/// <summary>
/// This method removes sequences not having the required number of chains.
/// </summary>
/// <returns></returns>
public static List <ISequence> RemoveSequencesWithIncorrectNumberOfChains(CancellationToken cancellationToken, List <ISequence> sequenceList, int numberOfChainsRequired = 2, ProgressActionSet progressActionSet = null)
{
if (sequenceList == null || sequenceList.Count == 0)
{
throw new ArgumentOutOfRangeException(nameof(sequenceList));
}
if (progressActionSet == null)
{
throw new ArgumentNullException(nameof(progressActionSet));
}
var pdbIdListNotDistinct = FilterProteins.SequenceListToPdbIdList(sequenceList, false);
ProgressActionSet.StartAction(pdbIdListNotDistinct.Count, progressActionSet);
var workDivision = new WorkDivision <List <string> >(pdbIdListNotDistinct.Count);
for (var threadIndex = 0; threadIndex < workDivision.ThreadCount; threadIndex++)
{
var localThreadIndex = threadIndex;
Task <List <string> > task = Task.Run(() =>
{
var taskResult = pdbIdListNotDistinct.Where((a, pdbIdIndex) =>
{
if (pdbIdIndex < workDivision.ThreadFirstIndex[localThreadIndex] || pdbIdIndex > workDivision.ThreadLastIndex[localThreadIndex])
{
return(false);
}
workDivision.IncrementItemsCompleted(1);
ProgressActionSet.ProgressAction(1, progressActionSet);
ProgressActionSet.EstimatedTimeRemainingAction(workDivision.StartTicks, workDivision.ItemsCompleted, workDivision.ItemsToProcess, progressActionSet);
return(pdbIdListNotDistinct.Count(b => a == b) != numberOfChainsRequired);
}).ToList();
return(taskResult);
}, cancellationToken);
workDivision.TaskList.Add(task);
}
workDivision.WaitAllTasks();
var sequencesWithIncorrectNumberOfChains = new List <string>();
foreach (var task in workDivision.TaskList.Where(t => t != null && t.Result != null && t.IsCompleted && !t.IsCanceled && !t.IsFaulted))
{
sequencesWithIncorrectNumberOfChains.AddRange(task.Result);
}
var result = RemoveSequences(cancellationToken, sequenceList, sequencesWithIncorrectNumberOfChains);
ProgressActionSet.FinishAction(true, progressActionSet);
return(result);
}
public static List <string> RemoveStructuresWithIncorrectNumberOfChains(CancellationToken cancellationToken, string[] pdbFolders, List <string> pdbIdList = null, Dictionary <string, List <string> > pdbIdChainIdList = null, int numberChainsRequired = 2, ProgressActionSet progressActionSet = null, int totalThreads = -1)
{
if (pdbFolders == null || pdbFolders.Length == 0)
{
throw new ArgumentOutOfRangeException(nameof(pdbFolders));
}
if (pdbIdList == null || pdbIdList.Count == 0)
{
throw new ArgumentOutOfRangeException(nameof(pdbIdList));
}
if (progressActionSet == null)
{
throw new ArgumentNullException(nameof(progressActionSet));
}
var pdbFilesArray = ProteinDataBankFileOperations.GetPdbFilesArray(pdbFolders);
pdbFilesArray = ProteinDataBankFileOperations.RemoveNonWhiteListedPdbIdFromPdbFilesArray(pdbIdList, pdbFilesArray);
ProteinDataBankFileOperations.ShowMissingPdbFiles(pdbFilesArray, pdbIdList, progressActionSet);
WorkDivision <List <string> > workDivision = new WorkDivision <List <string> >(pdbFilesArray.Length, totalThreads);
ProgressActionSet.StartAction(pdbFilesArray.Length, progressActionSet);
for (int threadIndex = 0; threadIndex < workDivision.ThreadCount; threadIndex++)
{
int localThreadIndex = threadIndex;
Task <List <string> > task = Task.Run(() =>
{
var taskResult = new List <string>();
for (int pdbFileNumber = workDivision.ThreadFirstIndex[localThreadIndex]; pdbFileNumber <= workDivision.ThreadLastIndex[localThreadIndex]; pdbFileNumber++)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
try
{
string pdbFilename = pdbFilesArray[pdbFileNumber];
string proteinId = ProteinDataBankFileOperations.PdbIdFromPdbFilename(pdbFilename);
// Check if the file found is included in the white list.
if (pdbIdList != null && !pdbIdList.Contains(proteinId))
{
continue;
}
var sequenceChainIdList = pdbIdChainIdList != null ? (pdbIdChainIdList.ContainsKey(proteinId) ? pdbIdChainIdList[proteinId].ToArray() : null) : null;
int chainCount = ProteinDataBankFileOperations.PdbAtomicChainsCount(pdbFilename, sequenceChainIdList, numberChainsRequired);
if (chainCount != numberChainsRequired)
{
if (!taskResult.Contains(proteinId))
{
taskResult.Add(proteinId);
}
}
}
finally
{
workDivision.IncrementItemsCompleted(1);
ProgressActionSet.ProgressAction(1, progressActionSet);
ProgressActionSet.EstimatedTimeRemainingAction(workDivision.StartTicks, workDivision.ItemsCompleted, workDivision.ItemsToProcess, progressActionSet);
}
}
return(taskResult);
}, cancellationToken);
workDivision.TaskList.Add(task);
}
workDivision.WaitAllTasks();
ProgressActionSet.FinishAction(true, progressActionSet);
var result = new List <string>();
foreach (var task in workDivision.TaskList.Where(t => t != null && t.Result != null && t.IsCompleted && !t.IsCanceled && !t.IsFaulted))
{
result.AddRange(task.Result);
}
result = result.Distinct().ToList();
return(result);
}
public static string[] MakeInteractionsOutput(CancellationToken cancellationToken, decimal maxAtomInterationDistance, string[] pdbFilesList, Dictionary <string, List <string> > pdbIdChainIdList, ProgressActionSet progressActionSet, bool outputToGui)
{
var interactionTasks = new List <Task <string> >();
ProgressActionSet.StartAction(pdbFilesList.Length, progressActionSet);
foreach (string pdbFilename in pdbFilesList)
{
string _pdbFilename = pdbFilename;
while (interactionTasks.Count(t => t != null && !t.IsCompleted) >= Environment.ProcessorCount * 10)
{
Task.WaitAny(interactionTasks.ToArray <Task>());
}
var interactionTask = Task.Run(() =>
{
if (cancellationToken.IsCancellationRequested)
{
return(null);
}
List <AtomPair> interactionsList = SearchInteractions.FindInteractions(cancellationToken, maxAtomInterationDistance, _pdbFilename, pdbIdChainIdList);
string pdbId = ProteinDataBankFileOperations.PdbIdFromPdbFilename(_pdbFilename);
if (string.IsNullOrEmpty(pdbId))
{
//pdbId = _pdbFilename;
throw new ArgumentNullException(nameof(pdbFilesList), "The file " + _pdbFilename + " has an invalid name.");
}
if (interactionsList == null)
{
interactionsList = new List <AtomPair>();
}
interactionsList = interactionsList.OrderBy(o => o.Distance).ToList();
var interactionsString = FormatInteractionOutput(pdbId, interactionsList);
if (outputToGui)
{
ProgressActionSet.Report(interactionsString, progressActionSet);
}
ProgressActionSet.ProgressAction(1, progressActionSet);
//ProgressActionSet.EstimatedTimeRemainingAction(startTicks, );
//////Console.WriteLine(_pdbFilename);
return(interactionsString);
}, cancellationToken);
interactionTasks.Add(interactionTask);
}
Task.WaitAll(interactionTasks.Where(t => t != null && !t.IsCompleted).ToArray <Task>());
var interactionsStringsList = interactionTasks.OrderBy(t => t.Id).Where(t => t != null && t.IsCompleted && !t.IsCanceled && !t.IsFaulted && t.Result != null).Select(t => t.Result).ToArray();
ProgressActionSet.FinishAction(true, progressActionSet);
return(interactionsStringsList);
}
/// <summary>
/// This method iterates through the provided FASTA files creating separate calculated outputs for each of them.
