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 });
}