private static string ProcessDataReadFromFile(IDataWithAnnotationRows mdata, Parameters param)
{
Parameter <string> fp = param.GetParam <string>("Input file");
string filename = fp.Value;
string[] colNames;
try {
colNames = TabSep.GetColumnNames(filename, '\t');
} catch (Exception) {
return("Could not open file " + filename + ". It maybe open in another program.");
}
int nameIndex = GetNameIndex(colNames);
if (nameIndex < 0)
{
return("Error: the file has to contain a column called 'Name'.");
}
if (colNames.Length < 2)
{
return("Error: the file does not contain a numerical column.");
}
string[] nameCol = TabSep.GetColumn(colNames[nameIndex], filename, '\t');
Dictionary <string, int> map = ArrayUtils.InverseMap(nameCol);
for (int i = 0; i < colNames.Length; i++)
{
if (i == nameIndex)
{
continue;
}
string groupName = colNames[i];
string[] groupCol = TabSep.GetColumn(groupName, filename, '\t');
double[] newCol = new double[mdata.ColumnCount];
for (int j = 0; j < newCol.Length; j++)
{
string colName = mdata.ColumnNames[j];
if (!map.ContainsKey(colName))
{
newCol[j] = double.NaN;
continue;
}
int ind = map[colName];
string group = groupCol[ind] ?? "";
group = group.Trim();
if (string.IsNullOrEmpty(group))
{
newCol[j] = double.NaN;
}
else
{
if (!Parser.TryDouble(group, out newCol[j]))
{
newCol[j] = double.NaN;
}
}
}
mdata.AddNumericRow(groupName, groupName, newCol);
}
return(null);
}
private static string ProcessDataReadFromFile(IDataWithAnnotationRows mdata, Parameters param)
{
Parameter <string> fp = param.GetParam <string>("Input file");
string filename = fp.Value;
string[] colNames;
try
{
colNames = TabSep.GetColumnNames(filename, '\t');
}
catch (Exception)
{
return("Could not open file " + filename + ". It maybe open in another program.");
}
int nameIndex = GetNameIndex(colNames);
if (nameIndex < 0)
{
return("Error: the file has to contain a column called 'Name'.");
}
if (colNames.Length < 2)
{
return("Error: the file does not contain a grouping column.");
}
string[] nameCol = TabSep.GetColumn(colNames[nameIndex], filename, '\t');
for (int i = 0; i < nameCol.Length; i++)
{
MessageBox.Show(nameCol[i].ToString());
}
return(null);
}
private static string ProcessDataReadFromFileList(IMatrixData mdata, Parameters param)
{
Parameter <string> fp = param.GetParam <string>("Input file");
string filename = fp.Value;
string[] colNames;
try
{
colNames = TabSep.GetColumnNames(filename, '\t');
}
catch (Exception)
{
return("Could not open file " + filename + ". It maybe open in another program.");
}
int nameIndex = GetNameIndex(colNames);
string[] nameCol = TabSep.GetColumn(colNames[nameIndex], filename, '\t');
MessageBox.Show(nameCol.Length.ToString());
MessageBox.Show(mdata.ColumnCount.ToString());
for (int i = 0; i < mdata.ColumnCount; i++)
{
MessageBox.Show(nameCol.Length.ToString());
MessageBox.Show(mdata.ColumnCount.ToString());
/* if (nameCol.Length != mdata.ColumnCount)
* {
* return "ERROR";
* }
* else { mdata.ColumnNames[i] = nameCol[i]; } */
}
return(null);
}
//gene enrichment analysis for both Basic PECA and PECA-N
public static IMatrixData GetGOEnr(IMatrixData mdata, string workingDir, int option)//, out string errString)
{
char separator = '\t';
string filename = Path.Combine(workingDir, @".\Goterms.txt");
IMatrixData mNew = (IMatrixData)mdata.CreateNewInstance();
string name = "GSA";
if (option == 0)
{
name = name + "_Degradation";
}
else if (option == 1)
{
name = name + "_Synthesis";
}
mNew.Clear();
mNew.Name = name;
mNew.AltName = name;
//update
//mNew.AltName = "Gene Set Enrichment Analysis";
//mNew.Description = "Gene Set Enrichment Analysis";
string[] colNames = TabSep.GetColumnNames(filename, 0, PerseusUtils.commentPrefix,
PerseusUtils.commentPrefixExceptions, null, separator);
string[][] cols = TabSep.GetColumns(colNames, filename, 0, PerseusUtils.commentPrefix,
PerseusUtils.commentPrefixExceptions, separator);
int nrows = TabSep.GetRowCount(filename);
mNew.Values.Init(nrows, 0);
mNew.SetAnnotationColumns(new List <string>(colNames), new List <string>(colNames), new List <string[]>(cols), new List <string>(),
new List <string>(), new List <string[][]>(), new List <string>(), new List <string>(), new List <double[]>(),
new List <string>(), new List <string>(), new List <double[][]>());
//convert the ones not matching regex to numeric
string pattern = @"^((?!id|name|members).)*$";
Regex numericReg = new Regex(pattern);
List <int> numericList = new List <int>();
for (int i = 0; i < colNames.Length; i++)
{
