public void TestSetGeneString()
{
AnnovarGenomeSummaryItem item = new AnnovarGenomeSummaryItem();
item.GeneString = "TP53";
Assert.AreEqual(1, item.Genes.Count);
Assert.AreEqual("TP53", item.Genes[0].Name);
item.GeneString = "FLJ20518(dist=2106),LOC401074(dist=2958)";
Assert.AreEqual(2, item.Genes.Count);
Assert.AreEqual("FLJ20518", item.Genes[0].Name);
Assert.AreEqual("dist=2106", item.Genes[0].Annotation);
Assert.AreEqual("LOC401074", item.Genes[1].Name);
Assert.AreEqual("dist=2958", item.Genes[1].Annotation);
item.GeneString = "FLJ20518(dist=2106),NONE(dist=NONE)";
Assert.AreEqual(1, item.Genes.Count);
Assert.AreEqual("FLJ20518", item.Genes[0].Name);
Assert.AreEqual("dist=2106", item.Genes[0].Annotation);
item.GeneString = "FRAS1(NM_025074:exon47:c.6584-2A>G)";
Assert.AreEqual(1, item.Genes.Count);
Assert.AreEqual("FRAS1", item.Genes[0].Name);
Assert.AreEqual("NM_025074:exon47:c.6584-2A>G", item.Genes[0].Annotation);
}
public override IEnumerable<string> Process(string fileName)
{
//Write the stream data of workbook to the root directory
using (FileStream file = new FileStream(this.targetFile, FileMode.Create))
{
var book = new HSSFWorkbook();
//cell style for hyperlinks
//by default hyperlinks are blue and underlined
var hlink_style = book.CreateCellStyle();
IFont hlink_font = book.CreateFont();
hlink_font.Underline = (byte)FontUnderlineType.SINGLE;
hlink_font.Color = HSSFColor.BLUE.index;
hlink_style.SetFont(hlink_font);
hlink_style.WrapText = true;
var wrap_style = book.CreateCellStyle();
wrap_style.WrapText = true;
var numeric_style = book.CreateCellStyle();
numeric_style.DataFormat = 0xb;
Dictionary<string, string> genenames = File.Exists(this.affyAnnotationFile) ? AnnotationFile.GetGeneSymbolDescriptionMap(this.affyAnnotationFile) : new Dictionary<string, string>();
ISheet all = book.CreateSheet("all");
AnnovarGenomeSummaryItem item = new AnnovarGenomeSummaryItem();
var sr = new StreamReader(fileName);
var headers = sr.ReadLine().Split(',').ToList();
var geneIndex = headers.FindIndex(m => m.Equals("Gene") || m.Equals("Gene.refGene"));
var funcIndex = FindIndex(geneIndex, headers.FindIndex(m => m.Equals("Func") || m.Equals("Func.refGene")));
var exonicIndex = FindIndex(geneIndex, headers.FindIndex(m => m.Equals("ExonicFunc") || m.Equals("ExonicFunc.refGene")));
var dbsnpIndex = FindIndex(geneIndex, headers.FindIndex(m => m.ToLower().StartsWith("dbsnp") || m.ToLower().StartsWith("snp")));
var chrIndex = headers.IndexOf("Chr");
var startIndex = headers.IndexOf("Start");
var endIndex = headers.IndexOf("End");
var otherInfoIndex = headers.IndexOf("Otherinfo");
//handle the headers. The length of headers may less than the data.
