public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int[] cols = param.GetMultiChoiceParam("Columns").Value;
HashSet<int> w = ArrayUtils.ToHashSet(param.GetMultiChoiceParam("Calculate").Value);
bool[] include = new bool[SummaryStatisticsRows.procs.Length];
double[][] rowws = new double[SummaryStatisticsRows.procs.Length][];
for (int i = 0; i < include.Length; i++){
include[i] = w.Contains(i);
if (include[i]){
rowws[i] = new double[cols.Length];
}
}
for (int i = 0; i < cols.Length; i++){
double[] vals = GetColumn(cols[i], mdata);
for (int j = 0; j < include.Length; j++){
if (include[j]){
rowws[j][i] = SummaryStatisticsRows.procs[j].Item2(vals);
}
}
}
List<double[]> ex = new List<double[]>();
List<string> names = new List<string>();
for (int i = 0; i < include.Length; i++){
if (include[i]){
ex.Add(rowws[i]);
names.Add(SummaryStatisticsRows.procs[i].Item1);
}
}
float[,] exVals = GetExVals(ex);
string[] colNames = GetColNames(mdata, cols);
mdata.SetData("Summary", new List<string>(names.ToArray()), exVals, new List<string>(new[]{"Columns"}),
new List<string[]>(new[]{colNames}), mdata.CategoryRowNames,
TransformCategories(mdata, cols, mdata.ExpressionColumnCount), mdata.NumericRowNames,
TransformNumeric(mdata.NumericRows, cols, mdata.ExpressionColumnCount), new List<string>(), new List<double[][]>());
}
public void ProcessData(IMatrixData data, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
bool falseAreIndicated = param.GetSingleChoiceParam("Indicated are").Value == 0;
int catCol = param.GetSingleChoiceParam("In column").Value;
string word = param.GetStringParam("Indicator").Value;
int[] scoreColumns = param.GetMultiChoiceParam("Scores").Value;
if (scoreColumns.Length == 0){
processInfo.ErrString = "Please specify at least one column with scores.";
return;
}
bool largeIsGood = param.GetBoolParam("Large values are good").Value;
int[] showColumns = param.GetMultiChoiceParam("Display quantity").Value;
if (showColumns.Length == 0){
processInfo.ErrString = "Please select at least one quantity to display";
return;
}
bool[] indCol = GetIndicatorColumn(falseAreIndicated, catCol, word, data);
List<string> expColNames = new List<string>();
List<float[]> expCols = new List<float[]>();
foreach (int scoreColumn in scoreColumns){
double[] vals = scoreColumn < data.NumericColumnCount
? data.NumericColumns[scoreColumn]
: ArrayUtils.ToDoubles(data.GetExpressionColumn(scoreColumn - data.NumericColumnCount));
string name = scoreColumn < data.NumericColumnCount
? data.NumericColumnNames[scoreColumn] : data.ExpressionColumnNames[scoreColumn - data.NumericColumnCount];
int[] order = GetOrder(vals, largeIsGood);
CalcCurve(ArrayUtils.SubArray(indCol, order), showColumns, name, expCols, expColNames);
}
float[,] expData = ToMatrix(expCols);
data.SetData(data.Name, expColNames, expData, new List<string>(), new List<string[]>(), new List<string>(),
new List<string[][]>(), new List<string>(), new List<double[]>(), new List<string>(), new List<double[][]>());
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables, ProcessInfo processInfo)
{
int nameCol = param.GetSingleChoiceParam("New column names").Value;
float[,] x = ArrayUtils.Transpose(mdata.ExpressionValues);
List<string> colNames = new List<string>(mdata.StringColumns[nameCol]);
List<string> rowNames = mdata.ExpressionColumnNames;
mdata.SetData(mdata.Name, colNames, x, new List<string>(new[]{"Name"}), new List<string[]>(new[]{rowNames.ToArray()}),
new List<string>(), new List<string[][]>(), new List<string>(), new List<double[]>(), new List<string>(),
new List<double[][]>());
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int nameCol = param.GetSingleChoiceParam("New column names").Value;
List<string> colNames;
if (nameCol >= 0){
HashSet<string> taken = new HashSet<string>();
colNames = new List<string>();
foreach (string n in mdata.StringColumns[nameCol]){
string n1 = GetNextAvailableName(n, taken);
taken.Add(n1);
colNames.Add(n1);
}
} else{
colNames = new List<string>();
for (int i = 0; i < mdata.RowCount; i++){
colNames.Add("Column" + (i + 1));
}
}
List<string> rowNames = mdata.ExpressionColumnNames;
mdata.SetData(mdata.Name, mdata.Description, colNames, colNames, ArrayUtils.Transpose(mdata.ExpressionValues),
ArrayUtils.Transpose(mdata.IsImputed), ArrayUtils.Transpose(mdata.QualityValues), mdata.QualityName,
