public IndexingEngine(List <Protein> proteinList, List <ModificationWithMass> variableModifications, List <ModificationWithMass> fixedModifications, List <ProductType> productTypes, int currentPartition, DecoyType decoyType, IEnumerable <DigestionParams> collectionOfDigestionParams, CommonParameters commonParams, double maxFragmentSize, List <string> nestedIds) : base(commonParams, nestedIds)
{
ProteinList = proteinList;
VariableModifications = variableModifications;
FixedModifications = fixedModifications;
ProductTypes = productTypes;
CurrentPartition = currentPartition + 1;
DecoyType = decoyType;
CollectionOfDigestionParams = collectionOfDigestionParams;
MaxFragmentSize = maxFragmentSize;
}
public IndexingEngine(List <Protein> proteinList, List <ModificationWithMass> variableModifications, List <ModificationWithMass> fixedModifications, List <ProductType> lp, int currentPartition, DecoyType decoyType, IEnumerable <IDigestionParams> CollectionOfDigestionParams, ICommonParameters commonParams, double maxFragmentSize, List <string> nestedIds) : base(nestedIds)
{
this.proteinList = proteinList;
this.variableModifications = variableModifications;
this.fixedModifications = fixedModifications;
this.lp = lp;
this.currentPartition = currentPartition + 1;
this.decoyType = decoyType;
this.CollectionOfDigestionParams = CollectionOfDigestionParams;
this.commonParams = commonParams;
this.maxFragmentSize = maxFragmentSize;
}
public IndexingEngine(List <Protein> proteinList, List <Modification> variableModifications, List <Modification> fixedModifications,
List <SilacLabel> silacLabels, int currentPartition, DecoyType decoyType, CommonParameters commonParams, double maxFragmentSize,
bool generatePrecursorIndex, List <FileInfo> proteinDatabases, List <string> nestedIds)
: base(commonParams, nestedIds)
{
ProteinList = proteinList;
VariableModifications = variableModifications;
FixedModifications = fixedModifications;
SilacLabels = silacLabels;
CurrentPartition = currentPartition + 1;
DecoyType = decoyType;
MaxFragmentSize = maxFragmentSize;
GeneratePrecursorIndex = generatePrecursorIndex;
this.ProteinDatabases = proteinDatabases;
}
/// <summary>
/// Generates decoys for a list of proteins
/// </summary>
/// <param name="proteins"></param>
/// <param name="decoyType"></param>
/// <param name="digestionParams"></param>
/// <param name="randomSeed">Used when decoy type is shuffle for shuffling the peptides</param>
/// <returns></returns>
public static List <Protein> GenerateDecoys(List <Protein> proteins, DecoyType decoyType, int maxThreads = -1)
{
if (decoyType == DecoyType.None)
{
return(new List <Protein>());
}
else if (decoyType == DecoyType.Reverse)
{
return(GenerateReverseDecoys(proteins, maxThreads));
}
else if (decoyType == DecoyType.Slide)
{
return(GenerateSlideDecoys(proteins, maxThreads));
}
else
{
throw new ArgumentException("Decoy type " + decoyType.ToString() + " is not implemented.");
}
}
public void MoreTests(string filename, DecoyType decoyType = DecoyType.None)
{
string xmlName = Path.Combine(TestContext.CurrentContext.TestDirectory, "TestData", filename);
var proteins = ProteinDbLoader.LoadProteinXML(xmlName, decoyType == DecoyType.None, decoyType, null, false, null, out var un);
var peps = proteins[1].Digest(CommonParameters.DigestionParams, null, null).ToList();
PeptideWithSetModifications pep = peps[peps.Count - 2];
string mzmlName = $"ajgdiv{filename}{decoyType.ToString()}.mzML";
MsDataFile myMsDataFile = new TestDataFile(new List <PeptideWithSetModifications> {
pep
});
IO.MzML.MzmlMethods.CreateAndWriteMyMzmlWithCalibratedSpectra(myMsDataFile, mzmlName, false);
string outputFolder = Path.Combine(TestContext.CurrentContext.TestDirectory, $"TestSearchWithVariants{filename}{decoyType.ToString()}");
Directory.CreateDirectory(outputFolder);
SearchTask st = new SearchTask
{
SearchParameters = new SearchParameters
{
DoParsimony = true,
DecoyType = decoyType,
SearchTarget = decoyType == DecoyType.None,
ModPeptidesAreDifferent = false
},
CommonParameters = new CommonParameters(scoreCutoff: 1, digestionParams: new DigestionParams(minPeptideLength: 2), precursorMassTolerance: new PpmTolerance(20)),
};
st.RunTask(outputFolder, new List <DbForTask> {
new DbForTask(xmlName, false)
}, new List <string> {
mzmlName
}, "");
var psms = File.ReadAllLines(Path.Combine(outputFolder, "AllPSMs.psmtsv"));
//Assert.IsTrue(psms.Any(line => line.Contains($"\t{variantPsmShort}\t" + (containsVariant ? variantPsmShort : "\t"))));
Directory.Delete(outputFolder, true);
File.Delete(mzmlName);
//Directory.Delete(Path.Combine(TestContext.CurrentContext.TestDirectory, @"Task Settings"), true);
}
public void InitializeDecoy(int pNumber, float decoyLifetime, float health, float movementSpeed,
float explosionRadius, float explosionDamage, VFXManager vfxManager, DecoyType type,
float secondsForTarget = 0, float targetRadius = 0, int numberOfDecoys = 0)
{
this.pNumber = pNumber;
this.decoyLifetime = decoyLifetime;
this.health = health;
this.movementSpeed = movementSpeed;
this.targetRadius = targetRadius;
this.explosionRadius = explosionRadius;
this.explosionDamage = explosionDamage;
this.secondsForAction = secondsForTarget;
this.numberOfDecoys = numberOfDecoys;
this.vfxManager = vfxManager;
this.type = type;
if (type == DecoyType.MovementDecoy)
{
enemiesList = FindObjectsOfType <Hero>().Where(p => p.PlayerNumber != pNumber).ToList();
}
agent.speed = movementSpeed;
}
