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);
}
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);
}
public static bool GetHasAddtlMatrices(StreamReader reader, IList <int> expressionColIndices, char separator)
{
if (expressionColIndices.Count == 0)
{
return(false);
}
int expressionColIndex = expressionColIndices[0];
reader.ReadLine();
string line;
bool hasAddtl = false;
while ((line = reader.ReadLine()) != null)
{
if (TabSep.IsCommentLine(line, commentPrefix, commentPrefixExceptions))
{
continue;
}
string[] w = SplitLine(line, separator);
if (expressionColIndex < w.Length)
{
string s = StringUtils.RemoveWhitespace(w[expressionColIndex]);
hasAddtl = s.Contains(";");
break;
}
}
reader.Close();
return(hasAddtl);
}
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(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);
}
//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);
}
public static void ParseKinaseSubstrate(string filename, out string[] seqWins, out string[] subAccs,
out string[] kinases, out string[] kinAccs, out string[] species)
{
seqWins = TabSep.GetColumn("SITE_+/-7_AA", filename, 3, '\t');
subAccs = TabSep.GetColumn("SUB_ACC_ID", filename, 3, '\t');
kinases = TabSep.GetColumn("KINASE", filename, 3, '\t');
kinAccs = TabSep.GetColumn("KIN_ACC_ID", filename, 3, '\t');
species = TabSep.GetColumn("KIN_ORGANISM", filename, 3, '\t');
}
public static void ParseKnownMods(string filename, out string[] seqWins, out string[] accs, out string[] pubmedLtp,
out string[] pubmedMs2, out string[] cstMs2, out string[] species)
{
seqWins = TabSep.GetColumn("SITE_+/-7_AA", filename, 3, '\t');
accs = TabSep.GetColumn("ACC_ID", filename, 3, '\t');
pubmedLtp = TabSep.GetColumn("LT_LIT", filename, 3, '\t');
pubmedMs2 = TabSep.GetColumn("MS_LIT", filename, 3, '\t');
cstMs2 = TabSep.GetColumn("MS_CST", filename, 3, '\t');
species = TabSep.GetColumn("ORGANISM", filename, 3, '\t');
}
public static string[] GetAllKinaseSubstrateOrganisms()
{
string filename = GetKinaseSubstrateFile();
if (filename == null)
{
return(null);
}
string[] species = TabSep.GetColumn("KIN_ORGANISM", filename, 3, '\t');
return(ArrayUtils.UniqueValues(species));
}
public static void ParseRegulatorySites(string filename, out string[] seqWins, out string[] accs,
out string[] function, out string[] process, out string[] protInteract, out string[] otherInteract,
out string[] notes, out string[] species)
{
seqWins = TabSep.GetColumn("SITE_+/-7_AA", filename, 3, '\t');
accs = TabSep.GetColumn("ACC_ID", filename, 3, '\t');
function = TabSep.GetColumn("ON_FUNCTION", filename, 3, '\t');
process = TabSep.GetColumn("ON_PROCESS", filename, 3, '\t');
protInteract = TabSep.GetColumn("ON_PROT_INTERACT", filename, 3, '\t');
otherInteract = TabSep.GetColumn("ON_OTHER_INTERACT", filename, 3, '\t');
notes = TabSep.GetColumn("NOTES", filename, 3, '\t');
species = TabSep.GetColumn("ORGANISM", filename, 3, '\t');
}
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 static int GetRowCount(StreamReader reader, StreamReader auxReader, int[] mainColIndices,
List <Tuple <Relation[], int[], bool> > filters, char separator)
{
reader.BaseStream.Seek(0, SeekOrigin.Begin);
reader.ReadLine();
int count = 0;
bool hasAddtlMatrices = auxReader != null && GetHasAddtlMatrices(auxReader, mainColIndices, separator);
string line;
while ((line = reader.ReadLine()) != null)
{
while (TabSep.IsCommentLine(line, commentPrefix, commentPrefixExceptions))
{
line = reader.ReadLine();
}
if (IsValidLine(line, separator, filters, hasAddtlMatrices))
{
count++;
}
}
return(count);
}
private static void LoadMatrixData(IList <string> colNames, IList <string> colDescriptions, IList <int> mainColIndices,