if (numericReg.Match(colNames[i]).Success)
{
numericList.Add(i);
}
}
StringToNumerical(numericList, mNew);
return(mNew);
}
private static string ProcessDataReadFromFile(IMatrixData mdata, Parameters param)
{
Parameter <string> fp = param.GetParam <string>("Input file");
string filename = fp.Value;
string[] colNames = TabSep.GetColumnNames(filename, '\t');
int nameIndex = GetNameIndex(colNames);
if (nameIndex < 0)
{
return("Error: the file has to contain a column called 'Name'.");
}
if (colNames.Length < 2)
{
return("Error: the file does not contain a grouping column.");
}
string[] nameCol = TabSep.GetColumn(colNames[nameIndex], filename, '\t');
Dictionary <string, int> map = ArrayUtils.InverseMap(nameCol);
for (int i = 0; i < colNames.Length; i++)
{
if (i == nameIndex)
{
continue;
}
string groupName = colNames[i];
string[] groupCol = TabSep.GetColumn(groupName, filename, '\t');
string[][] newCol = new string[mdata.ColumnCount][];
for (int j = 0; j < newCol.Length; j++)
{
string colName = mdata.ColumnNames[j];
if (!map.ContainsKey(colName))
{
newCol[j] = new string[0];
continue;
}
int ind = map[colName];
string group = groupCol[ind] ?? "";
group = group.Trim();
if (string.IsNullOrEmpty(group))
{
newCol[j] = new string[0];
}
else
{
string[] w = group.Split(';');
Array.Sort(w);
for (int k = 0; k < w.Length; k++)
{
w[k] = w[k].Trim();
}
newCol[j] = w;
}
}
mdata.AddCategoryRow(groupName, groupName, newCol);
}
return(null);
}
private static string ProcessDataReadFromFileList(IMatrixData mdata, Parameters param)
{
Parameter <string> fp = param.GetParam <string>("Input file");
string filename = fp.Value;
string[] colNames;
try{
colNames = TabSep.GetColumnNames(filename, '\t');
} catch (Exception) {
return("Could not open file " + filename + ". It maybe open in another program.");
}
int nameIndex = GetNameIndex(colNames);
string[] nameCol = TabSep.GetColumn(colNames[nameIndex], filename, '\t');
return(null);
}
private static void LoadSplit(IMatrixData mdata, string filename, bool csv)
{
char separator = csv ? ',' : '\t';
string[] colNames = TabSep.GetColumnNames(filename, 0, PerseusUtils.commentPrefix,
PerseusUtils.commentPrefixExceptions, null, separator);
string[][] cols = TabSep.GetColumns(colNames, filename, 0, PerseusUtils.commentPrefix,
PerseusUtils.commentPrefixExceptions, separator);
int nrows = TabSep.GetRowCount(filename);
mdata.Values.Init(nrows, 0);
mdata.SetAnnotationColumns(new List <string>(colNames), new List <string>(colNames), new List <string[]>(cols), new List <string>(),
new List <string>(), new List <string[][]>(), new List <string>(), new List <string>(), new List <double[]>(),
new List <string>(), new List <string>(), new List <double[][]>());
mdata.Origin = filename;
}
public void TestGetColumnNamesFromGzippedFile()
{
var lines = new[]
{
"Col_1 Col_2",
"a b",
};
var tmpFile = Path.GetTempFileName() + ".gz";
using (var memory = new MemoryStream(Encoding.UTF8.GetBytes(string.Join("\n", lines))))
using (var outFile = File.Create(tmpFile))
using (var gzip = new GZipStream(outFile, CompressionMode.Compress))
{
memory.CopyTo(gzip);
}
var columnNames = TabSep.GetColumnNames(tmpFile, ' ');
CollectionAssert.AreEqual(new [] { "Col_1", "Col_2" }, columnNames);
}
//this function is modified from PerseusPluginLib/Load/UnstructuredTxtUpload.cs LoadSplit function
//obtains the output from fdr.exe (so only applicable to PECA CORE and N)
public static void GetOutput(IMatrixData mdata, Parameters param, Parameters dataParam, string filename, string geneName, string expSeries1 = "Expression Series 1", int numOfSeries = 2)
{
char separator = '\t';
//gene name column name is not included in the file so need to replace it
//gene name
ReplaceFirstLine(filename, geneName);
string[] colNames = TabSep.GetColumnNames(filename, 0, PerseusUtils.commentPrefix,
PerseusUtils.commentPrefixExceptions, null, separator);
string[][] cols = TabSep.GetColumns(colNames, filename, 0, PerseusUtils.commentPrefix,
PerseusUtils.commentPrefixExceptions, separator);
int nrows = TabSep.GetRowCount(filename);
string[] expressionColumnsNames = ArrayUtils.Concat(mdata.ColumnNames, mdata.NumericColumnNames);
mdata.Clear();
mdata.Name = "PECA Analysis";
mdata.Values.Init(nrows, 0);
mdata.SetAnnotationColumns(new List <string>(colNames), new List <string>(colNames), new List <string[]>(cols), new List <string>(),
new List <string>(), new List <string[][]>(), new List <string>(), new List <string>(), new List <double[]>(),
new List <string>(), new List <string>(), new List <double[][]>());