public PrecursorIndexingEngine(List <Protein> proteinList, List <ModificationWithMass> variableModifications, List <ModificationWithMass> fixedModifications, List <ProductType> lp, int currentPartition, DecoyType decoyType, IEnumerable <DigestionParams> CollectionOfDigestionParams, CommonParameters commonParams, double maxFragmentSize, List <string> nestedIds) : base(proteinList, variableModifications, fixedModifications, lp, currentPartition, decoyType, CollectionOfDigestionParams, commonParams, maxFragmentSize, nestedIds)
{
}
public static List <Protein> LoadProteinXML(string proteinDbLocation, bool generateTargetProteins, DecoyType decoyType, IEnumerable <Modification> allKnownModifications, bool IsContaminant, IEnumerable <string> modTypesToExclude, out Dictionary <string, Modification> unknownModifications)
{
List <Modification> prespecified = GetPtmListFromProteinXml(proteinDbLocation);
Dictionary <string, IList <Modification> > mod_dict = new Dictionary <string, IList <Modification> >();
if (prespecified.Count > 0 || allKnownModifications.Count() > 0)
{
mod_dict = GetModificationDict(new HashSet <Modification>(prespecified.Concat(allKnownModifications)));
}
List <Protein> result = new List <Protein>();
unknownModifications = new Dictionary <string, Modification>();
using (var stream = new FileStream(proteinDbLocation, FileMode.Open, FileAccess.Read, FileShare.Read))
{
Regex substituteWhitespace = new Regex(@"\s+");
Stream uniprotXmlFileStream = proteinDbLocation.EndsWith(".gz") ?
(Stream)(new GZipStream(stream, CompressionMode.Decompress)) :
stream;
string[] nodes = new string[6];
string accession = null;
string name = null;
string full_name = null;
string organism = null;
string sequence = null;
string feature_type = null;
string feature_description = null;
string original_value = ""; // if no content is found, assume it is empty, not null (e.g. <original>A</original><variation/> for a deletion event)
string variation_value = "";
string dbReference_type = null;
string dbReference_id = null;
List <string> property_types = new List <string>();
List <string> property_values = new List <string>();
int oneBasedfeature_position = -1;
int? oneBasedbeginPosition = null;
int? oneBasedendPosition = null;
List <ProteolysisProduct> proteolysisProducts = new List <ProteolysisProduct>();
List <SequenceVariation> sequenceVariations = new List <SequenceVariation>();
List <DisulfideBond> disulfideBonds = new List <DisulfideBond>();
var oneBasedModifications = new Dictionary <int, List <Modification> >();
List <Tuple <string, string> > gene_names = new List <Tuple <string, string> >();
bool reading_gene = false;
bool reading_organism = false;
List <DatabaseReference> databaseReferences = new List <DatabaseReference>();
using (XmlReader xml = XmlReader.Create(uniprotXmlFileStream))
{
while (xml.Read())
{
switch (xml.NodeType)
{
case XmlNodeType.Element:
nodes[xml.Depth] = xml.Name;
int outValue;
switch (xml.Name)
{
case "accession":
if (accession == null)
{
accession = xml.ReadElementString();
}
break;
case "name":
if (xml.Depth == 2 && !reading_gene && !reading_organism)
{
name = xml.ReadElementString();
}
if (reading_gene && !reading_organism)
{
gene_names.Add(new Tuple <string, string>(xml.GetAttribute("type"), xml.ReadElementString()));
}
if (reading_organism)
{
if (xml.GetAttribute("type").Equals("scientific"))
{
organism = xml.ReadElementString();
}
}
break;
case "gene":
reading_gene = true;
break;
case "organism":
if (organism == null)
{
reading_organism = true;
}
break;
case "fullName":
if (full_name == null)
{
full_name = xml.ReadElementString();
}
break;
case "feature":
feature_type = xml.GetAttribute("type");
feature_description = xml.GetAttribute("description");
break;
case "original":
original_value = xml.ReadElementString();
break;
case "variation":
variation_value = xml.ReadElementString();
break;
case "dbReference":
property_types.Clear();
property_values.Clear();
dbReference_type = xml.GetAttribute("type");
dbReference_id = xml.GetAttribute("id");
break;
case "property":
property_types.Add(xml.GetAttribute("type"));
property_values.Add(xml.GetAttribute("value"));
break;
case "position":
oneBasedfeature_position = int.Parse(xml.GetAttribute("position"));
break;
case "begin":
oneBasedbeginPosition = int.TryParse(xml.GetAttribute("position"), out outValue) ? (int?)outValue : null;
break;
case "end":
oneBasedendPosition = int.TryParse(xml.GetAttribute("position"), out outValue) ? (int?)outValue : null;
break;
case "sequence":
sequence = substituteWhitespace.Replace(xml.ReadElementString(), "");
break;
}
break;
case XmlNodeType.EndElement:
switch (xml.Name)
{
case "feature":
if (feature_type == "modified residue")
{
feature_description = feature_description.Split(';')[0];
// Create new entry for this residue, if needed
if (!oneBasedModifications.TryGetValue(oneBasedfeature_position, out List <Modification> residue_modifications))
{
residue_modifications = new List <Modification>();
oneBasedModifications.Add(oneBasedfeature_position, residue_modifications);
}
if (mod_dict.ContainsKey(feature_description))
{
// Known and not of a type in the exclusion list
List <Modification> mods = mod_dict[feature_description].Where(m => !modTypesToExclude.Contains(m.modificationType)).ToList();
if (mods.Count == 0 && oneBasedModifications[oneBasedfeature_position].Count == 0)
{
oneBasedModifications.Remove(oneBasedfeature_position);
}
else
{
oneBasedModifications[oneBasedfeature_position].AddRange(mods);
}
}
else if (unknownModifications.ContainsKey(feature_description))
{
// Not known but seen
residue_modifications.Add(unknownModifications[feature_description]);
}
else
{
// Not known and not seen
unknownModifications[feature_description] = new Modification(feature_description, "unknown");