/// </summary>
/// <param name="fastaFiles">The FASTA files to process.</param>
/// <param name="pdbFilesFolders">The locations where PDB files may be found.</param>
/// <param name="spreadsheetSaveFilenameTemplate">A template filename to save the outputs.</param>
/// <param name="saveTsv"></param>
/// <param name="saveXl"></param>
/// <param name="cancellationToken"></param>
/// <param name="progressActionSet"></param>
/// <param name="fileExistsOptions"></param>
public static void MakeHomodimerStatisticsSpreadsheetsAndOutputFiles(decimal maxAtomInterationDistance, string[] fastaFiles, string[] pdbFilesFolders, string spreadsheetSaveFilenameTemplate, bool saveTsv, bool saveXl, CancellationToken cancellationToken, ProgressActionSet progressActionSet = null, FileExistsHandler.FileExistsOptions fileExistsOptions = FileExistsHandler.FileExistsOptions.AppendNumberToFilename)
{
if (fastaFiles == null || fastaFiles.Length == 0)
{
throw new ArgumentOutOfRangeException(nameof(fastaFiles));
}
if (pdbFilesFolders == null || pdbFilesFolders.Length == 0)
{
throw new ArgumentOutOfRangeException(nameof(pdbFilesFolders));
}
if (!saveTsv && !saveXl)
{
throw new ArgumentOutOfRangeException(nameof(saveTsv));
}
for (int fastaFileNumber = 0; fastaFileNumber < fastaFiles.Length; fastaFileNumber++)
{
string fastaFilename = fastaFiles[fastaFileNumber];
if (string.IsNullOrWhiteSpace(fastaFilename))
{
continue;
}
ProgressActionSet.Report("Attempting to open file: " + fastaFilename, progressActionSet);
List <ISequence> sequences = SequenceFileHandler.LoadSequenceFile(fastaFilename, StaticValues.MolNameProteinAcceptedValues);
var pdbIdChainIdList = ProteinDataBankFileOperations.PdbIdChainIdList(sequences);
if ((sequences == null) || (sequences.Count == 0))
{
ProgressActionSet.Report("Error could not load file: " + fastaFilename, progressActionSet);
continue;
}
ProgressActionSet.Report("Loaded " + sequences.Count + " sequences from file: " + fastaFilename, progressActionSet);
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequences);
string appendFilename = FileAndPathMethods.FullPathToFilename(fastaFilename);
ProgressActionSet.Report("Creating spreadsheets...", progressActionSet);
Stopwatch stopwatch = Stopwatch.StartNew();
var spreadsheetList = MakeHomodimerStatisticsSpreadsheetsList(cancellationToken, maxAtomInterationDistance, pdbFilesFolders, pdbIdList, pdbIdChainIdList, progressActionSet);
stopwatch.Stop();
ProgressActionSet.Report("Finished calculating spreadsheet data [Elapsed: " + stopwatch.Elapsed.ToString(@"dd\:hh\:mm\:ss") + "]", progressActionSet);
if (cancellationToken.IsCancellationRequested)
{
//UserProteinInterfaceOperations.ProgressBarReset(progressBar, 0, 100, 0);
////UserProteinInterfaceOperations.LabelEstimatedTimeRemainingUpdate(estimatedTimeRemaining, 0, 1, 1);
ProgressActionSet.StartAction(100, progressActionSet);
ProgressActionSet.ProgressAction(100, progressActionSet);
ProgressActionSet.FinishAction(false, progressActionSet);
ProgressActionSet.Report("Cancelled.", progressActionSet);
break;
}
for (int spreadsheetIndex = 0; spreadsheetIndex < spreadsheetList.Count; spreadsheetIndex++)
{
var spreadsheet = spreadsheetList[spreadsheetIndex];
if (cancellationToken.IsCancellationRequested)
{
break;
}
// Remove the first row (which has the name for use in a worksheet title, not currently used)
var sheetName = spreadsheet[0][0].CellData;
var spreadsheetName = spreadsheet[1][0].CellData;
spreadsheet.RemoveAt(0);
// "c:/dResults/Results - %date% %time% - %fasta_filename% - %spreadsheet_name%.tsv"
string saveFilename = spreadsheetSaveFilenameTemplate;
saveFilename = saveFilename.Replace("%spreadsheet_name%", spreadsheetName);
saveFilename = saveFilename.Replace("%fasta_filename%", appendFilename);
saveFilename = saveFilename.Replace("%date%", DateTime.Now.ToString("yyyy-MM-dd"));
saveFilename = saveFilename.Replace("%time%", DateTime.Now.ToString("HH.mm.ss"));
saveFilename = saveFilename.Replace("%batch_number%", ""); //string.Empty + (fastaFileNumber + 1));
saveFilename = saveFilename.Replace("%batch_letter%", ""); //SpreadsheetFileHandler.AlphabetLetterRollOver(spreadsheetIndex));
sheetName = sheetName.Replace("%spreadsheet_name%", spreadsheetName);
sheetName = sheetName.Replace("%fasta_filename%", appendFilename);
sheetName = sheetName.Replace("%date%", DateTime.Now.ToString("yyyy-MM-dd"));
sheetName = sheetName.Replace("%time%", DateTime.Now.ToString("HH.mm.ss"));
sheetName = sheetName.Replace("%batch_number%", ""); //string.Empty + (fastaFileNumber + 1));
sheetName = sheetName.Replace("%batch_letter%", ""); //SpreadsheetFileHandler.AlphabetLetterRollOver(spreadsheetIndex));
//var tsvFilename = new FileInfo(FileAndPathMethods.RemoveFileExtension(saveFilename) + ".tsv");
var xlFilename = new FileInfo(FileAndPathMethods.RemoveFileExtension(saveFilename) + ".xlsx");
var savedFiles = SpreadsheetFileHandler.SaveSpreadsheet(xlFilename.FullName, new[] { sheetName }, spreadsheet, null, saveTsv, saveXl, fileExistsOptions);
ProgressActionSet.ReportFilesSaved(savedFiles, progressActionSet);
}
}
ProgressActionSet.Report("Finished processing files.", progressActionSet);
}
public static void FilterProteinInterfaceLengths(
CancellationToken cancellationToken,
decimal maxAtomInterationDistance,
decimal minimumProteinInterfaceDensity,
string[] sequenceListFileArray,
string[] pdbFileDirectoryLocationArray,
string filterProteinInterfacesLengthOutputFilename,
bool filterProteinInterfaceCountsWithoutLengths,
bool filterProteinInterfaceCountsWithLengths,
FileExistsHandler.FileExistsOptions fileExistsOptions,
ProgressActionSet progressActionSet)
{
if (sequenceListFileArray == null)
{
throw new ArgumentNullException(nameof(sequenceListFileArray));
}
if (pdbFileDirectoryLocationArray == null)
{
throw new ArgumentNullException(nameof(pdbFileDirectoryLocationArray));
}
if (filterProteinInterfacesLengthOutputFilename == null)
{
throw new ArgumentNullException(nameof(filterProteinInterfacesLengthOutputFilename));
}
if (!filterProteinInterfaceCountsWithoutLengths && !filterProteinInterfaceCountsWithLengths)
{
ProgressActionSet.Report("Cancelled: No filter options selected.", progressActionSet);
return;
}
// Check all sequence files are found
var missingSequenceFiles = sequenceListFileArray.Where(sequenceFile => !string.IsNullOrWhiteSpace(sequenceFile) && !File.Exists(sequenceFile)).ToList();
if (missingSequenceFiles.Count > 0)
{
foreach (string missingSequenceFile in missingSequenceFiles)
{
//throw new FileNotFoundException(sequenceFile);
ProgressActionSet.Report("Warning: Sequence file missing: " + missingSequenceFile, progressActionSet);
}
ProgressActionSet.Report("Cancelled: missing sequence files.", progressActionSet);
return;
}
// Check all pdb folders are found
var missingDirectoryList = pdbFileDirectoryLocationArray.Where(pdbDirectory => !string.IsNullOrWhiteSpace(pdbDirectory) && !Directory.Exists(pdbDirectory)).ToList();
if (missingDirectoryList.Count > 0)
{
foreach (string pdbDirectory in missingDirectoryList)
{
//throw new DirectoryNotFoundException(pdbDirectory);
ProgressActionSet.Report("Warning: Structure file directory missing: " + pdbDirectory, progressActionSet);
}
ProgressActionSet.Report("Cancelled: missing structure file directory.", progressActionSet);
return;
}
const string proteinInterfacesTemplateText = "%proteinInterfaces%";
if (string.IsNullOrWhiteSpace(filterProteinInterfacesLengthOutputFilename) || !filterProteinInterfacesLengthOutputFilename.Contains(proteinInterfacesTemplateText))
{
throw new ArgumentOutOfRangeException(nameof(filterProteinInterfacesLengthOutputFilename));
}
// Load fasta sequence files
List <ISequence> sequenceList = SequenceFileHandler.LoadSequenceFileList(sequenceListFileArray, StaticValues.MolNameProteinAcceptedValues);
// Get a list of the PDB Unique IDs with unique chain IDs which are wanted, ignoring others which may be present e.g. dna
var pdbIdChainIdList = ProteinDataBankFileOperations.PdbIdChainIdList(sequenceList);
// Get list of PDB Unique IDs
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequenceList);
// Check PDB Unique IDs were successfully loaded
if (pdbIdList == null || pdbIdList.Count == 0)
{
//throw new Exception("PDB ID List is empty or could not be loaded.");
ProgressActionSet.Report("Error: Sequence list could not be loaded", progressActionSet);
return;
}
// 3: Get a list of PDB files found in user specified directory
string[] pdbFilesArray = ProteinDataBankFileOperations.RemoveNonWhiteListedPdbIdFromPdbFilesArray(pdbIdList, ProteinDataBankFileOperations.GetPdbFilesArray(pdbFileDirectoryLocationArray));
// Check all PDB files are found
List <string> missingPdbFilesList = ProteinDataBankFileOperations.CheckForMissingPdbFiles(pdbFilesArray, pdbIdList);
if (missingPdbFilesList != null && missingPdbFilesList.Count > 0)
{
ProgressActionSet.Report("Missing PDB Files: " + string.Join(", ", missingPdbFilesList), progressActionSet);
}
ProgressActionSet.StartAction(pdbFilesArray.Length, progressActionSet);
int progressIncrement = 0;
var proteinInterfacesCountResultWithLengths = new Dictionary <string, List <string> >();
var startTicks = DateTime.Now.Ticks;
// 4: Loop through each pdb file
for (int pdbFileNumber = 0; pdbFileNumber < pdbFilesArray.Length + 1; pdbFileNumber++) // +1 is for progress update
{
if (progressIncrement > 0)
{
ProgressActionSet.ProgressAction(progressIncrement, progressActionSet);
progressIncrement = 0;
if (pdbFileNumber >= pdbFilesArray.Length)
{
break;
}
}
ProgressActionSet.EstimatedTimeRemainingAction(startTicks, pdbFileNumber, pdbFilesArray.Length, progressActionSet);
progressIncrement++;
// get unique id of pdb file
string pdbFilename = pdbFilesArray[pdbFileNumber];
string proteinId = ProteinDataBankFileOperations.PdbIdFromPdbFilename(pdbFilename);
// check pdb unique id was in the loaded sequence list
if (!pdbIdList.Contains(proteinId))
{
continue;
}
// perform clustering to detect interaction proteinInterfaces
ClusterProteinDataBankFileResult clusterPdbFileResult = Clustering.ClusterProteinDataBankFile(cancellationToken, maxAtomInterationDistance, minimumProteinInterfaceDensity, pdbFilename, pdbIdChainIdList, ClusteringMethodOptions.ClusterWithResidueSequenceIndex, -1, -1, null);
if (clusterPdbFileResult == null)
{
continue;
}
int[] proteinInterfacesCount = new int[clusterPdbFileResult.ClusteringFullResultListContainer.ChainList.Count];
for (int chainIndex = 0; chainIndex < clusterPdbFileResult.ClusteringFullResultListContainer.ChainList.Count; chainIndex++)
{
int totalProteinInterfaces = clusterPdbFileResult.ProteinInterfaceAnalysisResultData.ProteinInterfacesSequenceAndPositionDataList.Count(proteinInterface => proteinInterface.FullProteinInterfaceId.ChainId == chainIndex);
proteinInterfacesCount[chainIndex] = totalProteinInterfaces;
}
var proteinInterfacesCountStr = string.Join(" ", proteinInterfacesCount.OrderBy(x => x));
List <ProteinInterfaceSequenceAndPositionData> proteinInterfaces = clusterPdbFileResult.ProteinInterfaceAnalysisResultData.ProteinInterfacesSequenceAndPositionDataList;
int[] proteinInterfacesLength = new int[proteinInterfaces.Count];
for (int index = 0; index < proteinInterfaces.Count; index++)
{
ProteinInterfaceSequenceAndPositionData proteinInterface = proteinInterfaces[index];
proteinInterfacesLength[index] = proteinInterface.ProteinInterfaceLength;
}
var proteinInterfacesLengthStr = string.Join(" ", proteinInterfacesLength.Distinct().OrderBy(x => x));
if (proteinInterfacesLength.Length == 0)
{
proteinInterfacesLengthStr = 0.ToString();
}
var chainsCountStr = clusterPdbFileResult.ClusteringFullResultListContainer.ChainList.Count;
if (filterProteinInterfaceCountsWithoutLengths)
{
var combinedKeyAll = "chains [" + chainsCountStr + "] proteinInterfaces [" + proteinInterfacesCountStr + "]";
if (!proteinInterfacesCountResultWithLengths.ContainsKey(combinedKeyAll))
{
proteinInterfacesCountResultWithLengths.Add(combinedKeyAll, new List <string>());
}
proteinInterfacesCountResultWithLengths[combinedKeyAll].Add(proteinId);
}
if (filterProteinInterfaceCountsWithLengths)
{
var combinedKeyWithLengths = "chains [" + chainsCountStr + "] proteinInterfaces [" + proteinInterfacesCountStr + "] lengths [" + proteinInterfacesLengthStr + "]";
if (!proteinInterfacesCountResultWithLengths.ContainsKey(combinedKeyWithLengths))
{
proteinInterfacesCountResultWithLengths.Add(combinedKeyWithLengths, new List <string>());
}
proteinInterfacesCountResultWithLengths[combinedKeyWithLengths].Add(proteinId);
}
}
var confirmSaveList = new List <string>();
foreach (var kvp in proteinInterfacesCountResultWithLengths)
{
var seq2 = new List <ISequence>(sequenceList);
seq2 = FilterProteins.RemoveSequences(cancellationToken, seq2, kvp.Value, FilterProteins.RemoveSequencesOptions.RemoveSequencesNotInList);
var saveFilename = filterProteinInterfacesLengthOutputFilename;
saveFilename = saveFilename.Replace(proteinInterfacesTemplateText, kvp.Key);
var actualSavedFilename = SequenceFileHandler.SaveSequencesAsFasta(seq2, saveFilename, true, fileExistsOptions, progressActionSet);
if (!string.IsNullOrWhiteSpace(actualSavedFilename))
{
confirmSaveList.Add(actualSavedFilename);
}
}
// Confirm the total number of sequences saved is equal to original number loaded
ConfirmSequencesSaved(pdbIdList, confirmSaveList, progressActionSet);
ProgressActionSet.FinishAction(true, progressActionSet);
}
/// <summary>
/// Makes spreadsheets with scientific data outputs about given proteins.
