public void LoadProteins(string fastaFile)
{
using (FastaReader reader = new FastaReader(fastaFile))
{
LoadProteins(reader.ReadNextProtein());
}
}
public static void Start(IProtease protease, double percentIdentified = 0.05, int maxMissed = 3, int minLength = 5, int maxLength = 35)
{
Console.WriteLine("**Start Protein Grouping**");
Stopwatch watch = new Stopwatch();
watch.Start();
List<Peptide> peps = new List<Peptide>();
List<Protein> proteins = new List<Protein>();
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
foreach (Protein protein in reader.ReadNextProtein())
{
foreach (Peptide peptide in protein.Digest(protease, maxMissed, minLength, maxLength))
{
peps.Add(peptide);
}
proteins.Add(protein);
}
}
Console.WriteLine("Loaded {0:N0} peptides from {1:N0} proteins in {2} ms", peps.Count, proteins.Count, watch.ElapsedMilliseconds);
// Fixed seed to make it reproducible
Random random = new Random(480912341);
// Take the first x % to act as our identified peptides
List<Peptide> identifiedPeptides = peps.OrderBy(x => random.Next()).Take((int) (peps.Count*percentIdentified)).ToList();
List<ProteinGroup> proteinGroups = ProteinGroup.GroupProteins(proteins, protease, identifiedPeptides, new AminoAcidLeucineSequenceComparer(), maxMissed).ToList();
watch.Stop();
Console.WriteLine("{0:N0} proteins produced {1:N0} protein groups from {2:N0} identified sequences", proteins.Count, proteinGroups.Count, identifiedPeptides.Count);
Console.WriteLine("Time elapsed: {0}", watch.Elapsed);
Console.WriteLine("Memory used: {0:N0} MB", System.Environment.WorkingSet/(1024*1024));
Console.WriteLine("**END Protein Grouping**");
}
public static void Start(IProtease protease, int maxMissed = 3, int minLength = 5, int maxLength = 35)
{
Console.WriteLine("**Start Morpheus Search**");
Stopwatch watch = new Stopwatch();
watch.Start();
List<int> hashCodes = new List<int>();
// Generate peptide candidates
HashSet<Peptide> peptides = new HashSet<Peptide>();
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
foreach (Protein protein in reader.ReadNextProtein())
{
foreach (Peptide peptide in protein.Digest(protease, maxMissed, minLength, maxLength))
{
peptides.Add(peptide);
}
}
}
MSSearchEngine engine = new MorpheusSearchEngine();
engine.PrecursorMassTolerance = Tolerance.FromPPM(100);
engine.ProductMassTolerance = Tolerance.FromPPM(10);
engine.LoadPeptides(peptides);
watch.Stop();
Console.WriteLine("Time elapsed: {0}", watch.Elapsed);
Console.WriteLine("Memory used: {0:N0} MB", Environment.WorkingSet/(1024*1024));
Console.WriteLine("**End Morpheus Search**");
}
public static void ExampleProteinGrouping(IProtease protease, double percentIdentified = 0.01, int maxMissed = 3, int minLength = 5, int maxLength = 50)
{
Stopwatch watch = new Stopwatch();
watch.Start();
List<Peptide> peps = new List<Peptide>(1000000);
List<Protein> proteins = new List<Protein>(7000);
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
foreach (Protein protein in reader.ReadNextProtein())
{
peps.AddRange(protein.Digest(protease, maxMissed, minLength, maxLength));
proteins.Add(protein);
}
}
Console.WriteLine("Loaded {0:N0} peptides from {1:N0} proteins in {2} ms", peps.Count, proteins.Count, watch.ElapsedMilliseconds);
watch.Restart();
Random random = new Random(480912341);
// Take the first x % to act as our identified peptides
List<Peptide> identifiedPeptides = peps.OrderBy(x => random.Next()).Take((int) (peps.Count*percentIdentified)).ToList();
