/// <summary>
/// Access the web to resolve protein metadata not directly found in fasta file.
/// The fasta text importer will have left search hints in ProteinMetadata.
/// </summary>
/// <param name="progressMonitor"></param>
/// <param name="fastaImporter">object that accesses the web, or pretends to if in a test</param>
/// <param name="polite">if true, don't try to resolve everything in one go, assume we can come back later</param>
/// <returns>true on success</returns>
public bool LookupProteinMetadata(ProgressMonitor progressMonitor, WebEnabledFastaImporter fastaImporter, bool polite = false)
{
var unsearchedProteins = new List <ProteinSearchInfo>();
List <DbProteinName> untaggedProteins;
using (ISession session = OpenSession())
{
if (!progressMonitor.Invoke(Resources.ProteomeDb_LookupProteinMetadata_looking_for_unresolved_protein_details, 0))
{
return(false);
}
// get a list of proteins with unresolved metadata websearches
var proteinNames = session.CreateCriteria(typeof(DbProteinName)).List <DbProteinName>();
var proteinsToSearch =
proteinNames.Where(proteinName => (proteinName.GetProteinMetadata().GetPendingSearchTerm().Length > 0))
.ToList();
// and a list of proteins which have never been considered for metadata search
untaggedProteins =
proteinNames.Where(proteinName => proteinName.WebSearchInfo.IsEmpty()).ToList();
foreach (var untaggedProtein in untaggedProteins)
{
untaggedProtein.SetWebSearchCompleted(); // by default take this out of consideration for next time
var metadata = untaggedProtein.GetProteinMetadata();
if (metadata.HasMissingMetadata())
{
var search = fastaImporter.ParseProteinMetaData(metadata);
if (search != null)
{
metadata = untaggedProtein.ChangeProteinMetadata(metadata.Merge(search)); // don't stomp name by accident
metadata = untaggedProtein.ChangeProteinMetadata(metadata.ChangeWebSearchInfo(search.WebSearchInfo));
}
}
if (metadata.NeedsSearch())
{
proteinsToSearch.Add(untaggedProtein); // add to the list of things to commit back to the db
}
}
// Get the lengths of the sequences without getting the sequences themselves, for best speed
var proteinIds = proteinsToSearch.Select(name => name.Protein.Id.Value).Distinct().ToArray();
var proteinLengths = new Dictionary <long, int>();
using (var cmd = session.Connection.CreateCommand())
{
string sql = "SELECT Id, LENGTH(Sequence) AS SequenceLength FROM ProteomeDbProtein P"; // Not L10N
if (proteinIds.Length < 1000)
{
sql += " WHERE P.Id IN (" + // Not L10N
string.Join(",", proteinIds) + ")"; // Not L10N
}
cmd.CommandText = sql;
using (var reader = cmd.ExecuteReader())
{
while (reader.Read())
{
var id = reader.GetValue(0);
var len = reader.GetValue(1);
proteinLengths.Add(Convert.ToInt64(id), Convert.ToInt32(len));
}
}
}
foreach (var p in proteinsToSearch)
{
int length;
proteinLengths.TryGetValue(p.Protein.Id.GetValueOrDefault(), out length);
unsearchedProteins.Add(new ProteinSearchInfo(p, length));
}
}
if (untaggedProteins.Any(untagged => !untagged.GetProteinMetadata().NeedsSearch())) // did any get set as unsearchable?
