public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
Random2 rand = new Random2();
double std = param.GetParam<double>("Standard deviation").Value;
int[] inds = param.GetParam<int[]>("Columns").Value;
List<int> mainInds = new List<int>();
List<int> numInds = new List<int>();
foreach (int ind in inds){
if (ind < mdata.ColumnCount){
mainInds.Add(ind);
} else{
numInds.Add(ind - mdata.ColumnCount);
}
}
foreach (int j in mainInds){
for (int i = 0; i < mdata.RowCount; i++){
mdata.Values.Set(i, j, mdata.Values.Get(i, j) + (float) rand.NextGaussian(0, std));
}
}
foreach (int j in numInds){
for (int i = 0; i < mdata.RowCount; i++){
mdata.NumericColumns[j][i] += (float) rand.NextGaussian(0, std);
}
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
Parameter<int> access = param.GetParam<int>("Matrix access");
bool rows = access.Value == 0;
UnitVectors(rows, mdata);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
const bool rows = false;
bool percentage;
int minValids = PerseusPluginUtils.GetMinValids(param, out percentage);
ParameterWithSubParams<int> modeParam = param.GetParamWithSubParams<int>("Mode");
int modeInd = modeParam.Value;
if (modeInd != 0 && mdata.CategoryRowNames.Count == 0){
processInfo.ErrString = "No grouping is defined.";
return;
}
if (modeInd != 0){
processInfo.ErrString = "Group-wise filtering can only be appled to rows.";
return;
}
FilteringMode filterMode;
double threshold;
double threshold2;
PerseusPluginUtils.ReadValuesShouldBeParams(param, out filterMode, out threshold, out threshold2);
if (modeInd != 0){
//TODO
} else{
PerseusPluginUtils.NonzeroFilter1(rows, minValids, percentage, mdata, param, threshold, threshold2, filterMode);
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
bool rows = param.GetParam<int>("Matrix access").Value == 0;
double min = param.GetParam<double>("Minimum").Value;
double max = param.GetParam<double>("Maximum").Value;
MapToInterval1(rows, mdata, min, max, processInfo.NumThreads);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string[] mods = param.GetParam<int[]>("Modifications").StringValue.Split(new[]{';'},
StringSplitOptions.RemoveEmptyEntries);
string[] up = mdata.StringColumns[param.GetParam<int>("Uniprot column").Value];
string[][] uprot = new string[up.Length][];
for (int i = 0; i < up.Length; i++){
uprot[i] = up[i].Length > 0 ? up[i].Split(';') : new string[0];
}
double[][] c = new double[mods.Length][];
for (int index = 0; index < mods.Length; index++){
string mod = mods[index];
string filename = PhosphoSitePlusParser.GetFilenameForMod(mod);
if (filename == null){
processInfo.ErrString = "File does not exist.";
return;
}
string[] seqWins;
string[] accs;
string[] pubmedLtp;
string[] pubmedMs2;
string[] cstMs2;
string[] species;
PhosphoSitePlusParser.ParseKnownMods(filename, out seqWins, out accs, out pubmedLtp, out pubmedMs2, out cstMs2, out species);
for (int i = 0; i < seqWins.Length; i++){
seqWins[i] = seqWins[i].ToUpper();
}
Dictionary<string, HashSet<string>> counts = new Dictionary<string, HashSet<string>>();
for (int i = 0; i < accs.Length; i++){
string acc = accs[i];
if (!counts.ContainsKey(acc)){
counts.Add(acc, new HashSet<string>());
}
counts[acc].Add(seqWins[i]);
}
c[index] = new double[up.Length];
for (int i = 0; i < up.Length; i++){
c[index][i] = CountSites(uprot[i], counts);
}
}
string[][] catCol = new string[up.Length][];
for (int i = 0; i < catCol.Length; i++){
List<string> x = new List<string>();
for (int j = 0; j < mods.Length; j++){
if (c[j][i] > 0){
x.Add(mods[j]);
}
}
x.Sort();
catCol[i] = x.ToArray();
}
mdata.AddCategoryColumn("Known modifications", "Known modifications", catCol);
for (int i = 0; i < mods.Length; i++){
mdata.AddNumericColumn(mods[i] + " count", mods[i] + " count", c[i]);
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
var vals = param.GetParam<Tuple<Regex, string>>("Regex").Value;
var pattern = vals.Item1;
string replacementStr = vals.Item2;
for (int i = 0; i < mdata.ColumnCount; i++){
mdata.ColumnNames[i] = pattern.Replace(mdata.ColumnNames[i], replacementStr);
