf.txt");
Assert.AreEqual(6789, map.Count);
Assert.AreEqual("TCGA-A8-A06R-01A-11D-A011-01", map["CUSKS_p_TCGAb47_SNP_1N_GenomeWideSNP_6_A02_628278.hg18.seg.txt"]);
Assert.AreEqual("TCGA-A8-A06R-01A-11D-A011-01", map["CUSKS_p_TCGAb47_SNP_1N_GenomeWideSNP_6_A02_628278.hg19.seg.txt"]);
Assert.AreEqual("TCGA-A8-A06R-01A-11D-A011-01", map["CUSKS_p_TCGAb47_SNP_1N_GenomeWideSNP_6_A02_628278.nocnv_hg18.seg.txt"]);
Assert.AreEqual("TCGA-A8-A06R-01A-11D-A011-01", map["CUSKS_p_TCGAb47_SNP_1N_GenomeWideSNP_6_A02_628278.nocnv_hg19.seg.txt"]);
}
public FindParticipantBySdrfFile(string sdrfFile, string fileColumn, string barCodeColumn)
{
this.sdrfFile = sdrfFile;
var anns = new MapReader(fileColumn, barCodeColumn).ReadFromFile(sdrfFile);
maps = new Dictionary<string, string>();
foreach (var ann in anns)
{
if (ann.Key.Equals("->"))
{
continue;
}
maps[ann.Key.ToLower()] = TCGAUtils.GetSampleBarCode(ann.Value);
}
}
public void ParseDataset(string datasetDirectory, Dictionary<string, BreastCancerSampleItem> sampleMap)
{
//var files = GeoUtils.GetGsmNameFileMap(datasetDirectory);
var dirname = Path.GetFileName(datasetDirectory);
var map = new MapReader("geo_accn", "er").ReadFromFile(datasetDirectory + @"\GSE2990_suppl_info.txt");
foreach (var m in map)
{
var f = datasetDirectory + "\\" + m.Key + ".cel";
if (File.Exists(f))
{
if (!sampleMap.ContainsKey(m.Key))
{
sampleMap[m.Key] = new BreastCancerSampleItem(dirname, m.Key);
}
var item = sampleMap[m.Key];
item.ER = m.Value;
}
}
}
/// <summary>
/// After chip type check, remove file from specified chip types
/// </summary>
public void FilterChipType(string chipTypeTableFileName, string[] removeChipTypes)
{
chipTypeTableFileName = CheckFileName(chipTypeTableFileName);
var dic = new MapReader(0, 1).ReadFromFile(chipTypeTableFileName);
foreach (var cel in dic)
{
var chiptype = cel.Value;
if (removeChipTypes.Contains(chiptype))
{
var celfile = cel.Key;
if (File.Exists(celfile))
{
Console.WriteLine("Removing {0} : {1}", chiptype, celfile);
ExcludeFile(celfile);
}
}
}
}
public override IEnumerable<string> Process()
{
Dictionary<string, List<GtfItem>> map = new Dictionary<string, List<GtfItem>>();
var namemap = new Dictionary<string, string>();
if (File.Exists(options.MapFile))
{
namemap = new MapReader(0, 1, hasHeader: false).ReadFromFile(options.MapFile);
}
using (var gtf = new GtfItemFile(options.InputFile))
{
GtfItem item;
int count = 0;
while ((item = gtf.Next()) != null)
{
count++;
if ((count % 100000) == 0)
{
Progress.SetMessage("{0} gtf item processed", count);
}
List<GtfItem> oldItems;
if (!map.TryGetValue(item.GeneId, out oldItems))
{
map[item.GeneId] = new[] { item }.ToList();
}
else
{
if (IsExon(item))
{
oldItems.RemoveAll(m => !IsExon(m));
oldItems.Add(item);
}
else
{
if (oldItems.All(m => !IsExon(m)))
{
oldItems.Add(item);
}
}
}
}
}
// map[item.GeneId] = item.Attributes.StringAfter("gene_name \"").StringBefore("\"");
var keys = (from key in map.Keys
orderby key
select key).ToList();
using (StreamWriter sw = new StreamWriter(options.OutputFile))
{
bool bHasGeneName = map.Values.Any(l => l.Any(m => m.Attributes.Contains("gene_name")));
if (!bHasGeneName && !File.Exists(options.MapFile))
{
throw new Exception(string.Format("No gene_name found in {0} and no id/name map file defined.", options.InputFile));
}
bool bHasGeneBiotype = map.Values.Any(l => l.Any(m => m.Attributes.Contains("gene_biotype")));
if (bHasGeneBiotype)
{
sw.WriteLine("gene_id\tgene_name\tlength\tgene_biotype");
}
else
{
sw.WriteLine("gene_id\tgene_name\tlength");
}
foreach (var key in keys)
{
var gtfs = map[key];
string name;
var gtf = gtfs.FirstOrDefault(m => m.Attributes.Contains("gene_name"));
gtfs.CombineCoordinates();
string biotype;
if (gtf == null)
{
biotype = string.Empty;
if (!namemap.TryGetValue(key, out name))
{
name = key;
}
}
else
{
biotype = gtf.GetBiotype();
name = gtf.Attributes.StringAfter("gene_name \"").StringBefore("\"");
}
if (bHasGeneBiotype)
{
sw.WriteLine("{0}\t{1}\t{2}\t{3}", key, name, gtfs.Sum(m => m.Length), biotype);
}
else
{
sw.WriteLine("{0}\t{1}\t{2}", key, name, gtfs.Sum(m => m.Length));
}
}