residue_modifications.Add(unknownModifications[feature_description]);
}
}
else if (feature_type == "peptide" || feature_type == "propeptide" || feature_type == "chain" || feature_type == "signal peptide")
{
proteolysisProducts.Add(new ProteolysisProduct(oneBasedbeginPosition, oneBasedendPosition, feature_type));
}
else if (feature_type == "sequence variant" && // Only keep if there is variant sequence information and position information
variation_value != null &&
variation_value != "")
{
if (oneBasedbeginPosition != null && oneBasedendPosition != null)
{
sequenceVariations.Add(new SequenceVariation((int)oneBasedbeginPosition, (int)oneBasedendPosition, original_value, variation_value, feature_description));
}
else if (oneBasedfeature_position >= 1)
{
sequenceVariations.Add(new SequenceVariation(oneBasedfeature_position, original_value, variation_value, feature_description));
}
}
else if (feature_type == "disulfide bond")
{
if (oneBasedbeginPosition != null && oneBasedendPosition != null)
{
disulfideBonds.Add(new DisulfideBond((int)oneBasedbeginPosition, (int)oneBasedendPosition, feature_description));
}
else if (oneBasedfeature_position >= 1)
{
disulfideBonds.Add(new DisulfideBond(oneBasedfeature_position, feature_description));
}
}
oneBasedbeginPosition = null;
oneBasedendPosition = null;
oneBasedfeature_position = -1;
original_value = "";
variation_value = "";
break;
case "dbReference":
databaseReferences.Add(new DatabaseReference(dbReference_type, dbReference_id, Enumerable.Range(0, property_types.Count).Select(i => new Tuple <string, string>(property_types[i], property_values[i])).ToList()));
property_types = new List <string>();
property_values = new List <string>();
dbReference_type = null;
dbReference_id = null;
break;
case "gene":
reading_gene = false;
break;
case "organism":
reading_organism = false;
break;
case "entry":
if (accession != null && sequence != null)
{
if (generateTargetProteins)
{
var protein = new Protein(sequence, accession, organism, gene_names, oneBasedModifications, proteolysisProducts, name, full_name, false, IsContaminant, databaseReferences, sequenceVariations, disulfideBonds, proteinDbLocation);
result.Add(protein);
}
switch (decoyType)
{
case DecoyType.Reverse:
char[] sequence_array = sequence.ToCharArray();
Dictionary <int, List <Modification> > decoy_modifications = null;
List <DisulfideBond> decoy_disulfides = new List <DisulfideBond>();
if (sequence.StartsWith("M", StringComparison.Ordinal))
{
// Do not include the initiator methionine in reversal!!!
Array.Reverse(sequence_array, 1, sequence.Length - 1);
decoy_modifications = new Dictionary <int, List <Modification> >(oneBasedModifications.Count);
foreach (var kvp in oneBasedModifications)
{
if (kvp.Key > 1)
{
decoy_modifications.Add(sequence.Length - kvp.Key + 2, kvp.Value);
}
else if (kvp.Key == 1)
{
decoy_modifications.Add(1, kvp.Value);
}
}
}
else
{
Array.Reverse(sequence_array);
decoy_modifications = new Dictionary <int, List <Modification> >(oneBasedModifications.Count);
foreach (var kvp in oneBasedModifications)
{
decoy_modifications.Add(sequence.Length - kvp.Key + 1, kvp.Value);
}
}
var reversed_sequence = new string(sequence_array);
List <ProteolysisProduct> decoyPP = new List <ProteolysisProduct>();
foreach (ProteolysisProduct pp in proteolysisProducts)
{
decoyPP.Add(new ProteolysisProduct(sequence.Length - pp.OneBasedEndPosition + 1, sequence.Length - pp.OneBasedBeginPosition, pp.Type));
}
foreach (DisulfideBond disulfideBond in disulfideBonds)
{
decoy_disulfides.Add(new DisulfideBond(sequence.Length - disulfideBond.OneBasedBeginPosition + 2, sequence.Length - disulfideBond.OneBasedEndPosition + 2, "DECOY DISULFIDE BOND: " + disulfideBond.Description));
}
List <SequenceVariation> decoy_variations = new List <SequenceVariation>();
foreach (SequenceVariation sv in sequenceVariations)
{
char[] original_array = sv.OriginalSequence.ToArray();
char[] variation_array = sv.VariantSequence.ToArray();
if (sv.OneBasedBeginPosition == 1)
{
bool orig_init_m = sv.OriginalSequence.StartsWith("M", StringComparison.Ordinal);
bool var_init_m = sv.VariantSequence.StartsWith("M", StringComparison.Ordinal);
if (orig_init_m && !var_init_m)
{
decoy_variations.Add(new SequenceVariation(1, "M", "", "DECOY VARIANT: Initiator Methionine Change in " + sv.Description));
}
original_array = sv.OriginalSequence.Substring(Convert.ToInt32(orig_init_m)).ToArray();
variation_array = sv.VariantSequence.Substring(Convert.ToInt32(var_init_m)).ToArray();
}
int decoy_end = sequence.Length - sv.OneBasedBeginPosition + 2 + Convert.ToInt32(sv.OneBasedEndPosition == reversed_sequence.Length) - Convert.ToInt32(sv.OneBasedBeginPosition == 1);
int decoy_begin = decoy_end - original_array.Length + 1;
Array.Reverse(original_array);
Array.Reverse(variation_array);
decoy_variations.Add(new SequenceVariation(decoy_begin, decoy_end, new string(original_array), new string(variation_array), "DECOY VARIANT: " + sv.Description));
}
var decoy_protein = new Protein(reversed_sequence, "DECOY_" + accession, organism, gene_names, decoy_modifications, decoyPP, name, full_name, true, IsContaminant, null, decoy_variations, decoy_disulfides, proteinDbLocation);
result.Add(decoy_protein);
break;
case DecoyType.Slide:
int numSlides = 20;
char[] sequence_array_unslided = sequence.ToCharArray();
char[] sequence_array_slided = sequence.ToCharArray();
decoy_modifications = null;
List <DisulfideBond> decoy_disulfides_slide = new List <DisulfideBond>();
if (sequence.StartsWith("M", StringComparison.Ordinal))
{
// Do not include the initiator methionine in shuffle!!!