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string[] seqWins;
string[] subAccs;
string[] kinases;
string[] kinAccs;
string[] species;
PhosphoSitePlusParser.ParseKinaseSubstrate(out seqWins, out subAccs, out kinases, out kinAccs, out species);
if (seqWins == null){
processInfo.ErrString = "File does not exist.";
return;
}
string[] up = mdata.StringColumns[param.GetParam<int>("Uniprot column").Value];
string[][] uprot = new string[up.Length][];
for (int i = 0; i < up.Length; i++){
uprot[i] = up[i].Length > 0 ? up[i].Split(';') : new string[0];
}
string[] win = mdata.StringColumns[param.GetParam<int>("Sequence window").Value];
Dictionary<string, List<Tuple<string, string, string>>> substrateProperties =
new Dictionary<string, List<Tuple<string, string, string>>>();
for (int i = 0; i < seqWins.Length; i++){
string subAcc = subAccs[i];
if (!substrateProperties.ContainsKey(subAcc)){
substrateProperties.Add(subAcc, new List<Tuple<string, string, string>>());
}
substrateProperties[subAcc].Add(new Tuple<string, string, string>(seqWins[i], kinases[i], kinAccs[i]));
}
string[] kinaseNameColumn = new string[uprot.Length];
string[] kinaseUniprotColumn = new string[uprot.Length];
for (int i = 0; i < kinaseNameColumn.Length; i++){
string[] win1 = AddKnownSites.TransformIl(win[i]).Split(';');
HashSet<string> kinaseNamesHits = new HashSet<string>();
HashSet<string> kinaseUniprotHits = new HashSet<string>();
foreach (string ux in uprot[i]){
if (substrateProperties.ContainsKey(ux)){
List<Tuple<string, string, string>> properties = substrateProperties[ux];
foreach (Tuple<string, string, string> property in properties){
string w = property.Item1;
if (AddKnownSites.Contains(win1, AddKnownSites.TransformIl(w.ToUpper().Substring(1, w.Length - 2)))){
kinaseNamesHits.Add(property.Item2);
kinaseUniprotHits.Add(property.Item3);
}
}
}
}
kinaseNameColumn[i] = kinaseNamesHits.Count > 0 ? StringUtils.Concat(";", ArrayUtils.ToArray(kinaseNamesHits)) : "";
kinaseUniprotColumn[i] = kinaseUniprotHits.Count > 0
? StringUtils.Concat(";", ArrayUtils.ToArray(kinaseUniprotHits))
: "";
}
mdata.AddStringColumn("PhosphoSitePlus kinase", "", kinaseNameColumn);
mdata.AddStringColumn("PhosphoSitePlus kinase uniprot", "", kinaseUniprotColumn);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string regexStr = param.GetParam<string>("Regular expression").Value;
Regex regex = new Regex(regexStr);
int[] inds = param.GetParam<int[]>("Columns").Value;
bool keepColumns = param.GetParam<bool>("Keep original columns").Value;
bool semicolons = param.GetParam<bool>("Strings separated by semicolons are independent").Value;
foreach (int col in inds){
ProcessCol(mdata, regex, col, keepColumns, semicolons);
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string colName = param.GetParam<string>("Name of new column").Value;
int[] columns = param.GetParam<int[]>("Categories").Value;
bool inverse = param.GetParam<bool>("Inverse").Value;
int[] catCols;
int[] stringCols;
Split(columns, out catCols, out stringCols, mdata.CategoryColumnCount);
string[] word1 = param.GetParam<string[]>("Search terms").Value;
if (word1.Length == 0){
processInfo.ErrString = "Please specify one or more search terms.";
return;
}
if (string.IsNullOrEmpty(colName)){
colName = word1[0];
}
string[] word = new string[word1.Length];
for (int i = 0; i < word.Length; i++){
word[i] = word1[i].ToLower().Trim();
}
bool[] indicator = new bool[mdata.RowCount];
foreach (int col in catCols){
for (int i = 0; i < mdata.RowCount; i++){
foreach (string s in mdata.GetCategoryColumnEntryAt(col, i)){
foreach (string s1 in word){
if (s.ToLower().Contains(s1)){
indicator[i] = true;
break;
}
}
}
}
}
foreach (string[] txt in stringCols.Select(col => mdata.StringColumns[col])){
for (int i = 0; i < txt.Length; i++){
string s = txt[i];
foreach (string s1 in word){
if (s.ToLower().Contains(s1)){
indicator[i] = true;
break;
}
}
}
}
string[][] newCol = new string[indicator.Length][];
for (int i = 0; i < newCol.Length; i++){
bool yes = inverse ? !indicator[i] : indicator[i];