/// </summary>
/// <param name="cancellationToken"></param>
/// <param name="pdbFolders">The location of the PDB files</param>
/// <param name="pdbIdList">The PDB files which should be used.</param>
/// <param name="consoleTextBox"></param>
/// <param name="progressBar">User proteinInterface progress bar for user feedback.</param>
/// <param name="estimatedTimeRemainingLabel">User proteinInterface estimated time remaining label for user feedback.</param>
/// <param name="requestedTotalThreads"></param>
/// <returns>Returns the generated spreadsheets with scientific data.</returns>
public static List <List <SpreadsheetCell[]> > MakeHomodimerStatisticsSpreadsheetsList(CancellationToken cancellationToken, decimal maxAtomInterationDistance, string[] pdbFolders, List <string> pdbIdList = null, Dictionary <string, List <string> > pdbIdChainIdList = null, ProgressActionSet progressActionSet = null, int requestedTotalThreads = -1)
{
if (pdbFolders == null || pdbFolders.Length == 0)
{
throw new ArgumentOutOfRangeException(nameof(pdbFolders));
}
if (pdbIdList == null || pdbIdList.Count == 0)
{
throw new ArgumentOutOfRangeException(nameof(pdbIdList));
}
if (progressActionSet == null)
{
throw new ArgumentNullException(nameof(progressActionSet));
}
// this method creates
// 1. a list of interactions
// 2. a list of symmetry percentage
// 3. an "expected" heatmap by combining every possible a/b amino acid combination
// 4. an actual heatmap for the proteinInterfaces
// 5. normalised versions of both of the heatmaps
string[] pdbFilesArray = ProteinDataBankFileOperations.RemoveNonWhiteListedPdbIdFromPdbFilesArray(pdbIdList, ProteinDataBankFileOperations.GetPdbFilesArray(pdbFolders));
//var interactionRecordList = new List<ProteinInteractionRecord>();
//var interactionMatchPercentageList = new List<InteractionMatchPercentage>();
//var wholeProteinChainsAminoAcidCounter = new List<AminoAcidChainComposition>();
//var interactionChainsAminoAcidCounter = new List<AminoAcidChainComposition>();
//var interactionsAminoAcidToAminoAcidCounter = new AminoAcidPairCompositionMatrix();
////var wholeProteinAminoAcidToAminoAcidCounter2x2 = new AminoAcidPairCompositionMatrix(); // composition of every amino acid paired in every possible combination
var workDivision = new WorkDivision <HomodimersStatisticsMinerTaskResult>(pdbFilesArray.Length, requestedTotalThreads);
ProgressActionSet.StartAction(pdbFilesArray.Length, progressActionSet);
int checkAllFilesProcessed = 0;
var lockCheckAllFilesProcessed = new object();
var pdbFilesProcessed = new bool[pdbFilesArray.Length];
Array.Clear(pdbFilesProcessed, 0, pdbFilesProcessed.Length);
for (int threadIndex = 0; threadIndex < workDivision.ThreadCount; threadIndex++)
{
int localThreadIndex = threadIndex;
Task <HomodimersStatisticsMinerTaskResult> task = Task.Run(() =>
{
var result = new HomodimersStatisticsMinerTaskResult();
for (int pdbFileNumber = workDivision.ThreadFirstIndex[localThreadIndex]; pdbFileNumber <= workDivision.ThreadLastIndex[localThreadIndex]; pdbFileNumber++)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
lock (lockCheckAllFilesProcessed)
{
checkAllFilesProcessed++;
pdbFilesProcessed[pdbFileNumber] = true;
}
try
{
string pdbFilename = pdbFilesArray[pdbFileNumber];
string proteinId = ProteinDataBankFileOperations.PdbIdFromPdbFilename(pdbFilename);
// Check if the file found is included in the white list.
if (/*pdbIdList != null && */ !pdbIdList.Contains(proteinId))
{
ProgressActionSet.Report("Error: " + proteinId + " was not in the PDB ID white list.", progressActionSet);
continue;
}
List <AtomPair> interactions = SearchInteractions.FindInteractions(cancellationToken, maxAtomInterationDistance, pdbFilename, pdbIdChainIdList);
// Make a list to save interactions found.
var interactionMatchPercentage = new InteractionMatchPercentage(proteinId);
var chainAminoAcidCounterA1X1 = new AminoAcidChainComposition(proteinId, "A");
var chainAminoAcidCounterB1X1 = new AminoAcidChainComposition(proteinId, "B");
var chainInteractionAminoAcidCounterA = new AminoAcidChainComposition(proteinId, "A");
var chainInteractionAminoAcidCounterB = new AminoAcidChainComposition(proteinId, "B");
if (interactions != null && interactions.Count > 0)
{
interactionMatchPercentage.IncrementTotalInteractions(interactions.Count);
for (int interactionsIndex = 0; interactionsIndex < interactions.Count; interactionsIndex++)
{
chainInteractionAminoAcidCounterA.IncrementAminoAcidCount(interactions[interactionsIndex].Atom1.resName.FieldValue);
chainInteractionAminoAcidCounterB.IncrementAminoAcidCount(interactions[interactionsIndex].Atom2.resName.FieldValue);
result.InteractionRecordList.Add(new ProteinInteractionRecord(proteinId, interactionsIndex + 1, interactions[interactionsIndex]));
interactionMatchPercentage.AddResidueSequenceIndex(StaticValues.ChainA, interactions[interactionsIndex].Atom1.resSeq.FieldValue);
interactionMatchPercentage.AddResidueSequenceIndex(StaticValues.ChainB, interactions[interactionsIndex].Atom2.resSeq.FieldValue);
result.InteractionsAminoAcidToAminoAcidCounter.IncrementAminoAcidCount(interactions[interactionsIndex].Atom1.resName.FieldValue, interactions[interactionsIndex].Atom2.resName.FieldValue);
}
}
var chainIdList = pdbIdChainIdList != null ? (pdbIdChainIdList.ContainsKey(proteinId) ? pdbIdChainIdList[proteinId].ToArray() : null) : null;
ProteinChainListContainer proteinFileChains = ProteinDataBankFileOperations.PdbAtomicChains(pdbFilename, chainIdList, 2, 2, true);
if (proteinFileChains == null || proteinFileChains.ChainList == null || proteinFileChains.ChainList.Count != 2 ||
proteinFileChains.ChainList[StaticValues.ChainA] == null || proteinFileChains.ChainList[StaticValues.ChainA].AtomList == null || proteinFileChains.ChainList[StaticValues.ChainA].AtomList.Count == 0 ||
proteinFileChains.ChainList[StaticValues.ChainB] == null || proteinFileChains.ChainList[StaticValues.ChainB].AtomList == null || proteinFileChains.ChainList[StaticValues.ChainB].AtomList.Count == 0)
{
if (!File.Exists(pdbFilename))
{
ProgressActionSet.Report("Error: " + pdbFilename + " (" + proteinId + ") file not found", progressActionSet);
}
else
{
int proteinFileChainCount = ProteinDataBankFileOperations.PdbAtomicChainsCount(pdbFilename);
ProgressActionSet.Report("Error: " + proteinId + " did not have exactly 2 chains (" + proteinFileChainCount + " chains found) - skipping.", progressActionSet);
}
continue;
}
// count total of how many of each type of amino acids are in Chain A.
for (int atomIndexA = 0; atomIndexA < proteinFileChains.ChainList[StaticValues.ChainA].AtomList.Count; atomIndexA++)
{
chainAminoAcidCounterA1X1.IncrementAminoAcidCount(proteinFileChains.ChainList[StaticValues.ChainA].AtomList[atomIndexA].resName.FieldValue);
}
// count total of how many of each type of amino acids are in Chain B.