if (numSlides % sequence_array_slided.Length - 1 == 0)
{
numSlides++;
}
for (int i = 1; i < sequence_array_slided.Length; i++)
{
sequence_array_slided[i] = sequence_array_unslided[GetOldShuffleIndex(i, numSlides, sequence.Length, true)];
}
decoy_modifications = new Dictionary <int, List <Modification> >(oneBasedModifications.Count);
foreach (var kvp in oneBasedModifications)
{
if (kvp.Key > 1)
{
decoy_modifications.Add(GetOldShuffleIndex(kvp.Key - 1, numSlides, sequence.Length, true) + 1, kvp.Value);
}
else if (kvp.Key == 1)
{
decoy_modifications.Add(1, kvp.Value);
}
}
}
else
{
if (numSlides % sequence_array_slided.Length == 0)
{
numSlides++;
}
for (int i = 0; i < sequence_array_slided.Length; i++)
{
sequence_array_slided[i] = sequence_array_unslided[GetOldShuffleIndex(i, numSlides, sequence.Length, false)];
}
decoy_modifications = new Dictionary <int, List <Modification> >(oneBasedModifications.Count);
foreach (var kvp in oneBasedModifications)
{
decoy_modifications.Add(GetOldShuffleIndex(kvp.Key - 1, numSlides, sequence.Length, false) + 1, kvp.Value);
}
}
var slided_sequence = new string(sequence_array_slided);
List <ProteolysisProduct> decoyPP_slide = new List <ProteolysisProduct>();
foreach (ProteolysisProduct pp in proteolysisProducts) //can't keep all aa like you can with reverse, just keep it the same length
{
decoyPP_slide.Add(pp);
}
foreach (DisulfideBond disulfideBond in disulfideBonds) //these actually need the same cysteines...
{
decoy_disulfides_slide.Add(new DisulfideBond(GetOldShuffleIndex(disulfideBond.OneBasedBeginPosition - 1, numSlides, slided_sequence.Length, false) + 1, GetOldShuffleIndex(disulfideBond.OneBasedEndPosition - 1, numSlides, slided_sequence.Length, false) + 1, "DECOY DISULFIDE BOND: " + disulfideBond.Description));
}
List <SequenceVariation> decoy_variations_slide = new List <SequenceVariation>();
foreach (SequenceVariation sv in sequenceVariations) //No idea what's going on here. Review is appreciated.
{
char[] original_array_unshuffled = sv.OriginalSequence.ToArray();
char[] variation_array_unslided = sv.VariantSequence.ToArray();
if (sv.OneBasedBeginPosition == 1)
{
bool orig_init_m = sv.OriginalSequence.StartsWith("M", StringComparison.Ordinal);
bool var_init_m = sv.VariantSequence.StartsWith("M", StringComparison.Ordinal);
if (orig_init_m && !var_init_m)
{
decoy_variations_slide.Add(new SequenceVariation(1, "M", "", "DECOY VARIANT: Initiator Methionine Change in " + sv.Description));
}
original_array_unshuffled = sv.OriginalSequence.Substring(Convert.ToInt32(orig_init_m)).ToArray();
variation_array_unslided = sv.VariantSequence.Substring(Convert.ToInt32(var_init_m)).ToArray();
}
int decoy_end = sequence.Length - sv.OneBasedBeginPosition + 2 + Convert.ToInt32(sv.OneBasedEndPosition == slided_sequence.Length) - Convert.ToInt32(sv.OneBasedBeginPosition == 1);
int decoy_begin = decoy_end - original_array_unshuffled.Length + 1;
char[] original_array_slided = sv.OriginalSequence.ToArray();
char[] variation_array_slided = sv.VariantSequence.ToArray();
if (numSlides % original_array_slided.Length == 0)
{
numSlides++;
}
for (int i = 0; i < original_array_slided.Length; i++)
{
original_array_slided[i] = original_array_unshuffled[GetOldShuffleIndex(i, numSlides, original_array_unshuffled.Length, false)];
}
if (numSlides % variation_array_slided.Length == 0)
{
numSlides++;
}
for (int i = 0; i < variation_array_slided.Length; i++)
{
variation_array_slided[i] = variation_array_unslided[GetOldShuffleIndex(i, numSlides, variation_array_unslided.Length, false)];
}
decoy_variations_slide.Add(new SequenceVariation(decoy_begin, decoy_end, new string(original_array_slided), new string(variation_array_slided), "DECOY VARIANT: " + sv.Description));
}
var decoy_protein_slide = new Protein(slided_sequence, "DECOY_" + accession, organism, gene_names, decoy_modifications, decoyPP_slide, name, full_name, true, IsContaminant, null, decoy_variations_slide, decoy_disulfides_slide, proteinDbLocation);
result.Add(decoy_protein_slide);
break;
default:
break;
}
}
accession = null;
name = null;
full_name = null;
sequence = null;
organism = null;
feature_type = null;
feature_description = null;
original_value = "";
variation_value = "";
dbReference_type = null;
dbReference_id = null;
property_types = new List <string>();
property_values = new List <string>();
oneBasedfeature_position = -1;
oneBasedModifications = new Dictionary <int, List <Modification> >();
proteolysisProducts = new List <ProteolysisProduct>();
sequenceVariations = new List <SequenceVariation>();
disulfideBonds = new List <DisulfideBond>();
databaseReferences = new List <DatabaseReference>();
gene_names = new List <Tuple <string, string> >();
reading_gene = false;
reading_organism = false;
break;
}
break;
}
}
}
}
return(result);
}
/// <summary>
/// Load a protein fasta database, using regular expressions to get various aspects of the headers. The first regex capture group is used as each field.