List<ProteinGroup> proteinGroups = ProteinGroup.GroupProteins(proteins, protease, identifiedPeptides, new AminoAcidLeucineSequenceComparer(), maxMissed).ToList();
watch.Stop();
Console.WriteLine("{0:N0} proteins produced {1:N0} protein groups from {2:N0} identified sequences", proteins.Count, proteinGroups.Count, identifiedPeptides.Count);
Console.WriteLine();
Console.WriteLine("Time elapsed: {0} ms", watch.ElapsedMilliseconds);
}
public static void Start(IProtease protease, int maxMissed = 1, int minLength = 0, int maxLength = int.MaxValue, bool storeSequenceString = true)
{
Console.WriteLine("**Start Digestion**");
Stopwatch watch = new Stopwatch();
watch.Start();
List<Peptide> peps = new List<Peptide>();
List<Protein> prots = new List<Protein>();
List<double> allMzs = new List<double>();
AminoAcidPolymer.StoreSequenceString = storeSequenceString;
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
foreach (Protein protein in reader.ReadNextProtein())
{
foreach (Peptide peptide in protein.Digest(protease, maxMissed, minLength, maxLength))
{
peps.Add(peptide);
allMzs.Add(peptide.ToMz(2)); // forces the calculation of the mass and thus chemical formula
}
prots.Add(protein);
}
}
watch.Stop();
Console.WriteLine("{0:N0} proteins produced {1:N0} peptides using {2:N0} missed cleavages", prots.Count, peps.Count, maxMissed);
Console.WriteLine("Time elapsed: {0}", watch.Elapsed);
Console.WriteLine("Memory used: {0:N0} MB", System.Environment.WorkingSet / (1024 * 1024));
Console.WriteLine("**End Digestion**");
}
public static void BasicCsvWriting(string outputFilePath)
{
Console.WriteLine("Writing file to: " + outputFilePath);
// Create a stream writer to output data to a stream. In this case, the stream points to a file path on the
// computer and is saved as a file. The using statement is the same as doing the following:
// Stream writer = new StreamWriter(outputFilePath);
// writer.Open();
// writer.WriteLine("Hello World");
// writer.Close();
using (StreamWriter writer = new StreamWriter(outputFilePath))
{
// Now that the stream is open, we can a line of text to it to serve as the header row.
// CSV formats are just text files with fields seperated by commas.
writer.WriteLine("Protein Name,# of Amino Acids,Mass (da)");
// Open a connection to a fasta file (very similar syntax as the StreamWriter above)
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
// Loop over each protein in the fasta file
foreach (Protein protein in reader.ReadNextProtein())
{
// StringBuilder objects are an effective tool for constructing strings to write to files.
StringBuilder sb = new StringBuilder();
// To add items to the string builder, just call the append method with whatever you want to add
sb.Append(protein.Description);
sb.Append(','); // we need to add the delimiter after each field we add as well.
// Add the next item
sb.Append(protein.Length);
sb.Append(',');
// The append method can take any object, it will simply call the .ToString() method on the object supplied.
sb.Append(protein.MonoisotopicMass);
// No delimiter is needed after the last field is added.
// Now to write this string to the file itself.
// We convert the StringBuilder object (named: sb) to a string, and then write it on its own line in the
// file writer (named: writer)
writer.WriteLine(sb.ToString());
}
}
} // The file is automatically written and closed after exiting the using {} block.