for (int atomIndexB = 0; atomIndexB < proteinFileChains.ChainList[StaticValues.ChainB].AtomList.Count; atomIndexB++)
{
chainAminoAcidCounterB1X1.IncrementAminoAcidCount(proteinFileChains.ChainList[StaticValues.ChainB].AtomList[atomIndexB].resName.FieldValue);
}
interactionMatchPercentage.CalculatePercentage();
result.InteractionMatchPercentageList.Add(interactionMatchPercentage);
result.WholeProteinChainsAminoAcidCounter.Add(chainAminoAcidCounterA1X1);
result.WholeProteinChainsAminoAcidCounter.Add(chainAminoAcidCounterB1X1);
result.InteractionChainsAminoAcidCounter.Add(chainInteractionAminoAcidCounterA);
result.InteractionChainsAminoAcidCounter.Add(chainInteractionAminoAcidCounterB);
}
finally
{
workDivision.IncrementItemsCompleted(1);
ProgressActionSet.ProgressAction(1, progressActionSet);
ProgressActionSet.EstimatedTimeRemainingAction(workDivision.StartTicks, workDivision.ItemsCompleted, workDivision.ItemsToProcess, progressActionSet);
}
}
return(result);
}, cancellationToken);
workDivision.TaskList.Add(task);
}
workDivision.WaitAllTasks();
ProgressActionSet.FinishAction(true, progressActionSet);
// merge all instances of the results
var spreadsheetTaskResult = new HomodimersStatisticsMinerTaskResult();
foreach (var task in workDivision.TaskList.Where(t => t != null && t.IsCompleted && !t.IsCanceled && !t.IsFaulted && t.Result != null))
{
if (task.Result.InteractionChainsAminoAcidCounter != null && task.Result.InteractionChainsAminoAcidCounter.Count > 0)
{
spreadsheetTaskResult.InteractionChainsAminoAcidCounter.AddRange(task.Result.InteractionChainsAminoAcidCounter);
}
if (task.Result.InteractionMatchPercentageList != null && task.Result.InteractionMatchPercentageList.Count > 0)
{
spreadsheetTaskResult.InteractionMatchPercentageList.AddRange(task.Result.InteractionMatchPercentageList);
}
if (task.Result.InteractionRecordList != null && task.Result.InteractionRecordList.Count > 0)
{
spreadsheetTaskResult.InteractionRecordList.AddRange(task.Result.InteractionRecordList);
}
if (task.Result.InteractionsAminoAcidToAminoAcidCounter != null)
{
if (task.Result.InteractionsAminoAcidToAminoAcidCounter.AminoAcidToAminoAcid != null)
{
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
var totalGroups = AminoAcidGroups.AminoAcidGroups.GetTotalSubgroups(enumAminoAcidGroups);
for (int x = 0; x < totalGroups; x++)
{
for (int y = 0; y < totalGroups; y++)
{
spreadsheetTaskResult.InteractionsAminoAcidToAminoAcidCounter.AminoAcidToAminoAcid[(int)enumAminoAcidGroups][x, y] +=
task.Result.InteractionsAminoAcidToAminoAcidCounter.AminoAcidToAminoAcid[(int)enumAminoAcidGroups][x, y];
}
}
}
}
}
if (task.Result.WholeProteinChainsAminoAcidCounter != null && task.Result.WholeProteinChainsAminoAcidCounter.Count > 0)
{
spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter.AddRange(task.Result.WholeProteinChainsAminoAcidCounter);
}
}
if (pdbFilesProcessed.Count(file => file == false) > 0)
{
ProgressActionSet.Report("ERROR: " + pdbFilesProcessed.Count(file => file == false) + " PDB FILES WERE SKIPPED! 0x01", progressActionSet);
}
else
{
ProgressActionSet.Report("CHECK: NO PDB FILES WERE SKIPPED! 0x01", progressActionSet);
}
if (checkAllFilesProcessed != pdbFilesArray.Length)
{
ProgressActionSet.Report("ERROR: " + (pdbFilesArray.Length - checkAllFilesProcessed) + " PDB FILES WERE SKIPPED! 0x02", progressActionSet);
}
else
{
ProgressActionSet.Report("CHECK: NO PDB FILES WERE SKIPPED! 0x02", progressActionSet);
}
spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter = spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter.OrderBy(a => a.ProteinId).ThenBy(b => b.ChainId).ToList();
spreadsheetTaskResult.InteractionChainsAminoAcidCounter = spreadsheetTaskResult.InteractionChainsAminoAcidCounter.OrderBy(a => a.ProteinId).ThenBy(b => b.ChainId).ToList();
AminoAcidChainComposition wholeProteinChainsTotals = AminoAcidChainComposition.TotalFromAminoAcidChainCompositionList(spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter);
AminoAcidChainComposition interactionChainsTotals = AminoAcidChainComposition.TotalFromAminoAcidChainCompositionList(spreadsheetTaskResult.InteractionChainsAminoAcidCounter);
AminoAcidPairCompositionMatrix wholeProteinAminoAcidToAminoAcidCounter1X1 = AminoAcidChainComposition.ConvertToMatrix(wholeProteinChainsTotals);
var results = new List <List <SpreadsheetCell[]> >();
{
/* start test */
var spreadsheet1 = new List <SpreadsheetCell[]>();
spreadsheet1.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% TEST SHEET 0"), });
spreadsheet1.Add(new[] { new SpreadsheetCell("TEST SHEET 0"), });
foreach (AminoAcidChainComposition item in spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter)
{
//spreadsheet1.Add(item.ProteinId);
//spreadsheet1.Add(item.ChainId);
spreadsheet1.Add(item.SpreadsheetDataRow());
}
results.Add(spreadsheet1);
spreadsheet1 = null;
/* end test */
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
{
var spreadsheet2 = new List <SpreadsheetCell[]>();
spreadsheet2.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Interaction Symmetry"), });
spreadsheet2.Add(new[] { new SpreadsheetCell("Homodimers - List - Interaction Count And Interaction Match Percentage (Symmetry Measurement)") });
spreadsheet2.Add(InteractionMatchPercentage.SpreadsheetColumnHeadersRow());
var range2 = spreadsheetTaskResult.InteractionMatchPercentageList.Select(record => record.SpreadsheetDataRow()).ToList();
//range2.Sort();
range2 = range2
.OrderBy(a => a[0].CellData)
.ThenBy(a => a[1].CellData)
.ThenBy(a => a[2].CellData)
.ThenBy(a => a[3].CellData)
.ThenBy(a => a[4].CellData)
.ThenBy(a => a[5].CellData)
.ThenBy(a => a[6].CellData)
.ThenBy(a => a[7].CellData)
.ThenBy(a => a[8].CellData)
.ToList();
spreadsheet2.AddRange(range2);
range2 = null;
results.Add(spreadsheet2);
var spreadsheetHistogram2 = new List <SpreadsheetCell[]>();
spreadsheetHistogram2.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% HG Interaction Symmetry"), });
spreadsheetHistogram2.Add(new[] { new SpreadsheetCell("Homodimers - List - Interaction Count And Interaction Match Percentage (Symmetry Measurement) Histogram") });
spreadsheetHistogram2.AddRange(Histogram.MatrixToHistogram(spreadsheet2.ToArray(), Histogram.MakeBinDecimals(0, 100, 9, 10), new[] { 6, 7, 8 }, 2, -1, true));
results.Add(spreadsheetHistogram2);
spreadsheet2 = null;
spreadsheetHistogram2 = null;
}
//
{
var spreadsheet3 = new List <SpreadsheetCell[]>();
spreadsheet3.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Interaction Records"), });
spreadsheet3.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Interaction Record"), });
spreadsheet3.Add(ProteinInteractionRecord.TsvColumnHeadersRow());
var range3 = spreadsheetTaskResult.InteractionRecordList.Select(record => record.SpreadsheetDataRow()).ToList();
//range3.Sort();
range3 = range3
.OrderBy(a => a[0].CellData)
.ThenBy(a => a[1].CellData)
.ThenBy(a => a[3].CellData)
.ThenBy(a => a[5].CellData)
.ThenBy(a => a[13].CellData)
.ThenBy(a => a[15].CellData)
.ToList();
spreadsheet3.AddRange(range3);
range3 = null;
results.Add(spreadsheet3);
var spreadsheetHistogram3 = new List <SpreadsheetCell[]>();
spreadsheetHistogram3.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Interaction Records Histogram"), });
spreadsheetHistogram3.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Interaction Record - Histogram"), });
spreadsheetHistogram3.AddRange(Histogram.MatrixToHistogram(spreadsheet3.ToArray(), Histogram.MakeBinDecimals(0m, 5m, 0m, 0.05m), new[] { 1 }, 2, -1, true));
results.Add(spreadsheetHistogram3);
//spreadsheet3 = Histogram.InsertMatrixOverwrite(spreadsheet3.ToArray(), histogram3, 2, Histogram.MaxColumns(spreadsheet3.ToArray()) + 1).ToList();
spreadsheet3 = null;
spreadsheetHistogram3 = null;
}
//
{
var spreadsheet4 = new List <SpreadsheetCell[]>();
spreadsheet4.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Interaction Count - A-Z"), });
spreadsheet4.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Amino Acid Count - Interactions - A to Z"), });
spreadsheet4.Add(AminoAcidChainComposition.SpreadsheetTitleRow());
var range4 = spreadsheetTaskResult.InteractionChainsAminoAcidCounter.Select(record => record.SpreadsheetDataRow()).ToList();
//range4.Sort();
range4 = range4
.OrderBy(a => a[0].CellData)
.ThenBy(a => a[1].CellData)
.ToList();
spreadsheet4.AddRange(range4);
range4 = null;
spreadsheet4.Add(interactionChainsTotals.SpreadsheetDataRow());
results.Add(spreadsheet4);
spreadsheet4 = null;
}
//
{
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
var spreadsheet5 = new List <SpreadsheetCell[]>();
spreadsheet5.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Interaction Count - Groups " + enumAminoAcidGroups), });
spreadsheet5.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Amino Acid Count - Interactions - Acid Groups " + enumAminoAcidGroups), });
spreadsheet5.Add(AminoAcidChainComposition.SpreadsheetGroupsTitleRow(enumAminoAcidGroups));
var range5 = spreadsheetTaskResult.InteractionChainsAminoAcidCounter.Select(record => record.SpreadsheetGroupsDataRow(enumAminoAcidGroups)).ToList();
//range4.Sort();
range5 = range5
.OrderBy(a => a[0].CellData)
.ThenBy(a => a[1].CellData)
.ToList();
spreadsheet5.AddRange(range5);
range5 = null;
spreadsheet5.Add(interactionChainsTotals.SpreadsheetGroupsDataRow(enumAminoAcidGroups));
results.Add(spreadsheet5);
spreadsheet5 = null;
}
}
//
{
var spreadsheet6 = new List <SpreadsheetCell[]>();
spreadsheet6.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Entire Count - A-Z"), });
spreadsheet6.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Amino Acid Count - All Atoms - A to Z"), });
spreadsheet6.Add(AminoAcidChainComposition.SpreadsheetTitleRow());
var range6 = spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter.Select(record => record.SpreadsheetDataRow()).ToList();
//range6.Sort();
range6 = range6
.OrderBy(a => a[0].CellData)
.ThenBy(a => a[1].CellData)
.ToList();
spreadsheet6.AddRange(range6);
range6 = null;
spreadsheet6.Add(wholeProteinChainsTotals.SpreadsheetDataRow());
results.Add(spreadsheet6);
var spreadsheetHistogram6 = new List <SpreadsheetCell[]>();
spreadsheetHistogram6.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Entire Count - A-Z - Historgram"), });
spreadsheetHistogram6.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Amino Acid Count - All Atoms - A to Z - Histogram"), });
spreadsheetHistogram6.AddRange(Histogram.MatrixToHistogram(spreadsheet6.ToArray(), Histogram.MakeBinDecimals(0, 10500, 0, 500), new[] { 28 }, 2, -1, true));
spreadsheetHistogram6.Add(new [] { new SpreadsheetCell(""), });
spreadsheetHistogram6.AddRange(Histogram.MatrixToHistogram(spreadsheet6.ToArray(), Histogram.MakeBinDecimals(0, 1000, 0, 100), new[] { 28 }, 2, -1, true));
results.Add(spreadsheetHistogram6);
spreadsheet6 = null;
spreadsheetHistogram6 = null;
}
//
{
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
var spreadsheet7 = new List <SpreadsheetCell[]>();
spreadsheet7.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% L Entire Count - Groups " + enumAminoAcidGroups), });
spreadsheet7.Add(new[] { new SpreadsheetCell("Homodimers - List - Protein Amino Acid Count - All Atoms - Acid Groups " + enumAminoAcidGroups), });
spreadsheet7.Add(AminoAcidChainComposition.SpreadsheetGroupsTitleRow(enumAminoAcidGroups));
var range7 = spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter.Select(record => record.SpreadsheetGroupsDataRow(enumAminoAcidGroups)).ToList();
//range7.Sort();
range7 = range7
.OrderBy(a => a[0].CellData)
.ThenBy(a => a[1].CellData)
.ToList();
spreadsheet7.AddRange(range7);
range7 = null;
spreadsheet7.Add(wholeProteinChainsTotals.SpreadsheetGroupsDataRow(enumAminoAcidGroups));
results.Add(spreadsheet7);
spreadsheet7 = null;
}
}
// convert to percentage for creating mean average protein composition
var meanProteinComposition = new AminoAcidChainComposition("Mean Composition", "-");
foreach (AminoAcidChainComposition aminoAcidChainComposition in spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter)
{
// get percentage for row
AminoAcidChainComposition percentage = AminoAcidChainComposition.ConvertToPercentage(aminoAcidChainComposition);
// add percentage to overall tally
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
for (int x = 0; x < AminoAcidGroups.AminoAcidGroups.GetTotalSubgroups(enumAminoAcidGroups); x++)
{
meanProteinComposition.AminoAcidGroupsCount[(int)enumAminoAcidGroups][x] += (percentage.AminoAcidGroupsCount[(int)enumAminoAcidGroups][x] / spreadsheetTaskResult.WholeProteinChainsAminoAcidCounter.Count);
}
}
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
{
/* start test */
var spreadsheet8 = new List <SpreadsheetCell[]>();
spreadsheet8.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% TEST SHEET 1"), }); // Worksheet name.