mdata.QualityBiggerIsBetter, new List<string>(new[]{"Name"}), new List<string>(new[]{"Name"}),
new List<string[]>(new[]{rowNames.ToArray()}), mdata.CategoryRowNames, mdata.CategoryRowDescriptions,
GetCategoryRows(mdata), mdata.NumericRowNames, mdata.NumericRowDescriptions, mdata.NumericRows, new List<string>(),
new List<string>(), new List<double[][]>(), mdata.CategoryColumnNames, mdata.CategoryColumnDescriptions,
GetCategoryColumns(mdata), mdata.NumericColumnNames, mdata.NumericColumnDescriptions, mdata.NumericColumns);
}
private static void LoadData(IList<string> colNames, IList<string> colDescriptions, IList<int> expressionColIndices,
IList<int> catColIndices, IList<int> numColIndices, IList<int> textColIndices, IList<int> multiNumColIndices,
string filename, IMatrixData matrixData, IDictionary<string, string[]> annotationRows, Action<int> progress,
Action<string> status)
{
Dictionary<string, string[]> catAnnotatRows;
Dictionary<string, string[]> numAnnotatRows;
status("Reading data");
SplitAnnotRows(annotationRows, out catAnnotatRows, out numAnnotatRows);
int nrows = TabSep.GetRowCount(filename, 0, commentPrefix, commentPrefixExceptions);
float[,] expressionValues = new float[nrows,expressionColIndices.Count];
List<string[][]> categoryAnnotation = new List<string[][]>();
foreach (int t in catColIndices){
categoryAnnotation.Add(new string[nrows][]);
}
List<double[]> numericAnnotation = new List<double[]>();
foreach (int t in numColIndices){
numericAnnotation.Add(new double[nrows]);
}
List<double[][]> multiNumericAnnotation = new List<double[][]>();
foreach (int t in multiNumColIndices){
multiNumericAnnotation.Add(new double[nrows][]);
}
List<string[]> stringAnnotation = new List<string[]>();
foreach (int t in textColIndices){
stringAnnotation.Add(new string[nrows]);
}
StreamReader reader = new StreamReader(filename);
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 = line.Split('\t');
for (int i = 0; i < expressionColIndices.Count; i++){
if (expressionColIndices[i] >= w.Length){
expressionValues[count, i] = float.NaN;
} else{
string s = StringUtils.RemoveWhitespace(w[expressionColIndices[i]]);
bool success = float.TryParse(s, out expressionValues[count, i]);
if (!success){
expressionValues[count, i] = float.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(';');
Array.Sort(ww);
categoryAnnotation[i][count] = ww;
}
}
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);
numericAnnotation[i][count] = success ? q : double.NaN;
}
}
for (int i = 0; i < textColIndices.Count; i++){
if (textColIndices[i] >= w.Length){
stringAnnotation[i][count] = "";
} else{
string q = w[textColIndices[i]].Trim();
stringAnnotation[i][count] = RemoveSplitWhitespace(RemoveQuotes(q));
}
}
count++;
}
reader.Close();
string[] columnNames = ArrayUtils.SubArray(colNames, expressionColIndices);
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.SetData(filename, RemoveQuotes(columnNames), expressionValues, RemoveQuotes(textColnames),
stringAnnotation, RemoveQuotes(catColnames), categoryAnnotation, RemoveQuotes(numColnames), numericAnnotation,
RemoveQuotes(multiNumColnames), multiNumericAnnotation);
if (colDescriptions != null){
string[] columnDesc = ArrayUtils.SubArray(colDescriptions, expressionColIndices);
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.ExpressionColumnDescriptions = 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 ArrayUtils.GetKeys(catAnnotatRows)){
string name = key;
string[] svals = ArrayUtils.SubArray(catAnnotatRows[key], expressionColIndices);
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];
}
matrixData.AddCategoryRow(name, name, cat);
}
foreach (string key in ArrayUtils.GetKeys(numAnnotatRows)){
string name = key;
string[] svals = ArrayUtils.SubArray(numAnnotatRows[key], expressionColIndices);
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 = filename;
status("");
}
public void ProcessData(IMatrixData mdata, Parameters param1, ref IMatrixData[] supplTables, ProcessInfo processInfo)
{
int[] multiNumCols = param1.GetMultiChoiceParam("Multi-numeric columns").Value;
Array.Sort(multiNumCols);
int[] stringCols = param1.GetMultiChoiceParam("String columns").Value;
Array.Sort(stringCols);
HashSet<int> multinumCols2 = new HashSet<int>(multiNumCols);
HashSet<int> stringCols2 = new HashSet<int>(stringCols);