IList <int> catColIndices, IList <int> numColIndices, IList <int> textColIndices, IList <int> multiNumColIndices,
string origin, IMatrixData matrixData, IDictionary <string, string[]> annotationRows, Action <int> progress,
Action <string> status, char separator, TextReader reader, StreamReader auxReader, int nrows,
bool shortenExpressionNames, List <Tuple <Relation[], int[], bool> > filters)
{
Dictionary <string, string[]> catAnnotatRows;
Dictionary <string, string[]> numAnnotatRows;
status("Reading data");
SplitAnnotRows(annotationRows, out catAnnotatRows, out numAnnotatRows);
List <string[][]> categoryAnnotation = new List <string[][]>();
for (int i = 0; i < catColIndices.Count; i++)
{
categoryAnnotation.Add(new string[nrows][]);
}
List <double[]> numericAnnotation = new List <double[]>();
for (int i = 0; i < numColIndices.Count; i++)
{
numericAnnotation.Add(new double[nrows]);
}
List <double[][]> multiNumericAnnotation = new List <double[][]>();
for (int i = 0; i < multiNumColIndices.Count; i++)
{
multiNumericAnnotation.Add(new double[nrows][]);
}
List <string[]> stringAnnotation = new List <string[]>();
for (int i = 0; i < textColIndices.Count; i++)
{
stringAnnotation.Add(new string[nrows]);
}
float[,] mainValues = new float[nrows, mainColIndices.Count];
float[,] qualityValues = null;
bool[,] isImputedValues = null;
bool hasAddtlMatrices = auxReader != null && GetHasAddtlMatrices(auxReader, mainColIndices, separator);
if (hasAddtlMatrices)
{
qualityValues = new float[nrows, mainColIndices.Count];
isImputedValues = new bool[nrows, mainColIndices.Count];
}
reader.ReadLine();
int count = 0;
string line;
while ((line = reader.ReadLine()) != null)
{
progress(100 * (count + 1) / nrows);
if (TabSep.IsCommentLine(line, commentPrefix, commentPrefixExceptions))
{
continue;
}
string[] w;
if (!IsValidLine(line, separator, filters, out w, hasAddtlMatrices))
{
continue;
}
for (int i = 0; i < mainColIndices.Count; i++)
{
if (mainColIndices[i] >= w.Length)
{
mainValues[count, i] = float.NaN;
}
else
{
string s = StringUtils.RemoveWhitespace(w[mainColIndices[i]]);
if (hasAddtlMatrices)
{
ParseExp(s, out mainValues[count, i], out isImputedValues[count, i], out qualityValues[count, i]);
}
else
{
if (count < mainValues.GetLength(0))
{
bool success = float.TryParse(s, out mainValues[count, i]);
if (!success)
{
mainValues[count, i] = float.NaN;
}
}
}
}
}
for (int i = 0; i < numColIndices.Count; i++)
{
if (numColIndices[i] >= w.Length)
{
numericAnnotation[i][count] = double.NaN;
}
else
{
double q;
bool success = double.TryParse(w[numColIndices[i]].Trim(), out q);
if (numericAnnotation[i].Length > count)
{
numericAnnotation[i][count] = success ? q : double.NaN;
}
}
}
for (int i = 0; i < multiNumColIndices.Count; i++)
{
if (multiNumColIndices[i] >= w.Length)
{
multiNumericAnnotation[i][count] = new double[0];
}
else
{
string q = w[multiNumColIndices[i]].Trim();
if (q.Length >= 2 && q[0] == '\"' && q[q.Length - 1] == '\"')
{
q = q.Substring(1, q.Length - 2);
}
if (q.Length >= 2 && q[0] == '\'' && q[q.Length - 1] == '\'')
{
q = q.Substring(1, q.Length - 2);
}
string[] ww = q.Length == 0 ? new string[0] : q.Split(';');
multiNumericAnnotation[i][count] = new double[ww.Length];
for (int j = 0; j < ww.Length; j++)
{
double q1;
bool success = double.TryParse(ww[j], out q1);
multiNumericAnnotation[i][count][j] = success ? q1 : double.NaN;
}
}
}
for (int i = 0; i < catColIndices.Count; i++)
{
if (catColIndices[i] >= w.Length)
{
categoryAnnotation[i][count] = new string[0];
}
else
{
string q = w[catColIndices[i]].Trim();
if (q.Length >= 2 && q[0] == '\"' && q[q.Length - 1] == '\"')
{
q = q.Substring(1, q.Length - 2);