spreadsheet8.Add(new[] { new SpreadsheetCell("TEST SHEET 1"), }); // Spreadsheet title
spreadsheet8.Add(new[] { new SpreadsheetCell(string.Empty), });
spreadsheet8.Add(meanProteinComposition.SpreadsheetDataRow());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet8.Add(meanProteinComposition.SpreadsheetGroupsDataRow(enumAminoAcidGroups));
}
results.Add(spreadsheet8);
spreadsheet8 = null;
/* end test */
}
AminoAcidPairCompositionMatrix meanProteinMatrix = AminoAcidChainComposition.ConvertToMatrix(meanProteinComposition);
{
var spreadsheet9 = new List <SpreadsheetCell[]>();
spreadsheet9.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% HM All Atoms 3x3"), }); // Worksheet name.
spreadsheet9.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Average Chain Composition"), }); // Spreadsheet title.
//spreadsheet9.Add(new[] { new SpreadsheetCell(string.Empty), });
//spreadsheet9.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Average Chain Composition - Percentage Composition - A to Z"), }); // Section title.
//spreadsheet9.AddRange(meanProteinMatrix.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet9.Add(new[] { new SpreadsheetCell(string.Empty), });
spreadsheet9.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Average Chain Composition - Percentage Composition - Acid Groups " + enumAminoAcidGroups), }); // Section title.
spreadsheet9.AddRange(meanProteinMatrix.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
results.Add(spreadsheet9);
spreadsheet9 = null;
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
//if (outputAllAtoms1x1)
//{
AminoAcidPairCompositionMatrix wholeProteinAminoAcidToAminoAcidCounterPercentage1X1 = AminoAcidPairCompositionMatrix.CalculatePercentageMatrix(wholeProteinAminoAcidToAminoAcidCounter1X1);
{
var spreadsheet10 = new List <SpreadsheetCell[]>();
spreadsheet10.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% HM All Atoms 1x1") }); // Worksheet name.
spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall Composition") }); // Spreadsheet title.
//spreadsheet10.Add(new[] { new SpreadsheetCell(string.Empty)});
//spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall Percentage Composition - A to Z")}); // Section title.
//spreadsheet10.AddRange(wholeProteinAminoAcidToAminoAcidCounterPercentage1X1.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet10.Add(new[] { new SpreadsheetCell(string.Empty) });
spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall Percentage Composition - Acid Groups " + enumAminoAcidGroups) }); // Section title.
spreadsheet10.AddRange(wholeProteinAminoAcidToAminoAcidCounterPercentage1X1.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
AminoAcidPairCompositionMatrix wholeProteinAminoAcidToAminoAcidCounterNormalised1X1 = AminoAcidPairCompositionMatrix.NormalizeWithCompositionMatrix(wholeProteinAminoAcidToAminoAcidCounterPercentage1X1, UniProtProteinDatabaseComposition.AminoAcidCompositionAsMatrix());
//spreadsheet10.Add(new[] { new SpreadsheetCell(string.Empty)});
//spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall UniProt Normalised - A to Z ")}); // Section title.
//spreadsheet10.AddRange(wholeProteinAminoAcidToAminoAcidCounterNormalised1X1.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet10.Add(new[] { new SpreadsheetCell(string.Empty) });
spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall UniProt Normalised - Acid Groups " + enumAminoAcidGroups) }); // Section title.
spreadsheet10.AddRange(wholeProteinAminoAcidToAminoAcidCounterNormalised1X1.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
AminoAcidPairCompositionMatrix wholeProteinAminoAcidToAminoAcidCounterDifference1X1 = AminoAcidPairCompositionMatrix.DifferenceWithCompositionMatrix(wholeProteinAminoAcidToAminoAcidCounterPercentage1X1, UniProtProteinDatabaseComposition.AminoAcidCompositionAsMatrix());
//spreadsheet10.Add(new[] { new SpreadsheetCell(string.Empty)});
//spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall A to Z - UniProt Difference")}); // Section title.
//spreadsheet10.AddRange(wholeProteinAminoAcidToAminoAcidCounterDifference1X1.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet10.Add(new[] { new SpreadsheetCell(string.Empty) });
spreadsheet10.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - All Atoms - Overall Acid Groups " + enumAminoAcidGroups + " - UniProt Difference") }); // Section title.
spreadsheet10.AddRange(wholeProteinAminoAcidToAminoAcidCounterDifference1X1.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
results.Add(spreadsheet10);
spreadsheet10 = null;
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
{
AminoAcidPairCompositionMatrix interactionsAminoAcidToAminoAcidCounterPercentage = AminoAcidPairCompositionMatrix.CalculatePercentageMatrix(spreadsheetTaskResult.InteractionsAminoAcidToAminoAcidCounter);
var spreadsheet11 = new List <SpreadsheetCell[]>();
spreadsheet11.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% HM Interactions Only") }); // Worksheet name.
spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only") }); // Spreadsheet title.
//spreadsheet11.Add(new[] { new SpreadsheetCell(string.Empty)});
//spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only - A to Z")}); // Section title.
//spreadsheet11.AddRange(spreadsheetTaskResult.InteractionsAminoAcidToAminoAcidCounter.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet11.Add(new[] { new SpreadsheetCell(string.Empty) });
spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only - Acid Groups " + enumAminoAcidGroups) }); // Section title.
spreadsheet11.AddRange(spreadsheetTaskResult.InteractionsAminoAcidToAminoAcidCounter.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
AminoAcidPairCompositionMatrix interactionsAminoAcidToAminoAcidCounterNormalised = AminoAcidPairCompositionMatrix.NormalizeWithCompositionMatrix(interactionsAminoAcidToAminoAcidCounterPercentage, UniProtProteinDatabaseComposition.AminoAcidCompositionAsMatrix());
//spreadsheet11.Add(new[] { new SpreadsheetCell(string.Empty)});
//spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only - A to Z - UniProt Normalised")}); // Section title.
//spreadsheet11.AddRange(interactionsAminoAcidToAminoAcidCounterNormalised.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet11.Add(new[] { new SpreadsheetCell(string.Empty) });
spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only - Acid Groups " + enumAminoAcidGroups + " - UniProt Normalised") }); // Section title.
spreadsheet11.AddRange(interactionsAminoAcidToAminoAcidCounterNormalised.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
AminoAcidPairCompositionMatrix interactionsAminoAcidToAminoAcidCounterDifference = AminoAcidPairCompositionMatrix.DifferenceWithCompositionMatrix(interactionsAminoAcidToAminoAcidCounterPercentage, UniProtProteinDatabaseComposition.AminoAcidCompositionAsMatrix());
//spreadsheet11.Add(new[] { new SpreadsheetCell(string.Empty)});
//spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only - A to Z - UniProt Difference")}); // Section title.