}
public static void Start(IProtease protease, int maxMissed = 3, int minLength = 5, int maxLength = 35)
{
Console.WriteLine("**Start Morpheus Search**");
Stopwatch watch = new Stopwatch();
watch.Start();
List<int> hashCodes = new List<int>();
// Generate peptide candidates
HashSet<Peptide> peptides = new HashSet<Peptide>();
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
foreach (Protein protein in reader.ReadNextProtein())
{
foreach (Peptide peptide in protein.Digest(protease, maxMissed, minLength, maxLength))
{
peptides.Add(peptide);
}
}
}
MSSearchEngine engine = new MorpheusSearchEngine();
engine.PrecursorMassTolerance = Tolerance.FromPPM(100);
engine.ProductMassTolerance = Tolerance.FromPPM(10);
engine.LoadPeptides(peptides);
using (MSDataFile msDataFile = new ThermoRawFile("Resources/ThermoRawFileMS1MS2.raw"))
{
//SortedMaxSizedContainer<PeptideSpectralMatch> psms = engine.Search(msDataFile.Where(scan => scan.MsnOrder > 1));
//foreach (MSDataScan scan in msDataFile.Where(scan => scan.MsnOrder > 1))
//{
// List<PeptideSpectralMatch> psms = engine.Search(scan);
// Console.WriteLine("{0} {1}", scan.SpectrumNumber, psms.Count);
//}
}
watch.Stop();
Console.WriteLine("Time elapsed: {0}", watch.Elapsed);
Console.WriteLine("Memory used: {0:N0} MB", System.Environment.WorkingSet / (1024 * 1024));
Console.WriteLine("**End Morpheus Search**");
}
public static void ExampleDigestion()
{
const string fastaFilePath = "Resources/yeast_uniprot_120226.fasta";
IProtease trypsin = Protease.GetProtease("Trypsin");
const int maxMissedCleavages = 3;
const int minPeptideLength = 5;
const int maxPeptideLength = 50;
List<double> masses = new List<double>();
Stopwatch watch = new Stopwatch();
watch.Start();
using (FastaReader reader = new FastaReader(fastaFilePath))
{
foreach (Protein protein in reader.ReadNextProtein())
{
foreach (Peptide peptide in protein.Digest(trypsin, maxMissedCleavages, minPeptideLength, maxPeptideLength))
{
masses.Add(peptide.MonoisotopicMass);
}
}
}
//Console.WriteLine("Average Peptide Mass = {0:F4}", masses.Average());
watch.Stop();
Console.WriteLine("Time elapsed: {0}", watch.Elapsed);
}
public void LoadProteins(string fastaFile)
{
using (FastaReader reader = new FastaReader(fastaFile))
{
LoadProteins(reader.ReadNextProtein());
}
}
/// <summary>
/// Performs an in silico digestion of all the proteins found within the fasta file.
/// </summary>
/// <param name="fastaFile">The fasta filename to perfrom the digestion on</param>
/// <param name="uniquePeptides">The unique peptides that were read in from the csv files</param>
/// <param name="proteases"></param>
/// <param name="semiDigestion">Perform a Semi Digestion</param>
/// <returns>True if all the unique peptides get isMapped to at least one protein, false otherwise</returns>
private List<Protein> GetMappedProteinsFromFasta(string fastaFile, Dictionary<string, Peptide> uniquePeptides, IList<Protease> proteases, bool semiDigestion = false)
{
string fastaFileniceName = Path.GetFileName(fastaFile);
StringBuilder sb = new StringBuilder();
foreach (Protease protease in proteases)
{
sb.Append(protease.Name);
sb.Append(',');
}
if (sb.Length > 0) { sb.Remove(sb.Length - 1, 1); }
Log("Performing {0}{1} digestion on {2}...", semiDigestion ? "semi " : "", sb, fastaFileniceName);
//Peptide.MappedCount = 0;
int forwardProteins = 0, decoyProteins = 0, forwardProteinsMapped = 0, decoyProteinsMapped = 0, fastaCounter = 0, pepsMapped = 0;
long totalBytes = new FileInfo(fastaFile).Length;
// A hashset of all proteins that have a peptide that was in the input files
Dictionary<Protein, Protein> proteins = new Dictionary<Protein, Protein>(1 << 13);
// Min and Max length of peptides