//be careful with changes of Number of time points in the future
int numOfExpCols = numOfSeries * dataParam.GetParam <int[]>(expSeries1).Value.Length;
//file format is structured so that expressions columns are before numeric ones
//so convert the numeric ones before expression columns
//first column guaranteed to be the name column
int[] expList = Enumerable.Range(1, numOfExpCols).ToArray();
int[] numericList = Enumerable.Range(numOfExpCols + 1, colNames.Count() - numOfExpCols - 1).ToArray();
StringToNumerical(numericList, mdata);
StringToExpression(expList, mdata);
}
public void LoadData(IMatrixData matrixData, Parameters parameters, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int ind = parameters.GetParam <int>("Organism").Value;
string filename = GetOrganismFiles()[ind];
List <string> stringColnames = new List <string>(TabSep.GetColumnNames(filename, '\t'));
List <string[]> stringCols = new List <string[]>();
foreach (string t in stringColnames)
{
string[] col = TabSep.GetColumn(t, filename, '\t');
stringCols.Add(col);
}
matrixData.Name = "Gene list";
matrixData.ColumnNames = new List <string>();
matrixData.Values.Init(stringCols[0].Length, 0);
matrixData.SetAnnotationColumns(stringColnames, stringCols,
new List <string>(), new List <string[][]>(), new List <string>(), new List <double[]>(), new List <string>(),
new List <double[][]>());
matrixData.Origin = "Gene list";
}
internal void UpdateFile(string filename)
{
textBox1.Text = filename;
bool csv = filename.ToLower().EndsWith(".csv");
char separator = csv ? ',' : '\t';
string[] colNames;
Dictionary <string, string[]> annotationRows = new Dictionary <string, string[]>();
try{
colNames = TabSep.GetColumnNames(filename, PerseusUtils.commentPrefix, PerseusUtils.commentPrefixExceptions,
annotationRows, separator);
} catch (Exception) {
MessageBox.Show("Could not open the file '" + filename +
"'. It is probably opened by another program.");
return;
}
string[] colTypes = null;
if (annotationRows.ContainsKey("Type"))
{
colTypes = annotationRows["Type"];
annotationRows.Remove("Type");
}
string msg = TabSep.CanOpen(filename);
if (msg != null)
{
MessageBox.Show(msg);
return;
}
multiListSelectorControl1.Init(colNames);
if (colTypes != null)
{
FormUtils.SelectExact(colNames, colTypes, multiListSelectorControl1);
}
else
{
FormUtils.SelectHeuristic(colNames, multiListSelectorControl1);
}
}
public void LoadData(IMatrixData mdata, Parameters parameters, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
PerseusLoadMatrixParam par = (PerseusLoadMatrixParam)parameters.GetParam("File");
string filename = par.Filename;
if (string.IsNullOrEmpty(filename))
{
processInfo.ErrString = "Please specify a filename";
return;
}
if (!File.Exists(filename))
{
processInfo.ErrString = "File '" + filename + "' does not exist.";
return;
}
string ftl = filename.ToLower();
bool csv = ftl.EndsWith(".csv") || ftl.EndsWith(".csv.gz");
char separator = csv ? ',' : '\t';
string[] colNames;
Dictionary <string, string[]> annotationRows = new Dictionary <string, string[]>();
try{
colNames = TabSep.GetColumnNames(filename, PerseusUtils.commentPrefix, PerseusUtils.commentPrefixExceptions,
annotationRows, separator);
} catch (Exception) {
processInfo.ErrString = "Could not open the file '" + filename + "'. It is probably opened in another program.";
return;
}
string origin = filename;
int[] eInds = par.MainColumnIndices;
int[] nInds = par.NumericalColumnIndices;
int[] cInds = par.CategoryColumnIndices;
int[] tInds = par.TextColumnIndices;
int[] mInds = par.MultiNumericalColumnIndices;
List <Tuple <Relation[], int[], bool> > filters = new List <Tuple <Relation[], int[], bool> >();
string errString;
foreach (Parameters p in par.MainFilterParameters)
{
PerseusUtils.AddFilter(filters, p, eInds, out errString);
if (errString != null)
{
processInfo.ErrString = errString;
return;
}
}
foreach (Parameters p in par.NumericalFilterParameters)
{
PerseusUtils.AddFilter(filters, p, nInds, out errString);
if (errString != null)
{
processInfo.ErrString = errString;
return;
}
}
int nrows = GetRowCount(filename, eInds, filters, separator);
StreamReader reader = FileUtils.GetReader(filename);
StreamReader auxReader = FileUtils.GetReader(filename);
PerseusUtils.LoadMatrixData(annotationRows, eInds, cInds, nInds, tInds, mInds, processInfo, colNames, mdata, reader,
auxReader, nrows, origin, separator, par.ShortenExpressionColumnNames, filters);
reader.Close();
auxReader.Close();
GC.Collect();
}