{
// Write back the ones that were formerly without search terms, but which now indicate no search is possible
using (ISession session = OpenWriteSession())
{
using (var transaction = session.BeginTransaction())
{
foreach (var untagged in untaggedProteins.Where(untagged => !untagged.GetProteinMetadata().NeedsSearch()))
{
session.SaveOrUpdate(untagged); // update the metadata
}
transaction.Commit();
}
}
}
if (unsearchedProteins.Any())
{
int resultsCount = 0;
int unsearchedCount = unsearchedProteins.Count;
for (bool success = true; success;)
{
success = false; // Until we see at least one succeed this round
var results = new List <DbProteinName>();
// The "true" arg means "do just one batch then return"
foreach (var result in fastaImporter.DoWebserviceLookup(unsearchedProteins, null, true))
{
if (result != null)
{
if (
!progressMonitor.Invoke(
string.Format(
Resources.ProteomeDb_LookupProteinMetadata_Retrieving_details_for__0__proteins,
unsearchedProteins.Count), 100 * resultsCount++ / unsearchedCount))
{
return(false);
}
success = true;
results.Add(result.ProteinDbInfo);
}
}
if (results.Any()) // save this batch
{
using (var session = OpenWriteSession())
{
using (var transaction = session.BeginTransaction())
{
foreach (var result in results)
{
session.SaveOrUpdate(result);
}
transaction.Commit();
session.Close();
}
}
}
// Edit this list rather than rederive with database access
var hits = unsearchedProteins.Where(p => !p.GetProteinMetadata().NeedsSearch()).ToList();
foreach (var hit in hits)
{
unsearchedProteins.Remove(hit);
}
}
}
return(true);
}
public Digestion Digest(IProtease protease, String name, String description, ProgressMonitor progressMonitor)
{
DbOrganism organism;
DbDigestion digestion;
List <DbProtein> proteins;
using (ISession session = ProteomeDb.OpenWriteSession())
{
organism = GetEntity(session);
session.BeginTransaction();
digestion = new DbDigestion
{
Name = name,
Description = description,
Organism = organism,
MaxMissedCleavages = protease.MaxMissedCleavages
};
session.Save(digestion);
if (!progressMonitor.Invoke("Listing proteins", 0))
{
return(null);
}
proteins = new List <DbProtein>(organism.Proteins);
Dictionary <String, long> digestedPeptideIds
= new Dictionary <string, long>();
const String sqlPeptide =
"INSERT INTO ProteomeDbDigestedPeptide (Digestion, MissedCleavages, Sequence, Version) VALUES(@Digestion,@MissedCleavages,@Sequence,1);select last_insert_rowid();";
var commandPeptide = session.Connection.CreateCommand();
commandPeptide.CommandText = sqlPeptide;
commandPeptide.Parameters.Add(new SQLiteParameter("@Digestion"));
commandPeptide.Parameters.Add(new SQLiteParameter("@MissedCleavages"));
commandPeptide.Parameters.Add(new SQLiteParameter("@Sequence"));
const String sqlPeptideProtein =
"INSERT INTO ProteomeDbDigestedPeptideProtein (StartIndex, Peptide, Protein, Version) VALUES(?,?,?,1);";
var commandProtein = session.Connection.CreateCommand();
commandProtein.CommandText = sqlPeptideProtein;
commandProtein.Parameters.Add(new SQLiteParameter("@StartIndex"));
commandProtein.Parameters.Add(new SQLiteParameter("@Peptide"));
commandProtein.Parameters.Add(new SQLiteParameter("@Protein"));
for (int i = 0; i < proteins.Count; i++)
{
if (!progressMonitor.Invoke("Digesting " + proteins.Count
+ " proteins", 100 * i / proteins.Count))
{
return(null);
}
Protein protein = new Protein(this, proteins[i]);
foreach (DigestedPeptide digestedPeptide in protease.Digest(protein))
{
if (digestedPeptide.Sequence.Length > MAX_PEPTIDE_LENGTH)
{
continue;
}
long digestedPeptideId;
if (!digestedPeptideIds.TryGetValue(digestedPeptide.Sequence, out digestedPeptideId))
{
((SQLiteParameter)commandPeptide.Parameters[0]).Value = digestion.Id;
((SQLiteParameter)commandPeptide.Parameters[1]).Value = digestedPeptide.MissedCleavages;
((SQLiteParameter)commandPeptide.Parameters[2]).Value = digestedPeptide.Sequence;
digestedPeptideId = Convert.ToInt64(commandPeptide.ExecuteScalar());
digestedPeptideIds.Add(digestedPeptide.Sequence, digestedPeptideId);
}
((SQLiteParameter)commandProtein.Parameters[0]).Value = digestedPeptide.Index;
((SQLiteParameter)commandProtein.Parameters[1]).Value = digestedPeptideId;
((SQLiteParameter)commandProtein.Parameters[2]).Value = proteins[i].Id;
commandProtein.ExecuteNonQuery();
}
}
if (!progressMonitor.Invoke("Committing transaction", 99))
{
return(null);
}
session.Transaction.Commit();