newCol[i] = yes ? new[]{"+"} : new string[0];
}
mdata.AddCategoryColumn(colName, "", newCol);
}
public IMatrixData ProcessData(IMatrixData[] inputData, Parameters parameters, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
IMatrixData mdata1 = inputData[0];
Dictionary<string, string> map = GetMap(inputData[1], parameters);
IMatrixData result = (IMatrixData) mdata1.Clone();
int ind = parameters.GetParam<int>("Column in matrix 1 to be edited").Value;
string[] x = mdata1.StringColumns[ind];
for (int i = 0; i < x.Length; i++){
x[i] = Process(x[i], map);
}
return result;
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int nameCol = param.GetParam<int>("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 = StringUtils.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.ColumnNames;
mdata.Values = mdata.Values.Transpose();
if (mdata.IsImputed != null){
mdata.IsImputed = mdata.IsImputed.Transpose();
}
if (mdata.Quality != null){
mdata.Quality = mdata.Quality.Transpose();
}
List<string> stringColumnNames = mdata.StringColumnNames;
List<string> categoryColumnNames = mdata.CategoryColumnNames;
List<string> numericColumnNames = mdata.NumericColumnNames;
List<string> multiNumericColumnNames = mdata.MultiNumericColumnNames;
List<string> stringColumnDescriptions = mdata.StringColumnDescriptions;
List<string> categoryColumnDescriptions = mdata.CategoryColumnDescriptions;
List<string> numericColumnDescriptions = mdata.NumericColumnDescriptions;
List<string> multiNumericColumnDescriptions = mdata.MultiNumericColumnDescriptions;
List<string[]> stringColumns = mdata.StringColumns;
List<string[][]> categoryColumns = GetCategoryColumns(mdata);
List<double[]> numericColumns = mdata.NumericColumns;
List<double[][]> multiNumericColumns = mdata.MultiNumericColumns;
mdata.SetAnnotationColumns(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.ColumnNames = colNames;
mdata.SetAnnotationRows(stringColumnNames, stringColumnDescriptions, stringColumns, categoryColumnNames,
categoryColumnDescriptions, categoryColumns, numericColumnNames, numericColumnDescriptions, numericColumns,
multiNumericColumnNames, multiNumericColumnDescriptions, multiNumericColumns);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string word = param.GetParam<string>("Find what").Value;
int colInd = param.GetParam<int>("Look in").Value;
bool matchCase = param.GetParam<bool>("Match case").Value;
bool matchWholeWord = param.GetParam<bool>("Match whole word").Value;
string scolName = mdata.StringColumnNames[colInd];
string[] scol = mdata.StringColumns[colInd];
string[][] catCol = new string[mdata.RowCount][];
for (int i = 0; i < catCol.Length; i++){
bool found = Find(scol[i], word, matchCase, matchWholeWord);
catCol[i] = found ? new[]{"+"} : new string[0];
}
mdata.AddCategoryColumn("Search: " + scolName, "Search: " + scolName, catCol);
}
public void ProcessData(IMatrixData data, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int[] numColInds = param.GetParam<int[]>("Numerical rows").Value;
int[] multiNumColInds = param.GetParam<int[]>("Multi-numerical rows").Value;
int[] catColInds = param.GetParam<int[]>("Categorical rows").Value;
int[] textColInds = param.GetParam<int[]>("Text rows").Value;
data.NumericRows = ArrayUtils.SubList(data.NumericRows, numColInds);
data.NumericRowNames = ArrayUtils.SubList(data.NumericRowNames, numColInds);
data.NumericRowDescriptions = ArrayUtils.SubList(data.NumericRowDescriptions, numColInds);
data.MultiNumericRows = ArrayUtils.SubList(data.MultiNumericRows, multiNumColInds);
data.MultiNumericRowNames = ArrayUtils.SubList(data.MultiNumericRowNames, multiNumColInds);
data.MultiNumericRowDescriptions = ArrayUtils.SubList(data.MultiNumericRowDescriptions, multiNumColInds);
data.CategoryRows = PerseusPluginUtils.GetCategoryRows(data, catColInds);
data.CategoryRowNames = ArrayUtils.SubList(data.CategoryRowNames, catColInds);
data.CategoryRowDescriptions = ArrayUtils.SubList(data.CategoryRowDescriptions, catColInds);
data.StringRows = ArrayUtils.SubList(data.StringRows, textColInds);
