// Creates a subsequence from a source sequence given the settings provided
private ISequence CreateSubsequence(SimulatorSettings settings, int index)
{
double err = (double)settings.ErrorFrequency;
// Set the length using the appropriate random number distribution type
int subLength = settings.SequenceLength;
if (settings.DistributionType == (int)Distribution.Uniform)
{
subLength += random.Next(settings.LengthVariation * 2) - settings.LengthVariation;
}
else if (settings.DistributionType == (int)Distribution.Normal)
{
subLength = (int)MBF.Util.Helper.GetNormalRandom((double)settings.SequenceLength,
(double)settings.LengthVariation);
}
// Quick sanity checks on the length of the subsequence
if (subLength <= 0)
{
subLength = 1;
}
if (subLength > Sequence.Count)
{
subLength = Sequence.Count;
}
// Set the start position
int startPosition = random.Next(Sequence.Count - subLength);
Sequence result = new Sequence(Sequence.Alphabet);
result.IsReadOnly = false;
List <ISequenceItem> errorSource = Sequence.Alphabet.LookupAll(true, false, settings.AllowAmbiguities, false);
for (int i = 0; i < subLength; i++)
{
// Apply Errors if applicable
if (random.NextDouble() < err)
{
result.Add(errorSource[random.Next(errorSource.Count - 1)]);
}
else
{
result.Add(Sequence[startPosition + i]);
}
}
result.ID = Sequence.ID + " (Split " + (index + 1) + ", " + result.Count + "bp)";
// Reverse Sequence if applicable
if (settings.ReverseHalf && random.NextDouble() < 0.5f)
{
return(result.Reverse);
}
return(result);
}
/// <summary>
/// Does the logic behind the sequence simulation
/// </summary>
internal void DoSimulation(SimulatorWindow window, string outputFileName, SimulatorSettings settings)
{
FileInfo file = new FileInfo(outputFileName);
if (!file.Directory.Exists)
{
throw new ArgumentException("Could not write to the output directory for " + outputFileName);
}
if (settings.OutputSequenceCount <= 0)
{
throw new ArgumentException("'Max Output Sequences Per File' should be greater than zero.");
}
if (settings.SequenceLength <= 0)
{
throw new ArgumentException("'Mean Output Length' should be greater than zero.");
}
string filePrefix;
if (String.IsNullOrEmpty(file.Extension))
{
filePrefix = file.FullName;
}
else
{
filePrefix = file.FullName.Substring(0, file.FullName.IndexOf(file.Extension));
}
string filePostfix = "_{0}.fa";
long seqCount = (settings.DepthOfCoverage * SequenceToSplit.Count) / settings.SequenceLength;
long fileCount = seqCount / settings.OutputSequenceCount;
if (seqCount % settings.OutputSequenceCount != 0)
{
fileCount++;
}
window.UpdateSimulationStats(seqCount, fileCount);
if (generatedSequenceList == null)
{
generatedSequenceList = new List <ISequence>();
}
else
{
generatedSequenceList.Clear();
}
int fileIndex = 1;
FastAFormatter formatter = null;
for (long i = 0; i < seqCount; i++)
{
generatedSequenceList.Add(CreateSubsequence(settings, i));
if (generatedSequenceList.Count >= settings.OutputSequenceCount)
{
FileInfo outFile = new FileInfo(filePrefix + string.Format(filePostfix, fileIndex++));
formatter = new FastAFormatter(outFile.FullName);
foreach (ISequence seq in generatedSequenceList)
{
formatter.Write(seq);
}
formatter.Close();
generatedSequenceList.Clear();
}
}
if (generatedSequenceList.Count > 0)
{
FileInfo outFile = new FileInfo(filePrefix + string.Format(filePostfix, fileIndex++));
formatter = new FastAFormatter(outFile.FullName);
foreach (ISequence seq in generatedSequenceList)
{
formatter.Write(seq);
}
formatter.Close();
window.NotifySimulationComplete(formatter.Name);
}
else
{
window.NotifySimulationComplete(string.Empty);
}
}
// Creates a subsequence from a source sequence given the settings provided
private ISequence CreateSubsequence(SimulatorSettings settings, long index)
{
double err = (double)settings.ErrorFrequency;
// Set the length using the appropriate random number distribution type
long subLength = settings.SequenceLength;
if (settings.DistributionType == (int)Distribution.Uniform)
{
subLength += random.Next(settings.LengthVariation * 2) - settings.LengthVariation;
}
else if (settings.DistributionType == (int)Distribution.Normal)
{
subLength = (long)Math.Floor(Bio.Util.Helper.GetNormalRandom((double)settings.SequenceLength,
(double)settings.LengthVariation));
}
// Quick sanity checks on the length of the subsequence
if (subLength <= 0)
{
subLength = 1;
}
if (subLength > SequenceToSplit.Count)
{
subLength = SequenceToSplit.Count;
}
// Set the start position