var firstrow = all.CreateRow(0);
for (int i = 0; i <= geneIndex; i++)
{
firstrow.CreateCell(i).SetCellValue(headers[i]);
}
firstrow.CreateCell(geneIndex + 1).SetCellValue("Description");
for (int i = geneIndex + 1; i < otherInfoIndex; i++)
{
firstrow.CreateCell(i + 1).SetCellValue(headers[i]);
}
firstrow.CreateCell(otherInfoIndex + 1).SetCellValue("Location");
bool? isMuTect = null;
bool isTableVersion = false; //using table_annovar.pl or summarize_annovar.pl
bool hasLOD = false;
double lod = 0.0;
//handle data
using (CsvReader csv = new CsvReader(sr, false))
{
int nRow = 0;
while (csv.ReadNextRecord())
{
if (!isMuTect.HasValue)
{
isTableVersion = csv.FieldCount == headers.Count;
isMuTect = mutectRegex.Match(csv[csv.FieldCount - 2]).Success;
hasLOD = double.TryParse(csv[csv.FieldCount - 1], out lod);
if (isMuTect.Value)
{
firstrow.CreateCell(otherInfoIndex + 2).SetCellValue("Normal");
firstrow.CreateCell(otherInfoIndex + 3).SetCellValue("Tumor");
firstrow.CreateCell(otherInfoIndex + 4).SetCellValue("FisherExactTest");
all.SetDefaultColumnStyle(otherInfoIndex + 4, numeric_style);
if (hasLOD)
{
firstrow.CreateCell(otherInfoIndex + 5).SetCellValue("LOD_FStar");
all.SetDefaultColumnStyle(otherInfoIndex + 5, numeric_style);
}
}
else
{
for (int i = otherInfoIndex; i < headers.Count; i++)
{
firstrow.CreateCell(i + 2).SetCellValue(headers[i]);
}
}
}
nRow++;
var row = all.CreateRow(nRow);
for (int i = 0; i < geneIndex; i++)
{
row.CreateCell(i).SetCellValue(csv[i]);
}
//add link for gene symbol
item.GeneString = csv[geneIndex];
var cell = row.CreateCell(geneIndex);
cell.Hyperlink = new HSSFHyperlink(HyperlinkType.URL)
{
Address = string.Format("http://www.genecards.org/cgi-bin/carddisp.pl?gene={0}", item.Genes[0].Name)
};
cell.CellStyle = hlink_style;
cell.SetCellValue((from g in item.Genes select g.Name).Merge("\n"));
//gene description
var desCell = row.CreateCell(geneIndex + 1);
desCell.CellStyle = wrap_style;
desCell.SetCellValue((from gene in item.Genes
let description = genenames.ContainsKey(gene.Name) ? genenames[gene.Name] : " "
select description).Merge("\n"));
//add location information
for (int i = geneIndex + 1; i < otherInfoIndex; i++)
{
row.CreateCell(i + 1).SetCellValue(csv[i]);
}
var locationCell = row.CreateCell(otherInfoIndex + 1);
locationCell.SetCellValue(string.Format("{0}:{1}-{2}", csv[chrIndex], csv[startIndex], csv[endIndex]));
if (isMuTect.Value)
{
Match normal, tumor;
if (isTableVersion)
{
var parts = csv[csv.FieldCount - 1].Split('\t');
if (hasLOD)
{
normal = mutectRegex.Match(parts[parts.Length - 3]);
tumor = mutectRegex.Match(parts[parts.Length - 2]);
lod = double.Parse(parts[parts.Length - 1]);
}
else
{
normal = mutectRegex.Match(parts[parts.Length - 2]);
tumor = mutectRegex.Match(parts[parts.Length - 1]);
}
}
else
{
if (hasLOD)
{
tumor = mutectRegex.Match(csv[csv.FieldCount - 3]);
normal = mutectRegex.Match(csv[csv.FieldCount - 2]);
lod = double.Parse(csv[csv.FieldCount - 1]);
}
else
{
tumor = mutectRegex.Match(csv[csv.FieldCount - 2]);
normal = mutectRegex.Match(csv[csv.FieldCount - 1]);
}
}
FisherExactTestResult fetr = new FisherExactTestResult();
fetr.Sample1.Succeed = int.Parse(normal.Groups[1].Value);
fetr.Sample1.Failed = int.Parse(normal.Groups[2].Value);
fetr.Sample2.Succeed = int.Parse(tumor.Groups[1].Value);
fetr.Sample2.Failed = int.Parse(tumor.Groups[2].Value);
row.CreateCell(otherInfoIndex + 2).SetCellValue(string.Format("{0}:{1}", fetr.Sample1.Succeed, fetr.Sample1.Failed));
row.CreateCell(otherInfoIndex + 3).SetCellValue(string.Format("{0}:{1}", fetr.Sample2.Succeed, fetr.Sample2.Failed));
row.CreateCell(otherInfoIndex + 4).SetCellValue(fetr.CalculateTwoTailPValue());
if (hasLOD)
{
row.CreateCell(otherInfoIndex + 5).SetCellValue(lod);
}