//spreadsheet11.AddRange(interactionsAminoAcidToAminoAcidCounterDifference.SpreadsheetAminoAcidColorGroupsHeatMap());
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet11.Add(new[] { new SpreadsheetCell(string.Empty) });
spreadsheet11.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Interactions Only - Acid Groups " + enumAminoAcidGroups + " - UniProt Difference") }); // Section title.
spreadsheet11.AddRange(interactionsAminoAcidToAminoAcidCounterDifference.SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
}
results.Add(spreadsheet11);
spreadsheet11 = null;
}
////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
{
var spreadsheet12 = new List <SpreadsheetCell[]>();
spreadsheet12.Add(new[] { new SpreadsheetCell("%batch_number%%batch_letter% HM Interactions v Homodimers") }); // Worksheet name.
spreadsheet12.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Difference between homodimer composition and homodimer interactions") }); // Spreadsheet title
spreadsheet12.Add(new[] { new SpreadsheetCell(string.Empty) });
//spreadsheet12.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Difference between homodimer composition and homodimer interactions - A to Z")}); // Section title
//spreadsheet12.AddRange(AminoAcidPairCompositionMatrix.DifferenceWithCompositionMatrix(wholeProteinAminoAcidToAminoAcidCounterPercentage1X1, spreadsheetTaskResult.InteractionsAminoAcidToAminoAcidCounter).SpreadsheetAminoAcidColorGroupsHeatMap());
//spreadsheet12.Add(new[] { new SpreadsheetCell(string.Empty)});
foreach (AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups enumAminoAcidGroups in Enum.GetValues(typeof(AminoAcidGroups.AminoAcidGroups.EnumAminoAcidGroups)))
{
spreadsheet12.Add(new[] { new SpreadsheetCell("Homodimers - Amino Acid Heat Map - Difference between homodimer composition and homodimer interactions - Acid Groups " + enumAminoAcidGroups) }); // Section title.
spreadsheet12.AddRange(AminoAcidPairCompositionMatrix.DifferenceWithCompositionMatrix(wholeProteinAminoAcidToAminoAcidCounterPercentage1X1, spreadsheetTaskResult.InteractionsAminoAcidToAminoAcidCounter).SpreadsheetAminoAcidColorGroupsHeatMap(enumAminoAcidGroups));
spreadsheet12.Add(new[] { new SpreadsheetCell(string.Empty) });
}
results.Add(spreadsheet12);
}
///////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////
return(results);
}
/// <summary>
/// Filters the given FASTA files and PDB files with the given options and saves the results to disk. Data needs to be
/// cleaned for two reasons, firstly to not pollute or distort the results, and secondly to save unnecessary processing
/// operations.
/// </summary>
/// <param name="cancellationToken"></param>
/// <param name="pdbFilesFolders"></param>
/// <param name="fastaFiles"></param>
/// <param name="proteinOperationOptionFlags"></param>
/// <param name="saveFastaFilenameTemplate"></param>
/// <param name="consoleTextBox"></param>
/// <param name="progressBar"></param>
/// <param name="estimatedTimeRemaining"></param>
public static void CleanProteins(CancellationToken cancellationToken, decimal maxAtomInterationDistance, string[] pdbFilesFolders, string[] fastaFiles, ProteinOperation proteinOperationOptionFlags, string saveFastaFilenameTemplate, ProgressActionSet progressActionSet, FileExistsHandler.FileExistsOptions fileExistsOptions = FileExistsHandler.FileExistsOptions.AppendNumberToFilename)
{
if (pdbFilesFolders == null || pdbFilesFolders.Length == 0)
{
if (proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveMultipleModelsInStructure) || proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveNonInteractingProteinsInStructure) || proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveNonSymmetricalInStructure) || proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveWrongNumberOfChainsInStructure))
{
throw new ArgumentOutOfRangeException(nameof(pdbFilesFolders));
}
}
if (fastaFiles == null || fastaFiles.Length == 0)
{
throw new ArgumentOutOfRangeException(nameof(fastaFiles));
}
if (string.IsNullOrWhiteSpace(saveFastaFilenameTemplate))
{
throw new ArgumentOutOfRangeException(nameof(saveFastaFilenameTemplate));
}
string[] pdbFilesArray = ProteinDataBankFileOperations.GetPdbFilesArray(pdbFilesFolders);
//List<string> pdbIdList = null;
int beforeCount = 0;
int afterCount = 0;
string saveFilename = saveFastaFilenameTemplate;
var currentProteinOperation = ProteinOperation.LoadFile;
int[] numberSequencesLoaded;
var sequences = new List <ISequence> [3];
//UserProteinInterfaceOperations.TextBoxClear(consoleTextBox);
ProgressActionSet.Report("Filtering proteins.", progressActionSet);
// Load fasta/sequence files.
sequences[0] = SequenceFileHandler.LoadSequenceFileList(fastaFiles, StaticValues.MolNameProteinAcceptedValues, out numberSequencesLoaded, true);
var pdbIdChainIdList = ProteinDataBankFileOperations.PdbIdChainIdList(sequences[0]);
for (int numberSequencesLoadedIndex = 0; numberSequencesLoadedIndex < numberSequencesLoaded.Length; numberSequencesLoadedIndex++)
{
if (numberSequencesLoaded[numberSequencesLoadedIndex] > 0)
{
ProgressActionSet.Report("Loaded " + numberSequencesLoaded[numberSequencesLoadedIndex] / 2 + " proteins from file: " + fastaFiles[numberSequencesLoadedIndex], progressActionSet);
}
else
{
ProgressActionSet.Report("Error could not load file: " + fastaFiles[numberSequencesLoadedIndex], progressActionSet);
}
}
if (numberSequencesLoaded.Count(a => a > 0) == 0)
{
return;
}
// Replace placeholder variable names.
saveFilename = saveFilename.Replace("%date%", DateTime.Now.ToString("yyyy-MM-dd"));
saveFilename = saveFilename.Replace("%time%", DateTime.Now.ToString("HH.mm.ss"));
// Save initial loaded sequences.
if (File.Exists(saveFilename))
{
if (fileExistsOptions == FileExistsHandler.FileExistsOptions.AppendNumberToFilename)
{
saveFilename = FileExistsHandler.FindNextFreeOutputFilename(saveFilename);
}
else if (fileExistsOptions == FileExistsHandler.FileExistsOptions.OverwriteFile)
{
}
else if (fileExistsOptions == FileExistsHandler.FileExistsOptions.SkipFile)
{
return;
}
}
// Removes any entries not having a protein alphabet.
while (currentProteinOperation != ProteinOperation.Finished)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
currentProteinOperation = (ProteinOperation)((int)currentProteinOperation * 2);
sequences[1] = null;
sequences[2] = null;
var sequencesDescriptions = new string[3];
if (currentProteinOperation == ProteinOperation.Finished)
{
break;
}
if (currentProteinOperation == ProteinOperation.RemoveNonProteinAlphabetInSequence && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveNonProteinAlphabetInSequence))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveWrongNumberOfChainsInSequence && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveWrongNumberOfChainsInSequence))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveExactDuplicatesInSequence && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveExactDuplicatesInSequence))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveNonHomodimersInSequence && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveNonHomodimersInSequence))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveWrongNumberOfChainsInStructure && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveWrongNumberOfChainsInStructure))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveMultipleModelsInStructure && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveMultipleModelsInStructure))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveNonInteractingProteinsInStructure && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveNonInteractingProteinsInStructure))
{
continue;
}
if (currentProteinOperation == ProteinOperation.RemoveNonSymmetricalInStructure && !proteinOperationOptionFlags.HasFlag(ProteinOperation.RemoveNonSymmetricalInStructure))
{
continue;
}
// Count sequences before operation.
beforeCount = sequences[0].Count / 2;
// Update user about what is happening.
ProgressActionSet.Report("", progressActionSet);
ProgressActionSet.Report("Removing " + ProteinOperationString(currentProteinOperation) + " entries [from " + beforeCount + " proteins]", progressActionSet);
// Start stopwatch to count duration of operation.
Stopwatch stopwatch = Stopwatch.StartNew();
// Perform specified operation.
switch (currentProteinOperation)
{
case ProteinOperation.RemoveNonProteinAlphabetInSequence:
{
sequencesDescriptions[0] = "01 - Removed non-protein sequences (sequence filter)";
sequences[0] = FilterProteins.RemoveNonProteinAlphabetSequences(cancellationToken, sequences[0], progressActionSet);
break;
}
case ProteinOperation.RemoveWrongNumberOfChainsInSequence:
{
sequencesDescriptions[0] = "02 - Removed non-dimers (sequence filter)";
sequences[0] = FilterProteins.RemoveSequencesWithIncorrectNumberOfChains(cancellationToken, sequences[0], 2, progressActionSet);
break;
}
case ProteinOperation.RemoveExactDuplicatesInSequence:
{
sequencesDescriptions[0] = "03 - Removed exact duplicates (sequence filter)";
sequences[0] = FilterProteins.RemoveDuplicates(cancellationToken, sequences[0], progressActionSet);
break;
}
case ProteinOperation.RemoveNonHomodimersInSequence:
{
// homodimers - all types - unfiltered for interactions or symmetry
var result = FilterProteins.SplitDimerTypes(cancellationToken, sequences[0], 30, 90, progressActionSet);
sequencesDescriptions[0] = "04 - Homodimers only (sequence filter)";
sequences[0] = result.HomoDimerPdbIdList;
sequencesDescriptions[1] = "04 - Heterodimers only (sequence filter)";
sequences[1] = result.HeteroDimerPdbIdList;
sequencesDescriptions[2] = "04 - Homology dimers only (sequence filter)";
sequences[2] = result.HomologyDimerPdbIdList;
break;
}
case ProteinOperation.RemoveMultipleModelsInStructure:
{
sequencesDescriptions[0] = "05 - Removed multiple models (structure filter)";
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequences[0]);
pdbIdList = FilterProteins.RemoveMultipleStructureModels(cancellationToken, pdbFilesFolders, pdbIdList, progressActionSet);
sequences[0] = FilterProteins.RemoveSequences(cancellationToken, sequences[0], pdbIdList, FilterProteins.RemoveSequencesOptions.RemoveSequencesInList);
break;
}
case ProteinOperation.RemoveWrongNumberOfChainsInStructure:
{
sequencesDescriptions[0] = "06 - Removed non-dimers (structure filter)";
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequences[0]);
//var pdbIdChainIdList = ProteinDataBankFileOperations.PdbIdChainIdList(sequences[0]);
pdbIdList = FilterProteins.RemoveStructuresWithIncorrectNumberOfChains(cancellationToken, pdbFilesFolders, pdbIdList, pdbIdChainIdList, 2, progressActionSet);
sequences[0] = FilterProteins.RemoveSequences(cancellationToken, sequences[0], pdbIdList, FilterProteins.RemoveSequencesOptions.RemoveSequencesInList);
break;
}
case ProteinOperation.RemoveNonInteractingProteinsInStructure:
{
// Make copy of sequences as we will split the list into two parts - with and without interactions.
sequences[1] = new List <ISequence>(sequences[0]);
// Get pdb id list from sequences, to check for pdb file, load, perform processing.
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequences[0]);
// Makes a list of sequences with interactions.
pdbIdList = FilterProteins.RemoveSequencesWithoutInteractions(cancellationToken, maxAtomInterationDistance, pdbFilesFolders, pdbIdList, pdbIdChainIdList, progressActionSet);
// Remove any protein not in the list, keep the ones in the list.
sequencesDescriptions[0] = "08 - dimers - with interactions - unfiltered for symmetry";
sequences[0] = FilterProteins.RemoveSequences(cancellationToken, sequences[0], pdbIdList, FilterProteins.RemoveSequencesOptions.RemoveSequencesNotInList);
sequencesDescriptions[1] = "07 - dimers - no observed interactions";
sequences[1] = FilterProteins.RemoveSequences(cancellationToken, sequences[1], pdbIdList, FilterProteins.RemoveSequencesOptions.RemoveSequencesInList);
break;
}
case ProteinOperation.RemoveNonSymmetricalInStructure:
{
// Make copy of sequences as we will split the list into two parts - with and without symmetry.