}
if (q.Length >= 2 && q[0] == '\'' && q[q.Length - 1] == '\'')
{
q = q.Substring(1, q.Length - 2);
}
string[] ww = q.Length == 0 ? new string[0] : q.Split(';');
List <int> valids = new List <int>();
for (int j = 0; j < ww.Length; j++)
{
ww[j] = ww[j].Trim();
if (ww[j].Length > 0)
{
valids.Add(j);
}
}
ww = ArrayUtils.SubArray(ww, valids);
Array.Sort(ww);
if (categoryAnnotation[i].Length > count)
{
categoryAnnotation[i][count] = ww;
}
}
}
for (int i = 0; i < textColIndices.Count; i++)
{
if (textColIndices[i] >= w.Length)
{
stringAnnotation[i][count] = "";
}
else
{
string q = w[textColIndices[i]].Trim();
if (stringAnnotation[i].Length > count)
{
stringAnnotation[i][count] = RemoveSplitWhitespace(RemoveQuotes(q));
}
}
}
count++;
}
reader.Close();
string[] columnNames = ArrayUtils.SubArray(colNames, mainColIndices);
if (shortenExpressionNames)
{
columnNames = StringUtils.RemoveCommonSubstrings(columnNames, true);
}
string[] catColnames = ArrayUtils.SubArray(colNames, catColIndices);
string[] numColnames = ArrayUtils.SubArray(colNames, numColIndices);
string[] multiNumColnames = ArrayUtils.SubArray(colNames, multiNumColIndices);
string[] textColnames = ArrayUtils.SubArray(colNames, textColIndices);
matrixData.Name = origin;
matrixData.ColumnNames = RemoveQuotes(columnNames);
matrixData.Values.Set(mainValues);
if (hasAddtlMatrices)
{
matrixData.Quality.Set(qualityValues);
matrixData.IsImputed.Set(isImputedValues);
}
else
{
matrixData.Quality.Set(new float[mainValues.GetLength(0), mainValues.GetLength(1)]);
matrixData.IsImputed.Set(new bool[mainValues.GetLength(0), mainValues.GetLength(1)]);
}
matrixData.SetAnnotationColumns(RemoveQuotes(textColnames), stringAnnotation, RemoveQuotes(catColnames),
categoryAnnotation, RemoveQuotes(numColnames), numericAnnotation, RemoveQuotes(multiNumColnames),
multiNumericAnnotation);
if (colDescriptions != null)
{
string[] columnDesc = ArrayUtils.SubArray(colDescriptions, mainColIndices);
string[] catColDesc = ArrayUtils.SubArray(colDescriptions, catColIndices);
string[] numColDesc = ArrayUtils.SubArray(colDescriptions, numColIndices);
string[] multiNumColDesc = ArrayUtils.SubArray(colDescriptions, multiNumColIndices);
string[] textColDesc = ArrayUtils.SubArray(colDescriptions, textColIndices);
matrixData.ColumnDescriptions = new List <string>(columnDesc);
matrixData.NumericColumnDescriptions = new List <string>(numColDesc);
matrixData.CategoryColumnDescriptions = new List <string>(catColDesc);
matrixData.StringColumnDescriptions = new List <string>(textColDesc);
matrixData.MultiNumericColumnDescriptions = new List <string>(multiNumColDesc);
}
foreach (string key in catAnnotatRows.Keys)
{
string name = key;
string[] svals = ArrayUtils.SubArray(catAnnotatRows[key], mainColIndices);
string[][] cat = new string[svals.Length][];
for (int i = 0; i < cat.Length; i++)
{
string s = svals[i].Trim();
cat[i] = s.Length > 0 ? s.Split(';') : new string[0];
List <int> valids = new List <int>();
for (int j = 0; j < cat[i].Length; j++)
{
cat[i][j] = cat[i][j].Trim();
if (cat[i][j].Length > 0)
{
valids.Add(j);
}
}
cat[i] = ArrayUtils.SubArray(cat[i], valids);
Array.Sort(cat[i]);
}
matrixData.AddCategoryRow(name, name, cat);
}
foreach (string key in numAnnotatRows.Keys)
{
string name = key;
string[] svals = ArrayUtils.SubArray(numAnnotatRows[key], mainColIndices);
double[] num = new double[svals.Length];
for (int i = 0; i < num.Length; i++)
{
string s = svals[i].Trim();
num[i] = double.NaN;
double.TryParse(s, out num[i]);
}
matrixData.AddNumericRow(name, name, num);
}
matrixData.Origin = origin;
progress(0);
status("");
}
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();
}