long startPosition = (long)Math.Floor(random.NextDouble() * (SequenceToSplit.Count - subLength));
byte[] sequenceBytes = new byte[subLength];
IAlphabet resultSequenceAlphabet = SequenceToSplit.Alphabet;
// Get ambiguity symbols
List <byte> errorSource = null;
//= Sequence.Alphabet.LookupAll(true, false, settings.AllowAmbiguities, false);
if (settings.AllowAmbiguities &&
(SequenceToSplit.Alphabet == DnaAlphabet.Instance || SequenceToSplit.Alphabet == RnaAlphabet.Instance || SequenceToSplit.Alphabet == ProteinAlphabet.Instance)
)
{
resultSequenceAlphabet = Alphabets.AmbiguousAlphabetMap[SequenceToSplit.Alphabet];
}
errorSource = resultSequenceAlphabet.GetValidSymbols().ToList();
// remove gap and termination symbol
HashSet <byte> gaps, terminations;
SequenceToSplit.Alphabet.TryGetGapSymbols(out gaps);
SequenceToSplit.Alphabet.TryGetTerminationSymbols(out terminations);
if (gaps != null)
{
errorSource.RemoveAll(a => gaps.Contains(a));
}
if (terminations != null)
{
errorSource.RemoveAll(a => terminations.Contains(a));
}
for (long i = 0; i < subLength; i++)
{
// Apply Errors if applicable
if (random.NextDouble() < err)
{
sequenceBytes[i] = errorSource[random.Next(errorSource.Count - 1)];
}
else
{
sequenceBytes[i] = SequenceToSplit[startPosition + i];
}
}
Sequence generatedSequence = new Sequence(resultSequenceAlphabet, sequenceBytes.ToArray());
generatedSequence.ID = SequenceToSplit.ID + " (Split " + (index + 1) + ", " + generatedSequence.Count + "bp)";
// Reverse Sequence if applicable
if (settings.ReverseHalf && random.NextDouble() < 0.5f)
{
return(new DerivedSequence(generatedSequence, true, false));
}
return(generatedSequence);
}
/// <summary>
/// Constructor
/// </summary>
public SimulatorController()
{
_seqRandom = new Random();
Settings = new SimulatorSettings();
}
/// <summary>
/// Creates a subsequence from a source sequence given the settings provided
/// </summary>
/// <param name="index"></param>
/// <param name="sequenceToSplit"></param>
/// <param name="simulatorSettings"></param>
/// <returns></returns>
private ISequence CreateSubsequence(long index, ISequence sequenceToSplit, SimulatorSettings simulatorSettings)
{
double err = simulatorSettings.ErrorFrequency;
// Set the length using the appropriate random number distribution type
long subLength = simulatorSettings.SequenceLength;
switch (simulatorSettings.DistributionType)
{
case (int)Distribution.Uniform:
subLength += _seqRandom.Next(simulatorSettings.LengthVariation * 2) - simulatorSettings.LengthVariation;
break;
case (int)Distribution.Normal:
subLength = (long)Math.Floor(Bio.Util.Helper.GetNormalRandom(simulatorSettings.SequenceLength, simulatorSettings.LengthVariation));
break;
}
// Quick sanity checks on the length of the subsequence
if (subLength <= 0)
{
subLength = 1;
}
if (subLength > sequenceToSplit.Count)
{
subLength = sequenceToSplit.Count;
}
// Set the start position
long startPosition = (long)Math.Floor(_seqRandom.NextDouble() * (sequenceToSplit.Count - subLength));
byte[] sequenceBytes = new byte[subLength];
IAlphabet resultSequenceAlphabet = sequenceToSplit.Alphabet;
// Get ambiguity symbols
if (simulatorSettings.AllowAmbiguities &&
(sequenceToSplit.Alphabet == DnaAlphabet.Instance || sequenceToSplit.Alphabet == RnaAlphabet.Instance ||
sequenceToSplit.Alphabet == ProteinAlphabet.Instance))
{
resultSequenceAlphabet = Alphabets.AmbiguousAlphabetMap[sequenceToSplit.Alphabet];
}
List <byte> errorSource = resultSequenceAlphabet.GetValidSymbols().ToList();
// remove gap and termination symbol
HashSet <byte> gaps, terminations;
sequenceToSplit.Alphabet.TryGetGapSymbols(out gaps);
sequenceToSplit.Alphabet.TryGetTerminationSymbols(out terminations);
if (gaps != null)
{
errorSource.RemoveAll(a => gaps.Contains(a));
}
if (terminations != null)
{
errorSource.RemoveAll(a => terminations.Contains(a));
}
for (long i = 0; i < subLength; i++)
{
// Apply Errors if applicable
sequenceBytes[i] = _seqRandom.NextDouble() < err
? errorSource[_seqRandom.Next(errorSource.Count - 1)]
: sequenceToSplit[startPosition + i];
}
ISequence generatedSequence = new Sequence(resultSequenceAlphabet, sequenceBytes.ToArray());
generatedSequence.ID = sequenceToSplit.ID + " (Split " + (index + 1) + ", " + generatedSequence.Count + "bp)";
// Reverse Sequence if applicable
return(simulatorSettings.ReverseHalf && _seqRandom.NextDouble() < 0.5f
? new DerivedSequence(generatedSequence, true, true)
: generatedSequence);
}