data.StringRowNames = ArrayUtils.SubList(data.StringRowNames, textColInds);
data.StringRowDescriptions = ArrayUtils.SubList(data.StringRowDescriptions, textColInds);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int stringColumnIndx = param.GetParam<int>("Sequence window").Value;
string[] win = mdata.StringColumns[stringColumnIndx];
int start = param.GetParam<int>("Start").Value - 1;
int length = param.GetParam<int>("Length").Value;
if (start < 0){
processInfo.ErrString = "Start position cannot be smaller than 1.";
return;
}
if (start + length > win[0].Length){
processInfo.ErrString = "Start + length cannot exceed the total length of the sequence.";
return;
}
string[] shortenedMotifs = new string[win.Length];
for (int i = 0; i < mdata.RowCount; ++i){
shortenedMotifs[i] = win[i].Substring(start, length);
}
mdata.AddStringColumn("Short sequence window", "", shortenedMotifs);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
ParameterWithSubParams<int> p = param.GetParamWithSubParams<int>("Column");
int colInd = p.Value;
if (colInd < 0){
processInfo.ErrString = "No categorical columns available.";
return;
}
Parameter<int[]> mcp = p.GetSubParameters().GetParam<int[]>("Values");
int[] inds = mcp.Value;
if (inds.Length == 0){
processInfo.ErrString = "Please select at least one term for filtering.";
return;
}
string[] values = new string[inds.Length];
string[] v = mdata.GetCategoryColumnValuesAt(colInd);
for (int i = 0; i < values.Length; i++){
values[i] = v[inds[i]];
}
HashSet<string> value = new HashSet<string>(values);
bool remove = param.GetParam<int>("Mode").Value == 0;
List<int> valids = new List<int>();
for (int i = 0; i < mdata.RowCount; i++){
bool valid = true;
foreach (string w in mdata.GetCategoryColumnEntryAt(colInd, i)){
if (value.Contains(w)){
valid = false;
break;
}
}
if ((valid && remove) || (!valid && !remove)){
valids.Add(i);
}
}
PerseusPluginUtils.FilterRows(mdata, param, valids.ToArray());
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string[][] col = mdata.GetCategoryColumnAt(param.GetParam<int>("Indicator column").Value);
string term = param.GetParam<string>("Value").Value;
List<int> inds = new List<int>();
for (int i = 0; i < col.Length; i++){
if (Contains(col[i], term)){
inds.Add(i);
}
}
double[][] profiles = new double[inds.Count][];
for (int i = 0; i < profiles.Length; i++){
profiles[i] = ArrayUtils.ToDoubles(mdata.Values.GetRow(inds[i]));
float mean = (float) ArrayUtils.Mean(profiles[i]);
for (int j = 0; j < profiles[i].Length; j++){
profiles[i][j] -= mean;
}
}
double[] totalProfile = new double[mdata.ColumnCount];
for (int i = 0; i < totalProfile.Length; i++){
List<double> vals = new List<double>();
foreach (double[] t in profiles){
double val = t[i];
if (double.IsNaN(val) || double.IsInfinity(val)){
continue;
}
vals.Add(val);
}
totalProfile[i] = vals.Count > 0 ? ArrayUtils.Median(vals) : double.NaN;
}
for (int i = 0; i < mdata.RowCount; i++){
for (int j = 0; j < mdata.ColumnCount; j++){
mdata.Values.Set(i, j, mdata.Values.Get(i, j)-(float) totalProfile[j]);
}
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
const bool rows = true;
bool percentage;
int minValids = PerseusPluginUtils.GetMinValids(param, out percentage);
ParameterWithSubParams<int> modeParam = param.GetParamWithSubParams<int>("Mode");
int modeInd = modeParam.Value;
if (modeInd != 0 && mdata.CategoryRowNames.Count == 0){
processInfo.ErrString = "No grouping is defined.";
return;
}
FilteringMode filterMode;
double threshold;
double threshold2;
PerseusPluginUtils.ReadValuesShouldBeParams(param, out filterMode, out threshold, out threshold2);
if (modeInd != 0){
int gind = modeParam.GetSubParameters().GetParam<int>("Grouping").Value;
string[][] groupCol = mdata.GetCategoryRowAt(gind);
NonzeroFilterGroup(minValids, percentage, mdata, param, modeInd == 2, threshold, threshold2, filterMode, groupCol);
} else{
PerseusPluginUtils.NonzeroFilter1(rows, minValids, percentage, mdata, param, threshold, threshold2, filterMode);