}
return new string[] { options.OutputFile };
}
public static List<CoverageRegion> GetTargetCoverageRegion(ITargetBuilderOptions options, IProgressCallback progress, bool removeRegionWithoutSequence = true)
{
List<CoverageRegion> result;
if (options.TargetFile.EndsWith(".xml"))
{
result = GetTargetCoverageRegionFromXml(options, progress);
}
else
{
result = GetTargetCoverageRegionFromBed(options, progress);
}
var dic = result.ToGroupDictionary(m => m.Seqname);
progress.SetMessage("Filling sequence from {0}...", options.GenomeFastaFile);
using (var sr = new StreamReader(options.GenomeFastaFile))
{
var ff = new FastaFormat();
Sequence seq;
while ((seq = ff.ReadSequence(sr)) != null)
{
progress.SetMessage("Processing chromosome {0} ...", seq.Reference);
var seqname = seq.Name.StringAfter("chr");
List<CoverageRegion> lst;
if (dic.TryGetValue(seqname, out lst))
{
foreach (var l in lst)
{
l.Sequence = seq.SeqString.Substring((int)(l.Start - 1), (int)l.Length);
if(l.Strand == '+')
{
l.ReverseComplementedSequence = SequenceUtils.GetReverseComplementedSequence(l.Sequence);
}
}
}
}
}
if (removeRegionWithoutSequence)
{
result.RemoveAll(l => string.IsNullOrEmpty(l.Sequence));
}
progress.SetMessage("Filling sequence finished.");
var namemap = new MapReader(1, 12).ReadFromFile(options.RefgeneFile);
result.ForEach(m =>
{
var gene = m.Name.StringBefore("_utr3");
m.GeneSymbol = namemap.ContainsKey(gene) ? namemap[gene] : string.Empty;
});
return result;
}
public override IEnumerable<string> Process()
{
this.Progress.SetMessage("Reading group sample map file ...");
var groupSampleMap = new MapReader(0, 1).ReadFromFile(options.MapFile);
Dictionary<string, SignificantItem> geneNameMap = new Dictionary<string, SignificantItem>();
this.Progress.SetMessage("Reading cuffdiff significant files ...");
var sigs = (from file in options.SignificantFiles
from line in File.ReadAllLines(file).Skip(1)
let parts = line.Split('\t')
where parts.Length >= 3
let gene = parts[1]
let name = parts[2]
where !name.Equals("-")
let location = parts[3]
select new SignificantItem() { Gene = gene, GeneName = name, GeneLocation = location }).ToList();
foreach (var gene in sigs)
{
if (!geneNameMap.ContainsKey(gene.Gene))
{
geneNameMap[gene.Gene] = gene;
}
}
Func<string, bool> acceptGene = m => geneNameMap.ContainsKey(m);
var countFile = options.OutputFilePrefix + ".count";
var fpkmFile = options.OutputFilePrefix + ".fpkm";
var items = new List<TrackingItem>();
foreach (var trackingFile in options.InputFiles)
{
this.Progress.SetMessage("Reading cuffdiff read_group_tracking file " + trackingFile + "...");
using (StreamReader sr = new StreamReader(trackingFile))
{
string line = sr.ReadLine();
while ((line = sr.ReadLine()) != null)
{
var parts = line.Split('\t');
if (parts.Length <= 7)
{
continue;
}
var gene = parts[0];
if (!acceptGene(gene))
{
continue;
}
var group_index = parts[1] + "_" + parts[2];
var sample = groupSampleMap.ContainsKey(group_index) ? groupSampleMap[group_index] : group_index;
var count = parts[3];
var fpkm = parts[6];
var item = new TrackingItem() { Gene = gene, Sample = sample, Count = count, FPKM = fpkm };
items.Add(item);
}
}
}
this.Progress.SetMessage("Preparing result ...");
var samples = new HashSet<string>(from item in items
select item.Sample).OrderBy(m => m).ToList();
this.Progress.SetMessage(string.Format("There are {0} samples", samples.Count));
var genes = new HashSet<string>(from item in items
select item.Gene).OrderBy(m => m).ToList();
this.Progress.SetMessage(string.Format("There are {0} genes", genes.Count));
var map = ToDoubleDirectory(items);
this.Progress.SetMessage("Removing empty genes ...");
foreach (var gene in genes)
{
if (map[gene].All(m => m.Value.Count == "0"))
{
map.Remove(gene);
}
}
var finalGenes = map.Keys.OrderBy(m => m).ToList();
this.Progress.SetMessage("Outputing result ...");
OutputFile(samples, finalGenes, map, geneNameMap, m => Math.Round(double.Parse(m.Count)).ToString(), countFile);
OutputFile(samples, finalGenes, map, geneNameMap, m => m.FPKM, fpkmFile);
this.Progress.End();
return new[] { countFile, fpkmFile };
}