}
else
{
for (int i = otherInfoIndex; i < csv.FieldCount; i++)
{
row.CreateCell(i + 2).SetCellValue(csv[i]);
}
}
if (dbsnpIndex > 0)
{
var dbsnpcell = row.GetCell(dbsnpIndex);
var dbsnp = dbsnpcell.StringCellValue;
if (!string.IsNullOrEmpty(dbsnp))
{
dbsnpcell.Hyperlink = new HSSFHyperlink(HyperlinkType.URL)
{
Address = string.Format("http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs={0}", dbsnp.Substring(2))
};
dbsnpcell.CellStyle = (hlink_style);
}
}
}
}
all.SetColumnWidth(chrIndex, 5 * 256);
all.SetColumnWidth(startIndex, 13 * 256);
all.SetColumnWidth(endIndex, 13 * 256);
all.SetColumnWidth(funcIndex, 15 * 256);
all.SetColumnWidth(geneIndex, 13 * 256);
all.SetColumnWidth(geneIndex + 1, 60 * 256);
all.SetColumnWidth(exonicIndex, 20 * 256);
all.SetColumnWidth(dbsnpIndex, 15 * 256);
all.SetColumnWidth(otherInfoIndex + 1, 22 * 256);
if (isMuTect.Value)
{
all.SetColumnWidth(otherInfoIndex + 2, 10 * 256);
all.SetColumnWidth(otherInfoIndex + 3, 10 * 256);
all.SetColumnWidth(otherInfoIndex + 4, 10 * 256);
if (hasLOD)
{
all.SetColumnWidth(otherInfoIndex + 5, 10 * 256);
}
}
book.Write(file);
}
return new string[] { targetFile };
}
public override IEnumerable<string> Process()
{
//Write the stream data of workbook to the root directory
using (FileStream file = new FileStream(options.OutputFile, FileMode.Create))
{
var book = new HSSFWorkbook();
//cell style for hyperlinks
//by default hyperlinks are blue and underlined
var hlinkStyle = book.CreateCellStyle();
var hlinkFont = book.CreateFont();
hlinkFont.Underline = (byte)FontUnderlineType.SINGLE;
hlinkFont.Color = HSSFColor.BLUE.index;
hlinkStyle.SetFont(hlinkFont);
hlinkStyle.WrapText = true;
var wrapStyle = book.CreateCellStyle();
wrapStyle.WrapText = true;
var numericStyle = book.CreateCellStyle();
numericStyle.DataFormat = 0xb;
var genenames = File.Exists(options.AffyAnnotationFile) ? AnnotationFile.GetGeneSymbolDescriptionMap(options.AffyAnnotationFile) : new Dictionary<string, string>();
var all = book.CreateSheet("all");
var item = new AnnovarGenomeSummaryItem();
using (var sr = new StreamReader(options.InputFile))
{
int nRow = 0;
bool isMutect = false;
bool hasLod = false;
string tumorSampleName = "", normalSampleName = "";
string line;
//ignore the comments
while ((line = sr.ReadLine()) != null)
{
if (!line.StartsWith("#"))
{
break;
}
var row = all.CreateRow(nRow++);
row.CreateCell(0).SetCellValue(line);
if (line.StartsWith("##INFO=<ID=LOD"))
{
hasLod = true;
}
if (line.StartsWith("##MuTect"))
{
isMutect = true;
tumorSampleName = line.StringAfter("tumor_sample_name=").StringBefore(" ");
normalSampleName = line.StringAfter("normal_sample_name=").StringBefore(" ");
}
}
if (line == null)
{
throw new Exception("No entries in file " + options.InputFile);
}
var headers = line.Split('\t').ToList();
//original index
var geneIndex = headers.FindIndex(m => m.Equals("Gene") || m.Equals("Gene.refGene"));
var oldInfoIndex = headers.FindIndex(m => m.ToLower().StartsWith("info"));
var newInfoIndex = FindIndex(oldInfoIndex, geneIndex);
//relative index
var funcIndex = FindIndex(geneIndex, headers.FindIndex(m => m.Equals("Func") || m.Equals("Func.refGene")));
var exonicIndex = FindIndex(geneIndex, headers.FindIndex(m => m.Equals("ExonicFunc") || m.Equals("ExonicFunc.refGene")));
var dbsnpIndex = FindIndex(geneIndex, headers.FindIndex(m => m.ToLower().StartsWith("dbsnp") || m.ToLower().StartsWith("snp")));
var chrIndex = headers.IndexOf("Chr");
var startIndex = headers.IndexOf("Start");
var endIndex = headers.IndexOf("End");
var tumorIndex = headers.IndexOf(tumorSampleName);
var normalIndex = headers.IndexOf(normalSampleName);
hasLod = hasLod && oldInfoIndex != -1;
//handle the headers. The length of headers may less than the data.