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int colInd = param.GetParam<int>("Column").Value;
string searchString = param.GetParam<string>("Search string").Value;
if (string.IsNullOrEmpty(searchString)){
processInfo.ErrString = "Please provide a search string";
return;
}
bool remove = param.GetParam<int>("Mode").Value == 0;
bool matchCase = param.GetParam<bool>("Match case").Value;
bool matchWholeWord = param.GetParam<bool>("Match whole word").Value;
string[] vals = mdata.StringColumns[colInd];
List<int> valids = new List<int>();
for (int i = 0; i < vals.Length; i++){
bool matches = Matches(vals[i], searchString, matchCase, matchWholeWord);
if (matches && !remove){
valids.Add(i);
} else if (!matches && remove){
valids.Add(i);
}
}
PerseusPluginUtils.FilterRows(mdata, param, valids.ToArray());
}
public IAnalysisResult AnalyzeData(IMatrixData mdata, Parameters param, ProcessInfo processInfo)
{
return new SelectRowsManuallyResult(mdata);
}
public static void ReadMatrixFromFile(IMatrixData mdata, ProcessInfo processInfo, string filename, char separator)
{
var annotationRows = new Dictionary<string, string[]>();
var colNames = TabSep.GetColumnNames(filename, commentPrefix,
commentPrefixExceptions, annotationRows, separator);
var typeRow = annotationRows["Type"];
int[] eInds, nInds, cInds, tInds, mInds;
ColumnIndices(typeRow, out eInds, out nInds, out cInds, out tInds, out mInds);
var filters = new List<Tuple<Relation[], int[], bool>>();
int nrows;
using(StreamReader reader = FileUtils.GetReader(filename))
using (StreamReader auxReader = FileUtils.GetReader(filename))
{
nrows = GetRowCount(reader, auxReader, eInds, filters, separator);
}
using (StreamReader reader = FileUtils.GetReader(filename))
using (StreamReader auxReader = FileUtils.GetReader(filename))
{
LoadMatrixData(annotationRows, eInds, cInds, nInds, tInds, mInds, processInfo, colNames, mdata, reader,
auxReader, nrows, filename, separator, false, filters);
}
}
public static void ReadMatrixFromFile(IMatrixData mdata, ProcessInfo processInfo, string filename,
int[] eInds, int[] nInds, int[] cInds, int[] tInds, int[] mInds,
Parameters[] mainFilterParameters, Parameters[] numericalFilterParameters, bool shortenExpressionColumnNames)
{
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, commentPrefix, commentPrefixExceptions,
annotationRows, separator);
}
catch (Exception)
{
processInfo.ErrString = "Could not open the file '" + filename + "'. It is probably opened in another program.";
return;
}
string origin = filename;
List<Tuple<Relation[], int[], bool>> filters = new List<Tuple<Relation[], int[], bool>>();
string errString;
foreach (Parameters p in mainFilterParameters)
{
AddFilter(filters, p, eInds, out errString);
if (errString != null)
{
processInfo.ErrString = errString;
return;
}
}
foreach (Parameters p in numericalFilterParameters)
{
AddFilter(filters, p, nInds, out errString);
if (errString != null)
{
processInfo.ErrString = errString;
return;
}
}
int nrows;
using(StreamReader reader = FileUtils.GetReader(filename))
using (StreamReader auxReader = FileUtils.GetReader(filename))
{
nrows = GetRowCount(reader, auxReader, eInds, filters, separator);
}
using (StreamReader reader = FileUtils.GetReader(filename))
using (StreamReader auxReader = FileUtils.GetReader(filename))
{
LoadMatrixData(annotationRows, eInds, cInds, nInds, tInds, mInds, processInfo, colNames, mdata, reader,
auxReader, nrows, origin, separator, shortenExpressionColumnNames, filters);
}
GC.Collect();
}
public static void LoadMatrixData(IDictionary<string, string[]> annotationRows, int[] eInds, int[] cInds, int[] nInds,
int[] tInds, int[] mInds, ProcessInfo processInfo, IList<string> colNames, IMatrixData mdata, StreamReader reader,
StreamReader auxReader, int nrows, string origin, char separator, bool shortenExpressionNames,
List<Tuple<Relation[], int[], bool>> filters)
{
string[] colDescriptions = null;
if (annotationRows.ContainsKey("Description")){
colDescriptions = annotationRows["Description"];
annotationRows.Remove("Description");
}