/// </summary>
/// <param name="proteinDbLocation"></param>
/// <param name="onTheFlyDecoys"></param>
/// <param name="IsContaminant"></param>
/// <param name="accession_expression"></param>
/// <param name="full_name_expression"></param>
/// <param name="name_expression"></param>
/// <param name="gene_expression"></param>
/// <returns></returns>
public static List <Protein> LoadProteinFasta(string proteinDbLocation, bool originalTarget, DecoyType onTheFlyDecoys, bool IsContaminant, Regex accession_expression, Regex full_name_expression, Regex name_expression, Regex gene_expression, Regex organism_expression, out List <string> errors)
{
HashSet <string> unique_accessions = new HashSet <string>();
int unique_identifier = 1;
string accession = null;
string name = null;
string full_name = null;
string organism = null;
List <Tuple <string, string> > gene_name = new List <Tuple <string, string> >();
errors = new List <string>();
Regex substituteWhitespace = new Regex(@"\s+");
List <Protein> result = new List <Protein>();
using (var stream = new FileStream(proteinDbLocation, FileMode.Open, FileAccess.Read, FileShare.Read))
{
Stream fastaFileStream = proteinDbLocation.EndsWith(".gz") ?
(Stream)(new GZipStream(stream, CompressionMode.Decompress)) :
stream;
StringBuilder sb = null;
StreamReader fasta = new StreamReader(fastaFileStream);
while (true)
{
string line = "";
line = fasta.ReadLine();
if (line == null)
{
break;
}
if (line.StartsWith(">"))
{
var accession_match = accession_expression.Match(line);
var full_name_match = full_name_expression.Match(line);
var name_match = name_expression.Match(line);
var gene_name_match = gene_expression.Match(line);
if (organism_expression != null)
{
var organism_match = organism_expression.Match(line);
if (organism_match.Groups.Count > 1)
{
organism = organism_expression.Match(line).Groups[1].Value;
}
}
if (accession_match.Groups.Count > 1)
{
accession = accession_expression.Match(line).Groups[1].Value;
}
if (full_name_match.Groups.Count > 1)
{
full_name = full_name_expression.Match(line).Groups[1].Value;
}
if (name_match.Groups.Count > 1)
{
name = name_expression.Match(line).Groups[1].Value;
}
if (gene_name_match.Groups.Count > 1)
{
gene_name.Add(new Tuple <string, string>("primary", gene_expression.Match(line).Groups[1].Value));
}
if (accession == null || accession == "")
{
accession = line.Substring(1).TrimEnd();
}
sb = new StringBuilder();
}
else if (sb != null)
{
sb.Append(line.Trim());
}
if ((fasta.Peek() == '>' || fasta.Peek() == -1) && accession != null && sb != null)
{
string sequence = substituteWhitespace.Replace(sb.ToString(), "");
while (unique_accessions.Contains(accession))
{
accession += "_" + unique_identifier.ToString();
unique_identifier++;
}
unique_accessions.Add(accession);
if (originalTarget)
{
Protein protein = new Protein(sequence, accession, organism, gene_name, name: name, full_name: full_name, isContaminant: IsContaminant, databaseFilePath: proteinDbLocation);
if (protein.Length == 0)
{
errors.Add("Line" + line + ", Protein Length of 0: " + protein.Name + " was skipped from database: " + proteinDbLocation);
}
else
{
result.Add(protein);
}
}
switch (onTheFlyDecoys)
{
case DecoyType.Reverse:
char[] sequence_array = sequence.ToCharArray();
int starts_with_met = sequence.StartsWith("M", StringComparison.Ordinal) ? 1 : 0;
Array.Reverse(sequence_array, starts_with_met, sequence.Length - starts_with_met); // Do not include the initiator methionine in reversal!!!