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequences[0]);
sequences[1] = new List <ISequence>(sequences[0]);
sequences[2] = new List <ISequence>(sequences[0]);
Dictionary <string, decimal> symmetryPercentage = FilterProteins.CalculateStructureSymmetry(cancellationToken, maxAtomInterationDistance, pdbFilesFolders, pdbIdList, pdbIdChainIdList, progressActionSet);
var pdbSymmetrical = new List <string>();
var pdbPartSymmetrical = new List <string>();
var pdbNonSymmetrical = new List <string>();
foreach (var symmetryPercentageKeyValuePair in symmetryPercentage)
{
if (symmetryPercentageKeyValuePair.Value == 0.0m)
{
pdbNonSymmetrical.Add(symmetryPercentageKeyValuePair.Key);
}
else if (symmetryPercentageKeyValuePair.Value == 100.0m)
{
pdbSymmetrical.Add(symmetryPercentageKeyValuePair.Key);
}
else if (symmetryPercentageKeyValuePair.Value > 0.0m && symmetryPercentageKeyValuePair.Value < 100.0m)
{
pdbPartSymmetrical.Add(symmetryPercentageKeyValuePair.Key);
}
else
{
ProgressActionSet.Report("Error: Out of bounds symmetry value of " + symmetryPercentageKeyValuePair.Value + " was found in " + symmetryPercentageKeyValuePair.Key + ".", progressActionSet);
}
}
sequencesDescriptions[0] = "11 - dimers - with interactions - 100% symmetrical";
sequences[0] = FilterProteins.RemoveSequences(cancellationToken, sequences[0], pdbSymmetrical, FilterProteins.RemoveSequencesOptions.RemoveSequencesNotInList);
sequencesDescriptions[1] = "10 - dimers - with interactions - 1% to 99% symmetrical";
sequences[1] = FilterProteins.RemoveSequences(cancellationToken, sequences[1], pdbPartSymmetrical, FilterProteins.RemoveSequencesOptions.RemoveSequencesNotInList);
sequencesDescriptions[2] = "09 - dimers - with interactions - 0% symmetrical";
sequences[2] = FilterProteins.RemoveSequences(cancellationToken, sequences[2], pdbNonSymmetrical, FilterProteins.RemoveSequencesOptions.RemoveSequencesNotInList);
break;
}
}
// Stop stopwatch immediately after operation.
stopwatch.Stop();
// Count sequences after operation.
afterCount = sequences[0].Count / 2;
if (!cancellationToken.IsCancellationRequested)
{
for (int sequencesIndex = sequences.GetLowerBound(0); sequencesIndex <= sequences.GetUpperBound(0); sequencesIndex++)
{
if (sequences[sequencesIndex] != null)
{
// Find free filename to save the latest sequence results of operations.
string localSaveFilename = saveFilename;
localSaveFilename = localSaveFilename.Replace("%fasta_filename%", sequencesDescriptions[sequencesIndex]);
bool skipFile = false;
if (File.Exists(localSaveFilename))
{
if (fileExistsOptions == FileExistsHandler.FileExistsOptions.AppendNumberToFilename)
{
localSaveFilename = FileExistsHandler.FindNextFreeOutputFilename(localSaveFilename);
}
else if (fileExistsOptions == FileExistsHandler.FileExistsOptions.OverwriteFile)
{
}
else if (fileExistsOptions == FileExistsHandler.FileExistsOptions.SkipFile)
{
skipFile = true;
}
}
if (!skipFile)
{
// Save the sequence results to previous set filename.
string savedFile /*s*/ = SequenceFileHandler.SaveSequencesAsFasta(sequences[sequencesIndex], localSaveFilename);
// Inform user that file has been saved.
//foreach (char savedFile in savedFiles)
//{
ProgressActionSet.Report("Saved file: " + savedFile, progressActionSet);
//}
}
}
}
// Update the user about the results.
ProgressActionSet.Report("Removed " + (beforeCount - afterCount) + " proteins. [" + afterCount + " proteins remaining]. Elapsed: " + stopwatch.Elapsed.ToString(@"dd\:hh\:mm\:ss"), progressActionSet);
}
}
if (!cancellationToken.IsCancellationRequested)
{
ProgressActionSet.Report("Finished all selected filtering operations.", progressActionSet);
}
else
{
ProgressActionSet.Report("Cancelled.", progressActionSet);
//UserProteinInterfaceOperations.ProgressBarReset(progressBar, 0, 100, 0);
////UserProteinInterfaceOperations.LabelEstimatedTimeRemainingUpdate(estimatedTimeRemaining, 0, 1, 1);
ProgressActionSet.StartAction(100, progressActionSet);
ProgressActionSet.ProgressAction(100, progressActionSet);
ProgressActionSet.FinishAction(false, progressActionSet);
}
}
public static void Upgma(List <List <decimal> > distanceMatrix, List <string> vectorNames, int minimumOutputTreeLeafs, out List <List <UpgmaNode> > nodeListList, out List <List <string> > treeListList, out List <string> finalTreeLeafOrderList, bool newickTreeEveryIteration = false, ProgressActionSet progressActionSet = null)
{
if (distanceMatrix == null || distanceMatrix.Count == 0 || distanceMatrix.Select(a => a.Count).Contains(0))
{
throw new ArgumentOutOfRangeException(nameof(distanceMatrix), "distance matrix is null or empty");
}
//var makeIterationTrees = false;
//var distanceMatrixCache = new List<decimal[,]>();
nodeListList = new List <List <UpgmaNode> >();
var nodeList = new List <UpgmaNode>();
treeListList = new List <List <string> >();
finalTreeLeafOrderList = new List <string>();
// store a list of every index merged to each index position in the distance matrix
var distanceMatrixMap = new List <List <int> >();
//var distanceMatrixMapCache = new List<List<List<int>>>();
for (var matrixIndex = 0; matrixIndex < distanceMatrix.Count; matrixIndex++)
{
distanceMatrixMap.Add(new List <int>());
distanceMatrixMap[matrixIndex].Add(matrixIndex);
var node = new UpgmaNode();
node.VectorIndexes.Add(matrixIndex);
//node.IsLeafNode = true;
node.CopyPropertiesToGenericNodeProperties();
nodeList.Add(node);
}
int itemsCompleted = 0;
int itemsTotal = distanceMatrix.Count;
var startTicks = DateTime.Now.Ticks;
ProgressActionSet.StartAction(itemsTotal, progressActionSet);
while (distanceMatrix.Count > 1 && distanceMatrix.Select(a => a.Count).Max() > 1)
{
// find which indexes to join
var lowestIndexes = UpgmaLowestDistanceIndexes(distanceMatrix);
var lowestValue = distanceMatrix[lowestIndexes.X][lowestIndexes.Y];
var lowerIndex = lowestIndexes.X < lowestIndexes.Y ? lowestIndexes.X : lowestIndexes.Y;
var higherIndex = lowestIndexes.X > lowestIndexes.Y ? lowestIndexes.X : lowestIndexes.Y;
var nodeDistance = lowestValue / 2;
if (lowerIndex == higherIndex)
{
throw new Exception("lower index and higher index have the same value");
}
// Take parent node identities
var parentNodeIdA = distanceMatrixMap[lowerIndex].OrderBy(o => o).ToList();
var parentNodeIdB = distanceMatrixMap[higherIndex].OrderBy(o => o).ToList();
var childNodeId = new List <int>();
childNodeId.AddRange(parentNodeIdA);
childNodeId.AddRange(parentNodeIdB);
childNodeId = childNodeId.Distinct().ToList();
// Find if parent nodes already exist
var parentNodeCandidatesA = nodeList.Where(a => a.VectorIndexes.OrderBy(o => o).SequenceEqual(parentNodeIdA)).ToList();
var parentNodeCandidatesB = nodeList.Where(a => a.VectorIndexes.OrderBy(o => o).SequenceEqual(parentNodeIdB)).ToList();
UpgmaNode parentNodeA;
parentNodeA = parentNodeCandidatesA[0];
UpgmaNode parentNodeB;
parentNodeB = parentNodeCandidatesB[0];
var childNode = new UpgmaNode();
nodeList.Add(childNode);
parentNodeA.ChildNode = childNode;
parentNodeB.ChildNode = childNode;
childNode.ParentNodeA = parentNodeA;
childNode.ParentNodeB = parentNodeB;
var parentTotalDistanceA = DistanceNodeToFinalParent(parentNodeA);
var parentTotalDistanceB = DistanceNodeToFinalParent(parentNodeB);
parentNodeA.DistanceChildNode = nodeDistance - parentTotalDistanceA;
parentNodeB.DistanceChildNode = nodeDistance - parentTotalDistanceB;
childNode.DistanceParentNodeA = nodeDistance - parentTotalDistanceA;
childNode.DistanceParentNodeB = nodeDistance - parentTotalDistanceB;
childNode.DistanceMatrixIterationNumber = itemsCompleted;// distanceMatrixCache.Count - 1;
childNode.VectorIndexes = childNodeId;
parentNodeA.CopyPropertiesToGenericNodeProperties();
parentNodeB.CopyPropertiesToGenericNodeProperties();
childNode.CopyPropertiesToGenericNodeProperties();
// rearrange the matrix map with the new indexes joined distanceMatrixMap[higherIndex]
distanceMatrixMap[lowerIndex].AddRange(distanceMatrixMap[higherIndex]);
distanceMatrixMap.RemoveAt(higherIndex);
// recalculate distance matrix with indexes joined with mean average
distanceMatrix = UpgmaDistanceMatrixNextIteration(distanceMatrix, lowestIndexes);
if (newickTreeEveryIteration)
{
List <string> treeLeafOrderList;
var iterationTree = Newick.NewickTreeFormat(nodeList.ToList <GenericNode>(), vectorNames, out treeLeafOrderList, minimumOutputTreeLeafs);
treeListList.Add(iterationTree);
finalTreeLeafOrderList = treeLeafOrderList;
nodeListList.Add(UpgmaNode.CopyNetwork(nodeList));
}
itemsCompleted++;
ProgressActionSet.ProgressAction(1, progressActionSet);
ProgressActionSet.EstimatedTimeRemainingAction(startTicks, itemsCompleted, itemsTotal, progressActionSet);
}
if (!newickTreeEveryIteration)
{
List <string> treeLeafOrderList;
var iterationTree = Newick.NewickTreeFormat(nodeList.ToList <GenericNode>(), vectorNames, out treeLeafOrderList, minimumOutputTreeLeafs);
treeListList.Add(iterationTree);