int[] allInds = ArrayUtils.Concat(new[]{eInds, cInds, nInds, tInds, mInds});
Array.Sort(allInds);
for (int i = 0; i < allInds.Length - 1; i++){
if (allInds[i + 1] == allInds[i]){
processInfo.ErrString = "Column '" + colNames[allInds[i]] + "' has been selected multiple times";
return;
}
}
string[] allColNames = ArrayUtils.SubArray(colNames, allInds);
Array.Sort(allColNames);
for (int i = 0; i < allColNames.Length - 1; i++){
if (allColNames[i + 1].Equals(allColNames[i])){
processInfo.ErrString = "Column name '" + allColNames[i] + "' occurs multiple times.";
return;
}
}
LoadMatrixData(colNames, colDescriptions, eInds, cInds, nInds, tInds, mInds, origin, mdata, annotationRows,
processInfo.Progress, processInfo.Status, separator, reader, auxReader, nrows, shortenExpressionNames, filters);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
bool keepEmpty = param.GetBoolParam("Keep rows without ID").Value;
AverageType atype = GetAverageType(param.GetSingleChoiceParam("Average type for expression columns").Value);
string[] ids2 = mdata.StringColumns[param.GetSingleChoiceParam("ID column").Value];
string[][] ids = SplitIds(ids2);
int[] present;
int[] absent;
GetPresentAbsentIndices(ids, out present, out absent);
ids = ArrayUtils.SubArray(ids, present);
int[][] rowInds = new int[present.Length][];
for (int i = 0; i < rowInds.Length; i++){
rowInds[i] = new[]{present[i]};
}
ClusterRows(ref rowInds, ref ids);
if (keepEmpty){
rowInds = ProlongRowInds(rowInds, absent);
}
int nrows = rowInds.Length;
int ncols = mdata.ExpressionColumnCount;
float[,] expVals = new float[nrows,ncols];
for (int j = 0; j < ncols; j++){
float[] c = mdata.GetExpressionColumn(j);
for (int i = 0; i < nrows; i++){
float[] d = ArrayUtils.SubArray(c, rowInds[i]);
expVals[i, j] = Average(d, atype);
}
}
mdata.ExpressionValues = expVals;
for (int i = 0; i < mdata.NumericColumnCount; i++){
string name = mdata.NumericColumnNames[i];
AverageType atype1 = GetAverageType(param.GetSingleChoiceParam("Average type for " + name).Value);
double[] c = mdata.NumericColumns[i];
double[] newCol = new double[nrows];
for (int k = 0; k < nrows; k++){
double[] d = ArrayUtils.SubArray(c, rowInds[k]);
newCol[k] = Average(d, atype1);
}
mdata.NumericColumns[i] = newCol;
}
for (int i = 0; i < mdata.CategoryColumnCount; i++){
string[][] c = mdata.GetCategoryColumnAt(i);
string[][] newCol = new string[nrows][];
for (int k = 0; k < nrows; k++){
string[][] d = ArrayUtils.SubArray(c, rowInds[k]);
newCol[k] = Average(d);
}
mdata.SetCategoryColumnAt(newCol,i);
}
for (int i = 0; i < mdata.StringColumnCount; i++){
string[] c = mdata.StringColumns[i];
string[] newCol = new string[nrows];
for (int k = 0; k < nrows; k++){
string[] d = ArrayUtils.SubArray(c, rowInds[k]);
newCol[k] = Average(d);
}
mdata.StringColumns[i] = newCol;
}
for (int i = 0; i < mdata.MultiNumericColumnCount; i++){
double[][] c = mdata.MultiNumericColumns[i];
double[][] newCol = new double[nrows][];
for (int k = 0; k < nrows; k++){
double[][] d = ArrayUtils.SubArray(c, rowInds[k]);
newCol[k] = Average(d);
}
mdata.MultiNumericColumns[i] = newCol;
}
}
public void ProcessData(IMatrixData mdata, Parameters param1, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int[] multiNumCols = param1.GetParam<int[]>("Multi-numeric columns").Value;
Array.Sort(multiNumCols);
int[] stringCols = param1.GetParam<int[]>("Text 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.ColumnCount];
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.ColumnCount; k++){
expVals[count + j, k] = mdata.Values.Get(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.GetCategoryColumnEntryAt(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.ColumnNames = mdata.ColumnNames;
mdata.Values.Set(expVals);
mdata.SetAnnotationColumns(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);
}
public void ParseFile(string path, ProcessInfo processInfo)
{
processInfo.Status("Parsing " + path);
string accession = "";
int sequenceCounter = 0;
StringBuilder sequence = new StringBuilder();
ProteinSequence protein = new ProteinSequence();