var firstrow = all.CreateRow(nRow++);
for (int i = 0; i <= geneIndex; i++)
{
firstrow.CreateCell(i).SetCellValue(headers[i]);
}
firstrow.CreateCell(geneIndex + 1).SetCellValue("Description");
for (int i = geneIndex + 1; i < headers.Count; i++)
{
if (isMutect)
{
if (i == tumorIndex)
{
firstrow.CreateCell(i + 1).SetCellValue("Tumor:" + tumorSampleName);
continue;
}
if (i == normalIndex)
{
firstrow.CreateCell(i + 1).SetCellValue("Normal:" + normalSampleName);
continue;
}
}
firstrow.CreateCell(i + 1).SetCellValue(headers[i]);
}
var lastcol = headers.Count + 1;
if (hasLod)
{
firstrow.CreateCell(lastcol++).SetCellValue("TLodFstar");
}
if (isMutect)
{
firstrow.CreateCell(lastcol++).SetCellValue("NormalAlleles");
firstrow.CreateCell(lastcol++).SetCellValue("TumorAlleles");
firstrow.CreateCell(lastcol++).SetCellValue("FisherExactTest");
}
while ((line = sr.ReadLine()) != null)
{
var parts = line.Split('\t');
if (parts.Length < geneIndex)
{
break;
}
var row = all.CreateRow(nRow++);
for (int i = 0; i < geneIndex; i++)
{
row.CreateCell(i).SetCellValue(parts[i]);
}
item.GeneString = parts[geneIndex];
if (item.Genes.Count > 0)
{
//add link for gene symbol
var cell = row.CreateCell(geneIndex);
cell.Hyperlink = new HSSFHyperlink(HyperlinkType.URL)
{
Address = string.Format("http://www.genecards.org/cgi-bin/carddisp.pl?gene={0}", item.Genes[0].Name)
};
cell.CellStyle = hlinkStyle;
cell.SetCellValue((from g in item.Genes select g.Name).Merge("\n"));
//gene description
var desCell = row.CreateCell(geneIndex + 1);
desCell.CellStyle = wrapStyle;
desCell.SetCellValue((from gene in item.Genes
let description = genenames.ContainsKey(gene.Name) ? genenames[gene.Name] : " "
select description).Merge("\n"));
}
//add other information
for (int i = geneIndex + 1; i < headers.Count; i++)
{
row.CreateCell(i + 1).SetCellValue(parts[i]);
}
lastcol = headers.Count + 1;
if (hasLod)
{
row.CreateCell(lastcol++).SetCellValue(parts[oldInfoIndex].StringAfter("LOD=").StringBefore(";"));
}
if (isMutect)
{
Match normal = SomaticMutationUtils.MutectPattern.Match(parts[normalIndex]);
Match tumor = SomaticMutationUtils.MutectPattern.Match(parts[tumorIndex]);
var fetr = new FisherExactTestResult();
fetr.Sample1.Succeed = int.Parse(normal.Groups[1].Value);
fetr.Sample1.Failed = int.Parse(normal.Groups[2].Value);
fetr.Sample2.Succeed = int.Parse(tumor.Groups[1].Value);
fetr.Sample2.Failed = int.Parse(tumor.Groups[2].Value);
row.CreateCell(lastcol++).SetCellValue(string.Format("{0}:{1}", fetr.Sample1.Succeed, fetr.Sample1.Failed));
row.CreateCell(lastcol++).SetCellValue(string.Format("{0}:{1}", fetr.Sample2.Succeed, fetr.Sample2.Failed));
row.CreateCell(lastcol).SetCellValue(fetr.CalculateTwoTailPValue());
}
if (dbsnpIndex > 0)
{
var dbsnpcell = row.GetCell(dbsnpIndex);
var dbsnp = dbsnpcell.StringCellValue;
if (!string.IsNullOrEmpty(dbsnp))
{
dbsnpcell.Hyperlink = new HSSFHyperlink(HyperlinkType.URL)
{
Address = string.Format("http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs={0}", dbsnp.Substring(2))
};
dbsnpcell.CellStyle = (hlinkStyle);
}
}
}
all.SetColumnWidth(chrIndex, 5 * 256);
all.SetColumnWidth(startIndex, 11 * 256);
all.SetColumnWidth(endIndex, 11 * 256);
all.SetColumnWidth(funcIndex, 15 * 256);
all.SetColumnWidth(geneIndex, 15 * 256);
all.SetColumnWidth(geneIndex + 1, 60 * 256);
all.SetColumnWidth(exonicIndex, 20 * 256);
all.SetColumnWidth(dbsnpIndex, 15 * 256);
lastcol = headers.Count + 1;
if (hasLod)
{
all.SetColumnWidth(newInfoIndex, 15 * 256);
all.SetColumnWidth(lastcol++, 10 * 256);
}
if (isMutect)
{
all.SetColumnWidth(lastcol++, 10 * 256);
all.SetColumnWidth(lastcol++, 10 * 256);
all.SetColumnWidth(lastcol, 10 * 256);
}
}
book.Write(file);
}
return new string[] { options.OutputFile };
}