finalTreeLeafOrderList = treeLeafOrderList;
nodeListList.Add(UpgmaNode.CopyNetwork(nodeList));
}
ProgressActionSet.FinishAction(true, progressActionSet);
}
/*
* public static void ClusterVectorDistanceMatrixUpgma(List<VectorProteinInterfaceWhole> vectorProteinInterfaceWholeList, decimal[,] vectorDistanceMatrix, int minimumOutputTreeLeafs, out List<string> vectorNames, out List<List<UpgmaNode>> nodeList, out List<List<string>> treeList, ProgressActionSet progressActionSet)
* {
* if (vectorProteinInterfaceWholeList == null) throw new ArgumentNullException(nameof(vectorProteinInterfaceWholeList));
* if (vectorDistanceMatrix == null) throw new ArgumentNullException(nameof(vectorDistanceMatrix));
*
* vectorNames = vectorProteinInterfaceWholeList.Select(VectorProteinInterfaceWholeTreeHeader).ToList();
*
* List<string> finalTreeLeafOrderList;
* UpgmaClustering.Upgma(vectorDistanceMatrix, vectorNames, minimumOutputTreeLeafs, out nodeList, out treeList, out finalTreeLeafOrderList, false, progressActionSet);
* }
*/
public static void BestDistanceMatrixWithPartsAlignment(CancellationToken cancellationToken, List <VectorProteinInterfaceWhole> vectorProteinInterfaceWholeList, VectorDistanceMeasurementValues vectorDistanceMeasurementValues, out double[,] optimisticDistanceMatrix, /* out double[,] pessimisticDistanceMatrix,*/ ProgressActionSet progressActionSet)
{
if (vectorProteinInterfaceWholeList == null)
{
throw new ArgumentNullException(nameof(vectorProteinInterfaceWholeList));
}
if (vectorDistanceMeasurementValues == null)
{
throw new ArgumentNullException(nameof(vectorDistanceMeasurementValues));
}
var totalVectors = vectorProteinInterfaceWholeList.Count;
var optimisticDistanceMatrix2 = new double[totalVectors, totalVectors];
//var pessimisticDistanceMatrix2 = new double[totalVectors, totalVectors];
var workDivision = new WorkDivision(vectorProteinInterfaceWholeList.Count, -1);
ProgressActionSet.StartAction(vectorProteinInterfaceWholeList.Count, progressActionSet);
for (int threadIndex = 0; threadIndex < workDivision.ThreadCount; threadIndex++)
{
int localThreadIndex = threadIndex;
var task = Task.Run(() =>
{
for (int indexX = workDivision.ThreadFirstIndex[localThreadIndex]; indexX <= workDivision.ThreadLastIndex[localThreadIndex]; indexX++)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
var vectorProteinInterfaceWholeX = vectorProteinInterfaceWholeList[indexX];
for (int indexY = 0; indexY < vectorProteinInterfaceWholeList.Count; indexY++)
{
if (indexX >= indexY)
{
continue;
}
var vectorProteinInterfaceWholeY = vectorProteinInterfaceWholeList[indexY];
if (vectorProteinInterfaceWholeX.FullProteinInterfaceId == vectorProteinInterfaceWholeY.FullProteinInterfaceId)
{
continue;
}
double optimisticDistance;
//double pessimisticDistance;
BestDistanceWithPartsAlignment(vectorProteinInterfaceWholeX, vectorProteinInterfaceWholeY, vectorDistanceMeasurementValues, out optimisticDistance /*, out pessimisticDistance*/);
var lengthDifference = Math.Abs(vectorProteinInterfaceWholeX.ProteinInterfaceLength - vectorProteinInterfaceWholeY.ProteinInterfaceLength);
var lengthDistance = lengthDifference * vectorDistanceMeasurementValues.DifferentLengthProteinInterface;
optimisticDistance += lengthDistance;
//pessimisticDistance += lengthDistance;
optimisticDistanceMatrix2[indexX, indexY] = optimisticDistance;
//pessimisticDistanceMatrix2[indexX, indexY] = pessimisticDistance;
optimisticDistanceMatrix2[indexY, indexX] = optimisticDistance;
//pessimisticDistanceMatrix2[indexY, indexX] = pessimisticDistance;
}
workDivision.IncrementItemsCompleted(1);
ProgressActionSet.ProgressAction(1, progressActionSet);
ProgressActionSet.EstimatedTimeRemainingAction(workDivision.StartTicks, workDivision.ItemsCompleted, workDivision.ItemsToProcess, progressActionSet);
}
}, cancellationToken);
workDivision.TaskList.Add(task);
}
workDivision.WaitAllTasks();
ProgressActionSet.FinishAction(true, progressActionSet);
optimisticDistanceMatrix = optimisticDistanceMatrix2;
//pessimisticDistanceMatrix = pessimisticDistanceMatrix2;
}
/// <summary>
/// This method returns a dictionary entry for each protein id (pdb id), with a list of interaction vectors
/// </summary>
/// <returns></returns>
public static List <VectorProteinInterfaceWhole> LoadProteinInterfaceVectorFromFiles(
CancellationToken cancellationToken,
decimal maxAtomInterationDistance,
decimal minimumProteinInterfaceDensity,
string[] sequenceListFileArray,
string[] pdbFileDirectoryLocationArray,
ProgressActionSet progressActionSet)
{
if (sequenceListFileArray == null)
{
throw new ArgumentNullException(nameof(sequenceListFileArray));
}
if (pdbFileDirectoryLocationArray == null)
{
throw new ArgumentNullException(nameof(pdbFileDirectoryLocationArray));
}
var vectorProteinInterfaceWholeList = new List <VectorProteinInterfaceWhole>();
// 1: Open list of sequences already cleaned to have only symmetrical homodimers (fasta file only contains 100% symmetrical homodimers with all other junk removed - but could have any number of proteinInterfaces per chain)
List <ISequence> sequenceList = SequenceFileHandler.LoadSequenceFileList(sequenceListFileArray, StaticValues.MolNameProteinAcceptedValues);
var pdbIdChainIdList = ProteinDataBankFileOperations.PdbIdChainIdList(sequenceList);
// 2: Get a list of the unique ids for the sequences
List <string> pdbIdList = FilterProteins.SequenceListToPdbIdList(sequenceList);
if (pdbIdList == null || pdbIdList.Count == 0)
{
throw new ArgumentOutOfRangeException(nameof(sequenceListFileArray), "Error loading PDB ID list");
}
// 3: Get a list of PDB files found in user specified directory
string[] pdbFilesArray = ProteinDataBankFileOperations.GetPdbFilesArray(pdbFileDirectoryLocationArray);
ProgressActionSet.StartAction(pdbFilesArray.Length, progressActionSet);
var startTicks = DateTime.Now.Ticks;
// 4: Loop through each pdb file
for (int pdbFileNumber = 0; pdbFileNumber < pdbFilesArray.Length; pdbFileNumber++) // +1 is for progress update
{
ProgressActionSet.ProgressAction(1, progressActionSet);
ProgressActionSet.EstimatedTimeRemainingAction(startTicks, pdbFileNumber + 1, pdbFilesArray.Length, progressActionSet);
// get unique id of pdb file
string pdbFilename = pdbFilesArray[pdbFileNumber];
string proteinId = ProteinDataBankFileOperations.PdbIdFromPdbFilename(pdbFilename);
// check pdb unique id was in the loaded sequence list
if (!pdbIdList.Contains(proteinId))
{
continue;
}
ClusterProteinDataBankFileResult clusterPdbFileResult = Clustering.ClusterProteinDataBankFile(cancellationToken, maxAtomInterationDistance, minimumProteinInterfaceDensity, pdbFilename, pdbIdChainIdList, ClusteringMethodOptions.ClusterWithResidueSequenceIndex, -1, -1, progressActionSet);
if (clusterPdbFileResult == null)
{
continue;
}
List <ProteinInterfaceSequenceAndPositionData> proteinInterfaceSequenceAndPositionDataList = clusterPdbFileResult.ProteinInterfaceAnalysisResultData.ProteinInterfacesSequenceAndPositionDataList;
proteinInterfaceSequenceAndPositionDataList = proteinInterfaceSequenceAndPositionDataList.OrderBy(a => a.FullProteinInterfaceId.ProteinId).ThenBy(a => a.FullProteinInterfaceId.ChainId).ThenBy(a => a.FullProteinInterfaceId.ProteinInterfaceId).ToList();
for (int proteinInterfaceSequenceAndPositionDataListIndex = 0; proteinInterfaceSequenceAndPositionDataListIndex < proteinInterfaceSequenceAndPositionDataList.Count; proteinInterfaceSequenceAndPositionDataListIndex++)
{
ProteinInterfaceSequenceAndPositionData proteinInterfaceSequenceAndPositionData = proteinInterfaceSequenceAndPositionDataList[proteinInterfaceSequenceAndPositionDataListIndex];
var seq = sequenceList.FirstOrDefault(a =>
{
var p = SequenceIdSplit.SequenceIdToPdbIdAndChainId(a.ID);
return(p.PdbId.ToUpperInvariant() == proteinInterfaceSequenceAndPositionData.FullProteinInterfaceId.ProteinId.ToUpperInvariant() && p.ChainId.ToUpperInvariant() == proteinInterfaceSequenceAndPositionData.ChainIdLetter.ToUpperInvariant());
});
var seqLen = seq != null ? seq.Count : -1;
var vectorProteinInterfaceWholeFwd = MakeVectorProteinInterfaceWhole(pdbFilename, proteinInterfaceSequenceAndPositionData, false, false);
vectorProteinInterfaceWholeFwd.FullSequenceLength = seqLen;
vectorProteinInterfaceWholeList.Add(vectorProteinInterfaceWholeFwd);
var vectorProteinInterfaceWholeRev = MakeVectorProteinInterfaceWhole(pdbFilename, proteinInterfaceSequenceAndPositionData, true, false);
vectorProteinInterfaceWholeRev.FullSequenceLength = seqLen;
vectorProteinInterfaceWholeList.Add(vectorProteinInterfaceWholeRev);
}
}
ProgressActionSet.FinishAction(true, progressActionSet);
vectorProteinInterfaceWholeList = vectorProteinInterfaceWholeList.OrderBy(a => a.FullProteinInterfaceId.ProteinId).ThenBy(a => a.FullProteinInterfaceId.ChainId).ThenBy(a => a.FullProteinInterfaceId.ProteinInterfaceId).ToList();
return(vectorProteinInterfaceWholeList);
}