var reversed_sequence = new string(sequence_array);
Protein decoy_protein = new Protein(reversed_sequence, "DECOY_" + accession, organism, gene_name, name: name, full_name: full_name, isDecoy: true, isContaminant: IsContaminant, databaseFilePath: proteinDbLocation);
result.Add(decoy_protein);
break;
case DecoyType.Slide:
int numSlides = 20;
char[] sequence_array_unslide = sequence.ToCharArray();
char[] sequence_array_slide = sequence.ToCharArray();
bool starts_with_met_slide = sequence.StartsWith("M", StringComparison.Ordinal);
for (int i = starts_with_met_slide ? 1 : 0; i < sequence.Length; i++)
{
sequence_array_slide[i] = sequence_array_unslide[GetOldShuffleIndex(i, numSlides, sequence.Length, starts_with_met_slide)];
}
string slide_sequence = new string(sequence_array_slide);
Protein decoy_protein_slide = new Protein(slide_sequence, "DECOY_" + accession, organism, gene_name, name: name, full_name: full_name, isDecoy: true, isContaminant: IsContaminant, databaseFilePath: proteinDbLocation);
result.Add(decoy_protein_slide);
break;
default:
break;
}
accession = null;
name = null;
full_name = null;
organism = null;
gene_name = new List <Tuple <string, string> >();
}
// no input left
if (fasta.Peek() == -1)
{
break;
}
}
}
if (!result.Any())
{
errors.Add("Error: No proteins could be read from the database: " + proteinDbLocation);
}
return(result);
}
public static List <Protein> LoadProteinXML(string proteinDbLocation, bool generateTargets, DecoyType decoyType, IEnumerable <Modification> allKnownModifications,
bool isContaminant, IEnumerable <string> modTypesToExclude, out Dictionary <string, Modification> unknownModifications, int maxThreads = -1)
{
List <Modification> prespecified = GetPtmListFromProteinXml(proteinDbLocation);
allKnownModifications = allKnownModifications ?? new List <Modification>();
modTypesToExclude = modTypesToExclude ?? new List <string>();
//Dictionary<string, IList<Modification>> modsDictionary = new Dictionary<string, IList<Modification>>();
if (prespecified.Count > 0 || allKnownModifications.Count() > 0)
{
//modsDictionary = GetModificationDict(new HashSet<Modification>(prespecified.Concat(allKnownModifications)));
IdToPossibleMods = GetModificationDict(new HashSet <Modification>(prespecified.Concat(allKnownModifications)));
IdWithMotifToMod = GetModificationDictWithMotifs(new HashSet <Modification>(prespecified.Concat(allKnownModifications)));
}
List <Protein> targets = new List <Protein>();
unknownModifications = new Dictionary <string, Modification>();
using (var stream = new FileStream(proteinDbLocation, FileMode.Open, FileAccess.Read, FileShare.Read))
{
Regex substituteWhitespace = new Regex(@"\s+");
Stream uniprotXmlFileStream = proteinDbLocation.EndsWith("gz") ? // allow for .bgz and .tgz, which are (rarely) used
(Stream)(new GZipStream(stream, CompressionMode.Decompress)) :
stream;
ProteinXmlEntry block = new ProteinXmlEntry();
using (XmlReader xml = XmlReader.Create(uniprotXmlFileStream))
{
while (xml.Read())
{
if (xml.NodeType == XmlNodeType.Element)
{
block.ParseElement(xml.Name, xml);
}
if (xml.NodeType == XmlNodeType.EndElement || xml.IsEmptyElement)
{
var newProteinEntries = block.ParseEndElement(xml, modTypesToExclude, unknownModifications,
isContaminant, proteinDbLocation);
targets.AddRange(newProteinEntries);
}
}
}
}
List <Protein> decoys = DecoyProteinGenerator.GenerateDecoys(targets, decoyType, maxThreads);
List <Protein> intermediateProteinList = (generateTargets ? targets : new List <Protein>()).Concat(decoyType != DecoyType.None ? decoys : new List <Protein>()).ToList();
return((generateTargets ? targets : new List <Protein>()).Concat(decoyType != DecoyType.None ? decoys : new List <Protein>()).ToList());
}
/// <summary>
/// Load a protein fasta database, using regular expressions to get various aspects of the headers. The first regex capture group is used as each field.