try{
StreamReader file = new StreamReader(path);
string line;
while ((line = file.ReadLine()) != null){ // valid line
if (sequenceCounter%500 == 0){
processInfo.Status("Parsing " + path + ", " + (int) ((float) file.BaseStream.Position/file.BaseStream.Length*100) +
"%");
}
bool lineIsHeader = line.StartsWith(">");
// skip all lines until the first header is found
if (sequenceCounter == 0 && !lineIsHeader){
continue;
}
// line is a piece of a sequence
if (sequenceCounter > 0 && !lineIsHeader){
sequence.Append(line.Trim());
continue;
}
// line is a fasta header
if (lineIsHeader){
if (sequenceCounter > 0)
// this is not the first header, i.e. the previous sequence is now completely read in
{
// add the previous protein
protein.SetSequence(sequence.ToString());
entries.Add(accession, protein);
}
// initialize a new protein
protein = new ProteinSequence();
sequenceCounter++;
// then parse the new header
string header = line;
Match m = regexUniprotAccession.Match(header);
if (m.Success){ // uniprot header
accession = m.Groups[1].Value;
protein.Accession = accession;
protein.Header = header;
} else{ // fallback position: take entire header after the > as accession
accession = header.Substring(1).Trim();
protein.Accession = accession;
protein.Header = header;
}
sequence = new StringBuilder();
}
} //end while
file.Close();
//add the last protein
if (sequenceCounter > 0){ // make sure there is at least one sequence in the file
protein.SetSequence(sequence.ToString());
entries.Add(accession, protein);
}
} catch (Exception){
processInfo.ErrString =
"Something went wrong while parsing the fasta file.\nMake sure the path is correct and the " +
"file is not opened in another application.\nMake sure the fasta file is valid.";
}
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
double width = param.GetParam<double>("Width").Value;
double shift = param.GetParam<double>("Down shift").Value;
bool separateColumns = param.GetParam<int>("Mode").Value == 1;
int[] cols = param.GetParam<int[]>("Columns").Value;
if (cols.Length == 0){
return;
}
if (separateColumns){
ReplaceMissingsByGaussianByColumn(width, shift, mdata, cols);
} else{
string err = ReplaceMissingsByGaussianWholeMatrix(width, shift, mdata, cols);
if (err != null){
processInfo.ErrString = err;
}
}
}
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,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
string[] seqWins;
string[] accs;
string[] function;
string[] process;
string[] protInteract;
string[] otherInteract;
string[] notes;
string[] species;
PhosphoSitePlusParser.ParseRegulatorySites(out seqWins, out accs, out function, out process, out protInteract,
out otherInteract, out notes, out species);
if (seqWins == null){
processInfo.ErrString = "File does not exist.";
return;
}
string[] up = mdata.StringColumns[param.GetParam<int>("Uniprot column").Value];
string[][] uprot = new string[up.Length][];
for (int i = 0; i < up.Length; i++){
uprot[i] = up[i].Length > 0 ? up[i].Split(';') : new string[0];
}
string[] win = mdata.StringColumns[param.GetParam<int>("Sequence window").Value];
Dictionary<string, List<int>> map = new Dictionary<string, List<int>>();
for (int i = 0; i < seqWins.Length; i++){
string acc = accs[i];
if (!map.ContainsKey(acc)){
map.Add(acc, new List<int>());
}
map[acc].Add(i);
}
string[][] newCatCol = new string[uprot.Length][];
string[][] function2 = new string[uprot.Length][];
string[][] process2 = new string[uprot.Length][];
string[][] protInteract2 = new string[uprot.Length][];
string[][] otherInteract2 = new string[uprot.Length][];
string[][] notes2 = new string[uprot.Length][];
for (int i = 0; i < uprot.Length; i++){
string[] win1 = TransformIl(win[i]).Split(';');
HashSet<string> wins = new HashSet<string>();
HashSet<string> function1 = new HashSet<string>();
HashSet<string> process1 = new HashSet<string>();
HashSet<string> protInteract1 = new HashSet<string>();
HashSet<string> otherInteract1 = new HashSet<string>();
HashSet<string> notes1 = new HashSet<string>();
foreach (string ux in uprot[i]){