int minLength = semiDigestion ? 1 : _smallestPeptide - 1;
int maxLength = semiDigestion ? int.MaxValue : _largestPeptide + 1;
// Open the reader for the protein database in the .fasta format
using (FastaReader reader = new FastaReader(fastaFile))
{
// Read in each protein one-by-one
foreach (Fasta fasta in reader.ReadNextFasta())
{
// The number of fasta (proteins) read in (for progress bar feedback)
fastaCounter++;
// Create a new protein from the fasta
Protein prot = new Protein(fasta.Description, fasta.Sequence);
// Check if the protein is a decoy protein or not
if (prot.IsDecoy)
{
decoyProteins++;
}
else
{
forwardProteins++;
}
// Loop over each protease
foreach (Protease protease in proteases)
{
// Digest the protein's leucine sequences (all I's are now L's) with the given proteases, max missed cleavages, limiting it to the smallest and largest peptide observed (speed improvement)
// *Note each peptide sequence (pep_seq) will be leucine sequences as well
foreach (string pepSeq in AminoAcidPolymer.Digest(prot.Sequence, protease, MaxMissedCleavages, minLength, maxLength, semiDigestion: semiDigestion))
{
// Is this one of the unique peptide sequences in the csv files? If not, we don't care about it
Peptide pep;
if (!uniquePeptides.TryGetValue(pepSeq.Replace('I', 'L'), out pep))
continue;
// Check to see if this protein has already been added to the list of proteins hit
if(!proteins.ContainsKey(prot)) // returns true if the protein is new to the hashset of proteins
{
proteins.Add(prot, prot);
if (prot.IsDecoy)
{
decoyProteinsMapped++;
}
else
{
forwardProteinsMapped++;
}
}
// Add the peptide to the protein (internally hashed, so don't worry about duplicates)
prot.AddPeptide(pep);
// Mark that this peptide was successfully mapped, this is for error checking purposes
if (!pep.IsMapped)
{
pepsMapped++;
pep.IsMapped = true;
}
//pep.MarkAsMapped();
}
}
// Only call every 100 proteins otherwise you are wasting a lot of time refreshing and not doing actual work
//if (fastaCounter > 100)
//{
// fastaCounter = 0;
// ProgressUpdate((double)reader.Position / totalBytes);
//}
}
}
// Check to see if every peptide is matched, if not try using a brute force search method instead
if (uniquePeptides.Count > pepsMapped)
{
// Get all the peptides that weren't mapped
List<Peptide> unMapedPeptides = uniquePeptides.Values.Where(p => !p.IsMapped).ToList();
Log("[WARNING] Couldn't find every peptide using digestion method (wrong enzyme perhaps?), trying brute force search instead on the remaining {0} peptides...", unMapedPeptides.Count);
ProgressUpdate(0.1);
using (FastaReader reader = new FastaReader(fastaFile))
{
fastaCounter = 0;
// Read in each protein one-by-one
foreach (Fasta fasta in reader.ReadNextFasta())
{
string seq = fasta.Sequence.Replace('I', 'L');
foreach (Peptide pep2 in unMapedPeptides)
{
if (!seq.Contains(pep2.LeucineSequence))
continue;
Protein prot = new Protein(fasta.Description, fasta.Sequence);
Protein realProt;
if (proteins.TryGetValue(prot, out realProt))
{
// Add the peptide to the protein
realProt.AddPeptide(pep2);
}
else
{
proteins.Add(prot, prot);
if (prot.IsDecoy)
{
decoyProteinsMapped++;
}
else
{
forwardProteinsMapped++;
}
// Add the peptide to the protein
prot.AddPeptide(pep2);
}
// Mark that this peptide was successfully isMapped, this is for error checking purposes
pep2.IsMapped = true;
//pep2.MarkAsMapped();
}
fastaCounter++;
//// Only call every 100 proteins otherwise you are wasting a lot of time refreshing and not doing actual work
//if (fastaCounter > 100)
//{
// fastaCounter = 0;
// ProgressUpdate((double)reader.BaseStream.Position / totalBytes);
//}
}
}
// Still missing peptides?