/// </summary>
public static List <Protein> LoadProteinFasta(string proteinDbLocation, bool generateTargets, DecoyType decoyType, bool isContaminant,
FastaHeaderFieldRegex accessionRegex, FastaHeaderFieldRegex fullNameRegex, FastaHeaderFieldRegex nameRegex,
FastaHeaderFieldRegex geneNameRegex, FastaHeaderFieldRegex organismRegex, out List <string> errors, int maxThreads = -1)
{
HashSet <string> unique_accessions = new HashSet <string>();
int unique_identifier = 1;
string accession = null;
string name = null;
string fullName = null;
string organism = null;
List <Tuple <string, string> > geneName = new List <Tuple <string, string> >();
errors = new List <string>();
Regex substituteWhitespace = new Regex(@"\s+");
List <Protein> targets = new List <Protein>();
using (var stream = new FileStream(proteinDbLocation, FileMode.Open, FileAccess.Read, FileShare.Read))
{
Stream fastaFileStream = proteinDbLocation.EndsWith("gz") ? // allow for .bgz and .tgz, which are (rarely) used
(Stream)(new GZipStream(stream, CompressionMode.Decompress)) :
stream;
StringBuilder sb = null;
StreamReader fasta = new StreamReader(fastaFileStream);
while (true)
{
string line = "";
line = fasta.ReadLine();
if (line == null)
{
break;
}
if (line.StartsWith(">"))
{
accession = ApplyRegex(accessionRegex, line);
fullName = ApplyRegex(fullNameRegex, line);
name = ApplyRegex(nameRegex, line);
organism = ApplyRegex(organismRegex, line);
string geneNameString = ApplyRegex(geneNameRegex, line);
if (geneNameString != null)
{
geneName.Add(new Tuple <string, string>("primary", geneNameString));
}
if (accession == null || accession == "")
{
accession = line.Substring(1).TrimEnd();
}
sb = new StringBuilder();
}
else if (sb != null)
{
sb.Append(line.Trim());
}
if ((fasta.Peek() == '>' || fasta.Peek() == -1) && accession != null && sb != null)
{
string sequence = substituteWhitespace.Replace(sb.ToString(), "");
while (unique_accessions.Contains(accession))
{
accession += "_" + unique_identifier.ToString();
unique_identifier++;
}
unique_accessions.Add(accession);
Protein protein = new Protein(sequence, accession, organism, geneName, name: name, fullName: fullName,
isContaminant: isContaminant, databaseFilePath: proteinDbLocation);
if (protein.Length == 0)
{
errors.Add("Line" + line + ", Protein Length of 0: " + protein.Name + " was skipped from database: " + proteinDbLocation);
}
else
{
targets.Add(protein);
}
accession = null;
name = null;
fullName = null;
organism = null;
geneName = new List <Tuple <string, string> >();
}
// no input left
if (fasta.Peek() == -1)
{
break;
}
}
}
if (!targets.Any())
{
errors.Add("Error: No proteins could be read from the database: " + proteinDbLocation);
}
List <Protein> decoys = DecoyProteinGenerator.GenerateDecoys(targets, decoyType, maxThreads);
return((generateTargets ? targets : new List <Protein>()).Concat(decoyType != DecoyType.None ? decoys : new List <Protein>()).ToList());
}
[TestCase(9, 0, true, "EDITPEPEDITP2PPP", DecoyType.Reverse)] // MPEPPP becomes MPPPEP with the variant beginning at position 2
public static void SearchTests(int proteinIdx, int peptideIdx, bool containsVariant, string variantPsmShort, DecoyType decoyType = DecoyType.None)
{
Stopwatch stopwatch = new Stopwatch();
stopwatch.Start();
// Make sure can run the complete search task when multiple compact peptides may correspond to a single PWSM
SearchTask st = new SearchTask
{
SearchParameters = new SearchParameters
{
DoParsimony = true,
DecoyType = decoyType,
SearchTarget = decoyType == DecoyType.None,
ModPeptidesAreDifferent = false
},
CommonParameters = new CommonParameters(scoreCutoff: 1, digestionParams: new DigestionParams(minPeptideLength: 2), precursorMassTolerance: new PpmTolerance(20)),
};
ModificationMotif.TryGetMotif("V", out ModificationMotif motifV);
Modification mv = new Modification("mod", null, "type", null, motifV, "Anywhere.", null, 42.01, new Dictionary <string, IList <string> >(), null, null, null, null, null);
ModificationMotif.TryGetMotif("P", out ModificationMotif motifP);
Modification mp = new Modification("mod", null, "type", null, motifP, "Anywhere.", null, 42.01, new Dictionary <string, IList <string> >(), null, null, null, null, null);
List <Protein> proteins = new List <Protein>
{
new Protein("MPEPTIDE", "protein1", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 4, "P", "V", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEPTIDE", "protein2", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 5, "PT", "KT", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEPTIDE", "protein3", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 4, "P", "PPP", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEPPPTIDE", "protein3", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 6, "PPP", "P", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEPKPKTIDE", "protein3", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 7, "PKPK", "PK", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEPTAIDE", "protein2", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 5, "PTA", "KT", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEKKAIDE", "protein2", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 4, "KKA", "K", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
new Protein("MPEPTIDE", "protein1", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 4, "P", "V", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", new Dictionary <int, List <Modification> > {
{ 4, new[] { mv }.ToList() }
})
}),
new Protein("MPEPTIDE", "protein3", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 4, "P", "PPP", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", new Dictionary <int, List <Modification> > {
{ 5, new[] { mp }.ToList() }
})
}),
new Protein("MPEPTIDEPEPTIDE", "protein3", sequenceVariations: new List <SequenceVariation> {
new SequenceVariation(4, 4, "PTIDEPEPTIDE", "PPP", @"1\t50000000\t.\tA\tG\t.\tPASS\tANN=G||||||||||||||||\tGT:AD:DP\t1/1:30,30:30", null)
}),
};
PeptideWithSetModifications pep = proteins[proteinIdx].GetVariantProteins().SelectMany(p => p.Digest(CommonParameters.DigestionParams, null, null)).ToList()[peptideIdx];
string xmlName = $"andguiaheov{proteinIdx.ToString()}.xml";
ProteinDbWriter.WriteXmlDatabase(null, new List <Protein> {
proteins[proteinIdx]
}, xmlName);
string mzmlName = $"ajgdiv{proteinIdx.ToString()}.mzML";
MsDataFile myMsDataFile = new TestDataFile(new List <PeptideWithSetModifications> {
pep
});
IO.MzML.MzmlMethods.CreateAndWriteMyMzmlWithCalibratedSpectra(myMsDataFile, mzmlName, false);
string outputFolder = Path.Combine(TestContext.CurrentContext.TestDirectory, $"TestSearchWithVariants{proteinIdx.ToString()}");
Directory.CreateDirectory(outputFolder);
st.RunTask(outputFolder, new List <DbForTask> {
new DbForTask(xmlName, false)
}, new List <string> {
mzmlName
}, "");
var psms = File.ReadAllLines(Path.Combine(outputFolder, "AllPSMs.psmtsv"));
Assert.IsTrue(psms.Any(line => line.Contains(containsVariant ? variantPsmShort : "\t")));
Directory.Delete(outputFolder, true);
File.Delete(mzmlName);
File.Delete(xmlName);
Directory.Delete(Path.Combine(TestContext.CurrentContext.TestDirectory, @"Task Settings"), true);
Console.WriteLine($"Analysis time for VariantSearchTests.SearchTests({proteinIdx.ToString()},{peptideIdx.ToString()},{containsVariant.ToString()},{variantPsmShort}): {stopwatch.Elapsed.Hours}h {stopwatch.Elapsed.Minutes}m {stopwatch.Elapsed.Seconds}s");
}
/// <summary>
/// Load a protein fasta database, using regular expressions to get various aspects of the headers. The first regex capture group is used as each field.