if (map.ContainsKey(ux)){
List<int> n = map[ux];
foreach (int ind in n){
string s = seqWins[ind];
if (Contains(win1, TransformIl(s.ToUpper().Substring(1, s.Length - 2)))){
wins.Add(s);
if (function[ind].Length > 0){
function1.Add(function[ind]);
}
if (process[ind].Length > 0){
process1.Add(process[ind]);
}
if (protInteract[ind].Length > 0){
protInteract1.Add(protInteract[ind]);
}
if (otherInteract[ind].Length > 0){
otherInteract1.Add(otherInteract[ind]);
}
if (notes[ind].Length > 0){
notes1.Add(notes[ind]);
}
}
}
}
}
if (wins.Count > 0){
newCatCol[i] = new[]{"+"};
function2[i] = ArrayUtils.ToArray(function1);
process2[i] = ArrayUtils.ToArray(process1);
protInteract2[i] = ArrayUtils.ToArray(protInteract1);
otherInteract2[i] = ArrayUtils.ToArray(otherInteract1);
notes2[i] = ArrayUtils.ToArray(notes1);
} else{
newCatCol[i] = new string[0];
function2[i] = new string[0];
process2[i] = new string[0];
protInteract2[i] = new string[0];
otherInteract2[i] = new string[0];
notes2[i] = new string[0];
}
}
mdata.AddCategoryColumn("Regulatory site", "", newCatCol);
mdata.AddCategoryColumn("Regulatory site function", "", function2);
mdata.AddCategoryColumn("Regulatory site process", "", process2);
mdata.AddCategoryColumn("Regulatory site protInteract", "", protInteract2);
mdata.AddCategoryColumn("Regulatory site otherInteract", "", otherInteract2);
mdata.AddCategoryColumn("Regulatory site notes", "", notes2);
}
public void ProcessData(IMatrixData mdata, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo)
{
int[] colIndx = param.GetParam<int[]>("x").Value;
int[] colIndy = param.GetParam<int[]>("y").Value;
if (colIndx.Length == 0){
processInfo.ErrString = "Please select some columns";
return;
}
if (colIndx.Length != colIndy.Length){
processInfo.ErrString = "Please select the same number of columns in the boxes for the first and second columns.";
return;
}
int typeInd = param.GetParam<int>("Distribution type").Value;
int points = param.GetParam<int>("Number of points").Value;
for (int k = 0; k < colIndx.Length; k++){
float[] xvals = GetColumn(mdata, colIndx[k]);
float[] yvals = GetColumn(mdata, colIndy[k]);
float[] xvals1;
float[] yvals1;
GetValidPairs(xvals, yvals, out xvals1, out yvals1);
double xmin;
double xmax;
double ymin;
double ymax;
DensityEstimation.CalcRanges(xvals1, yvals1, out xmin, out xmax, out ymin, out ymax);
float[,] values = DensityEstimation.GetValuesOnGrid(xvals1, xmin, (xmax - xmin)/points, points, yvals1, ymin,
(ymax - ymin)/points, points);
if (typeInd == 1){
MakeConditional1(values);
}
if (typeInd == 2){
MakeConditional2(values);
}
if (typeInd == 3){
MakeConditional3(values);
}
DensityEstimation.DivideByMaximum(values);
double[] xmat = new double[points];
for (int i = 0; i < points; i++){
xmat[i] = xmin + i*(xmax - xmin)/points;
}
double[] ymat = new double[points];
for (int i = 0; i < points; i++){
ymat[i] = ymin + i*(ymax - ymin)/points;
}
float[,] percvalues = CalcExcludedPercentage(values);
double[] dvals = new double[xvals.Length];
double[] pvals = new double[xvals.Length];
for (int i = 0; i < dvals.Length; i++){
double xx = xvals[i];
double yy = yvals[i];
if (!double.IsNaN(xx) && !double.IsNaN(yy)){
int xind = ArrayUtils.ClosestIndex(xmat, xx);
int yind = ArrayUtils.ClosestIndex(ymat, yy);
dvals[i] = values[xind, yind];
pvals[i] = percvalues[xind, yind];
} else{
dvals[i] = double.NaN;
pvals[i] = double.NaN;
}
}
string xname = GetColumnName(mdata, colIndx[k]);
string yname = GetColumnName(mdata, colIndy[k]);
mdata.AddNumericColumn("Density_" + xname + "_" + yname,
"Density of data points in the plane spanned by the columns " + xname + " and " + yname + ".", dvals);
mdata.AddNumericColumn("Excluded fraction_" + xname + "_" + yname,
"Percentage of points with a point density smaller than at this point in the plane spanned by the columns " + xname +
" and " + yname + ".", pvals);
}
}
public abstract IMatrixData ProcessData(IMatrixData[] inputData, Parameters param, ref IMatrixData[] supplTables,
ref IDocumentData[] documents, ProcessInfo processInfo);