if (unMapedPeptides.Any(p => !p.IsMapped))
{
int count = 0;
foreach (Peptide pep2 in unMapedPeptides)
{
if (pep2.IsMapped)
continue;
count++;
Log("[ERROR]\tPeptide {0} was not isMapped", pep2);
}
throw new ArgumentException(
string.Format(
"[ERROR] Unable to map every peptide ({0}) to {1}. You might be using either the wrong database, enzyme, or max missed cleavages!",
count, fastaFileniceName));
}
}
Log("Every unique peptide was successfully mapped to at least one protein");
Log("{0:N0} of {1:N0} ({2:F2}%) target proteins were mapped at least once", forwardProteinsMapped, forwardProteins, 100.0 * (double)forwardProteinsMapped / (double)forwardProteins);
Log("{0:N0} of {1:N0} ({2:F2}%) decoy proteins were mapped at least once", decoyProteinsMapped, decoyProteins, 100.0 * (double)decoyProteinsMapped / (double)decoyProteins);
// force the progress bar to go into marquee mode
ProgressUpdate(0.0);
// Return a list of all the proteins that were isMapped at least once
return proteins.Values.ToList();
}
public static void StartRamp(IProtease protease, double percentIdentifiedSteps = 0.05, int maxMissed = 3, int minLength = 5, int maxLength = 35)
{
List<Peptide> peps = new List<Peptide>();
List<Protein> proteins = new List<Protein>();
using (FastaReader reader = new FastaReader("Resources/yeast_uniprot_120226.fasta"))
{
foreach (Protein protein in reader.ReadNextProtein())
{
foreach (Peptide peptide in protein.Digest(protease, maxMissed, minLength, maxLength))
{
peps.Add(peptide);
}
proteins.Add(protein);
}
}
// Fixed seed to make it reproducible
Random random = new Random(480912341);
peps = peps.OrderBy(x => random.Next()).ToList();
for (double percentIdentified = 0; percentIdentified <= 1; percentIdentified += percentIdentifiedSteps)
{
// Take the first x % to act as our identified peptides
List<Peptide> identifiedPeptides = peps.Take((int) (peps.Count*percentIdentified)).ToList();
List<ProteinGroup> proteinGroups = ProteinGroup.GroupProteins(proteins, protease, identifiedPeptides, new AminoAcidLeucineSequenceComparer(), maxMissed).ToList();
Console.WriteLine("{0} peptides {1} protein groups", identifiedPeptides.Count, proteinGroups.Count);
}
}
public void WriteFasta(string fasta_file, FastaWriter Writer)
{
bool MakeDecoy = false;
if (Options.OutputType == DatabaseType.Target || Options.OutputType == DatabaseType.Concatenated)
{
MakeDecoy = false;
}
else if (Options.OutputType == DatabaseType.Decoy || Options.OutputType == DatabaseType.Concatenated)
{
MakeDecoy = true;
}
using (FastaReader reader = new FastaReader(fasta_file))
{
foreach (Fasta fasta in reader.ReadNextFasta())
{
Regex uniprotRegex = new Regex(@"(.+)\|(.+)\|(.+?)\s(.+?)\sOS=(.+?)(?:\sGN=(.+?))?(?:$|PE=(\d+)\sSV=(\d+))", RegexOptions.ExplicitCapture);
Match UniprotMatch = uniprotRegex.Match(fasta.Description);
string HeaderFile = "InvalidUniprotheaders.txt";
string headerFolder = Path.GetDirectoryName(Options.InputFiles[0]);
if (Options.EnforceUniprot && !UniprotMatch.Success)
{
using (StreamWriter log = new StreamWriter(Path.Combine(headerFolder, HeaderFile), true))
{
log.WriteLine("Invalid Header:");
log.WriteLine();
log.WriteLine(fasta.Description);
log.WriteLine();
InvalidHeader(fasta);
}
}
if (UniprotMatch.Success)
{
bool excludeMethionine = false;
if (Options.ExcludeNTerminalMethionine && !Options.ExcludeNTerminalResidue)
{
excludeMethionine = true;
}
if (MakeDecoy)
{
Writer.Write(fasta.ToDecoy(Options.DecoyPrefix, Options.DecoyType, (excludeMethionine || Options.ExcludeNTerminalResidue), Options.ExcludeNTerminalMethionine));
}
else
{
Writer.Write(fasta);
}
}
}
}
}