if (multiNumCols.Length + stringCols.Length == 0){
processInfo.ErrString = "Please select some columns.";
return;
}
int rowCount = GetNewRowCount(mdata, multiNumCols, stringCols);
float[,] expVals = new float[rowCount,mdata.ExpressionColumnCount];
List<string[]> stringC = new List<string[]>();
for (int i = 0; i < mdata.StringColumnCount; i++){
stringC.Add(new string[rowCount]);
}
List<double[]> numC = new List<double[]>();
for (int i = 0; i < mdata.NumericColumnCount; i++){
numC.Add(new double[rowCount]);
}
List<string[][]> catC = new List<string[][]>();
for (int i = 0; i < mdata.CategoryColumnCount; i++){
catC.Add(new string[rowCount][]);
}
List<double[][]> multiNumC = new List<double[][]>();
for (int i = 0; i < mdata.MultiNumericColumnCount; i++){
multiNumC.Add(new double[rowCount][]);
}
int count = 0;
for (int i = 0; i < mdata.RowCount; i++){
string err;
int entryCount = GetEntryCount(i, mdata, multiNumCols, stringCols, out err);
if (err != null){
processInfo.ErrString = err;
return;
}
bool empty = entryCount == 0;
entryCount = Math.Max(entryCount, 1);
for (int j = 0; j < entryCount; j++){
for (int k = 0; k < mdata.ExpressionColumnCount; k++){
expVals[count + j, k] = mdata[i, k];
}
for (int k = 0; k < mdata.NumericColumnCount; k++){
numC[k][count + j] = mdata.NumericColumns[k][i];
}
for (int k = 0; k < mdata.CategoryColumnCount; k++){
catC[k][count + j] = mdata.CategoryColumns[k][i];
}
}
for (int k = 0; k < mdata.MultiNumericColumnCount; k++){
if (multinumCols2.Contains(k)){
if (empty){
multiNumC[k][count] = new double[0];
} else{
double[] vals = mdata.MultiNumericColumns[k][i];
for (int j = 0; j < entryCount; j++){
multiNumC[k][count + j] = new[]{vals[j]};
}
}
} else{
for (int j = 0; j < entryCount; j++){
multiNumC[k][count + j] = mdata.MultiNumericColumns[k][i];
}
}
}
for (int k = 0; k < mdata.StringColumnCount; k++){
if (stringCols2.Contains(k)){
if (empty){
stringC[k][count] = "";
} else{
string[] vals = mdata.StringColumns[k][i].Split(';');
for (int j = 0; j < entryCount; j++){
stringC[k][count + j] = vals[j];
}
}
} else{
for (int j = 0; j < entryCount; j++){
stringC[k][count + j] = mdata.StringColumns[k][i];
}
}
}
count += entryCount;
}
int[] multiNumComplement = ArrayUtils.Complement(multiNumCols, mdata.MultiNumericColumnCount);
List<double[][]> toBeTransformed = ArrayUtils.SubList(multiNumC, multiNumCols);
multiNumC = ArrayUtils.SubList(multiNumC, multiNumComplement);
foreach (double[][] d in toBeTransformed){
numC.Add(Transform(d));
}
mdata.SetData(mdata.Name, mdata.ExpressionColumnNames, expVals, mdata.StringColumnNames, stringC,
mdata.CategoryColumnNames, catC,
new List<string>(ArrayUtils.Concat(mdata.NumericColumnNames,
ArrayUtils.SubList(mdata.MultiNumericColumnNames, multiNumCols))), numC,
new List<string>(ArrayUtils.SubArray(mdata.MultiNumericColumnNames, multiNumComplement)), multiNumC);
}
private static void CreateMatrixData(CountingResult result, IMatrixData data, int minCount, IEnumerable selection)
{
List<string[]> type = new List<string[]>();
List<string[]> name = new List<string[]>();
List<double> count = new List<double>();
List<double> percOfTotal = new List<double>();
List<double> selCount = new List<double>();
List<double> selPerc = new List<double>();
for (int i = 0; i < result.Count; i++){
int c = result.GetTotalCountAt(i);
if (c < minCount){
continue;
}
type.Add(new[]{result.GetType1At(i)});
name.Add(new[]{result.GetName1At(i)});
count.Add(c);
percOfTotal.Add(Math.Round(10000.0*c/data.RowCount)/100.0);
if (selection != null){
int c1 = result.GetSelectCountAt(i);
selCount.Add(c1);
selPerc.Add(Math.Round(1000.0*c1/c)/10.0);
}
}
float[,] ex = new float[type.Count,0];
List<string[][]> catCols = new List<string[][]>{type.ToArray(), name.ToArray()};
List<string> catColNames = new List<string>(new[]{"Type", "Name"});
List<double[]> numCols = new List<double[]>{count.ToArray(), percOfTotal.ToArray()};
if (selection != null){
numCols.Add(selCount.ToArray());
numCols.Add(selPerc.ToArray());
}
List<string> numColNames = new List<string>(new[]{"Count", "Percentage of total"});
if (selection != null){
numColNames.AddRange(new[]{"Selection count", "Selection percentage"});
}
data.SetData("Count", new List<string>(), ex, new List<string>(), new List<string[]>(), catColNames, catCols,
numColNames, numCols, new List<string>(), new List<double[][]>());
}