/// </summary>
public static List <Protein> LoadProteinFasta(string proteinDbLocation, bool generateTargets, DecoyType decoyType, bool isContaminant,
FastaHeaderFieldRegex accessionRegex, FastaHeaderFieldRegex fullNameRegex, FastaHeaderFieldRegex nameRegex,
FastaHeaderFieldRegex geneNameRegex, FastaHeaderFieldRegex organismRegex, out List <string> errors, int maxThreads = -1)
{
HashSet <string> unique_accessions = new HashSet <string>();
int unique_identifier = 2;//for isoforms. the first will be "accession", the next will be "accession_2"
string accession = null;
string name = null;
string fullName = null;
string organism = null;
List <Tuple <string, string> > geneName = new List <Tuple <string, string> >();
errors = new List <string>();
Regex substituteWhitespace = new Regex(@"\s+");
List <Protein> targets = new List <Protein>();
using (var stream = new FileStream(proteinDbLocation, FileMode.Open, FileAccess.Read, FileShare.Read))
{
Stream fastaFileStream = proteinDbLocation.EndsWith("gz") ? // allow for .bgz and .tgz, which are (rarely) used
(Stream)(new GZipStream(stream, CompressionMode.Decompress)) :
stream;
StringBuilder sb = null;
StreamReader fasta = new StreamReader(fastaFileStream);
while (true)
{
string line = "";
line = fasta.ReadLine();
if (line == null)
{
break;
}
if (line.StartsWith(">"))
{
accession = ApplyRegex(accessionRegex, line);
fullName = ApplyRegex(fullNameRegex, line);
name = ApplyRegex(nameRegex, line);
organism = ApplyRegex(organismRegex, line);
string geneNameString = ApplyRegex(geneNameRegex, line);
if (geneNameString != null)
{
geneName.Add(new Tuple <string, string>("primary", geneNameString));
}
if (accession == null || accession == "")
{
accession = line.Substring(1).TrimEnd();
}
sb = new StringBuilder();
}
else if (sb != null)
{
sb.Append(line.Trim());
}
if ((fasta.Peek() == '>' || fasta.Peek() == -1) && accession != null && sb != null)
{
string sequence = substituteWhitespace.Replace(sb.ToString(), "");
// sanitize the sequence to replace unexpected characters with X (unknown amino acid)
// sometimes strange characters get added by RNA sequencing software, etc.
sequence = SanitizeAminoAcidSequence(sequence, 'X');
if (unique_accessions.Contains(accession)) //this will happen for isoforms
{
string originalAccession = accession; //save the original
accession += "_" + unique_identifier.ToString(); //add a number onto it
while (unique_accessions.Contains(accession)) //if that number was already added
{
unique_identifier++; //keep increasing it
accession = originalAccession + "_" + unique_identifier.ToString(); //try the new number
}
unique_identifier = 2; //reset
}
unique_accessions.Add(accession);
Protein protein = new Protein(sequence, accession, organism, geneName, name: name, fullName: fullName,
isContaminant: isContaminant, databaseFilePath: proteinDbLocation);
if (protein.Length == 0)
{
errors.Add("Line" + line + ", Protein Length of 0: " + protein.Name + " was skipped from database: " + proteinDbLocation);
}
else
{
targets.Add(protein);
}
accession = null;
name = null;
fullName = null;
organism = null;
geneName = new List <Tuple <string, string> >();
}
// no input left
if (fasta.Peek() == -1)
{
break;
}
}
}
if (!targets.Any())
{
errors.Add("Error: No proteins could be read from the database: " + proteinDbLocation);
}
List <Protein> decoys = DecoyProteinGenerator.GenerateDecoys(targets, decoyType, maxThreads);
return(generateTargets ? targets.Concat(decoys).ToList() : decoys);
}
public static List <Protein> LoadProteinDb(string fileName, bool generateTargets, DecoyType decoyType, int MaxThreadsToUse)
{
List <string> dbErrors = new List <string>();
List <Protein> proteinList = new List <Protein>();
string theExtension = Path.GetExtension(fileName).ToLowerInvariant();
bool compressed = theExtension.EndsWith("gz"); // allows for .bgz and .tgz, too which are used on occasion
theExtension = compressed ? Path.GetExtension(Path.GetFileNameWithoutExtension(fileName)).ToLowerInvariant() : theExtension;
if (theExtension.Equals(".fasta") || theExtension.Equals(".fa"))
{
proteinList = ProteinDbLoader.LoadProteinFasta(fileName, generateTargets, decoyType, false, ProteinDbLoader.UniprotAccessionRegex, ProteinDbLoader.UniprotFullNameRegex, ProteinDbLoader.UniprotFullNameRegex, ProteinDbLoader.UniprotGeneNameRegex,
ProteinDbLoader.UniprotOrganismRegex, out dbErrors, MaxThreadsToUse);
}
return(proteinList.Where(p => p.BaseSequence.Length > 0).ToList());
}