progressMonitor.Invoke(
"Digested " + proteins.Count + " proteins into " + digestedPeptideIds.Count + " unique peptides",
100);
return(new Digestion(this, digestion));
}
}
public Digestion Digest(IProtease protease, ProgressMonitor progressMonitor)
{
using (ISession session = OpenWriteSession())
{
DbDigestion dbDigestion = GetDbDigestion(protease.Name);
HashSet <string> existingSequences = new HashSet <string>();
using (var transaction = session.BeginTransaction())
{
if (dbDigestion != null)
{
if (dbDigestion.MaxSequenceLength >= MAX_SEQUENCE_LENGTH)
{
return(new Digestion(this, dbDigestion));
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_Digest_Listing_existing_peptides, 0))
{
return(null);
}
IQuery query = session.CreateQuery("SELECT P.Sequence FROM " // Not L10N
+ typeof(DbDigestedPeptide) + " P WHERE P.Digestion = :Digestion") // Not L10N
.SetParameter("Digestion", dbDigestion); // Not L10N
List <String> listSequences = new List <string>();
query.List(listSequences);
existingSequences.UnionWith(listSequences);
dbDigestion.MaxSequenceLength = MAX_SEQUENCE_LENGTH;
session.Update(dbDigestion);
}
else
{
dbDigestion = new DbDigestion
{
Name = protease.Name,
MinSequenceLength = MIN_SEQUENCE_LENGTH,
MaxSequenceLength = MAX_SEQUENCE_LENGTH,
};
session.Save(dbDigestion);
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_Digest_Listing_proteins, 0))
{
return(null);
}
List <DbProtein> proteins = new List <DbProtein>();
session.CreateCriteria(typeof(DbProtein)).List(proteins);
Dictionary <String, long> digestedPeptideIds
= new Dictionary <string, long>();
const String sqlPeptide =
"INSERT INTO ProteomeDbDigestedPeptide (Digestion, Sequence) VALUES(?,?);select last_insert_rowid();"; // Not L10N
using (var commandPeptide = session.Connection.CreateCommand())
using (var commandProtein = session.Connection.CreateCommand())
{
commandPeptide.CommandText = sqlPeptide;
commandPeptide.Parameters.Add(new SQLiteParameter());
commandPeptide.Parameters.Add(new SQLiteParameter());
const String sqlPeptideProtein =
"INSERT INTO ProteomeDbDigestedPeptideProtein (Peptide, Protein) VALUES(?,?);"; // Not L10N
commandProtein.CommandText = sqlPeptideProtein;
commandProtein.Parameters.Add(new SQLiteParameter());
commandProtein.Parameters.Add(new SQLiteParameter());
commandProtein.Parameters.Add(new SQLiteParameter());
for (int i = 0; i < proteins.Count; i++)
{
var proteinSequences = new HashSet <string>();
if (!progressMonitor.Invoke(string.Format(Resources.ProteomeDb_Digest_Digesting__0__proteins, proteins.Count), 100 * i / proteins.Count))
{
return(null);
}
Protein protein = new Protein(ProteomeDbPath, proteins[i]);
foreach (DigestedPeptide digestedPeptide in protease.Digest(protein))
{
if (digestedPeptide.Sequence.Length < dbDigestion.MinSequenceLength)
{
continue;
}
String truncatedSequence = digestedPeptide.Sequence.Substring(
0, Math.Min(digestedPeptide.Sequence.Length, dbDigestion.MaxSequenceLength));
if (existingSequences.Contains(truncatedSequence))
{
continue;
}
if (proteinSequences.Contains(truncatedSequence))
{
continue;
}
proteinSequences.Add(truncatedSequence);
long digestedPeptideId;
if (!digestedPeptideIds.TryGetValue(truncatedSequence, out digestedPeptideId))
{
((SQLiteParameter)commandPeptide.Parameters[0]).Value = dbDigestion.Id;
((SQLiteParameter)commandPeptide.Parameters[1]).Value = truncatedSequence;
digestedPeptideId = Convert.ToInt64(commandPeptide.ExecuteScalar());
digestedPeptideIds.Add(truncatedSequence, digestedPeptideId);
}
((SQLiteParameter)commandProtein.Parameters[0]).Value = digestedPeptideId;
((SQLiteParameter)commandProtein.Parameters[1]).Value = protein.Id;
commandProtein.ExecuteNonQuery();
}
}
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_AddFastaFile_Saving_changes, 99))
{
return(null);
}
transaction.Commit();
AnalyzeDb(session);
progressMonitor.Invoke(
string.Format(Resources.ProteomeDb_Digest_Digested__0__proteins_into__1__unique_peptides,
proteins.Count, digestedPeptideIds.Count),
100);
}
return(new Digestion(this, dbDigestion));
}
}
public void AddFastaFile(StreamReader reader, ProgressMonitor progressMonitor)
{
Dictionary <string, ProtIdNames> proteinIds = new Dictionary <string, ProtIdNames>();
using (ISession session = OpenWriteSession())
{
foreach (DbProtein protein in session.CreateCriteria(typeof(DbProtein)).List())
{
if (protein.Id.HasValue)
{
proteinIds.Add(protein.Sequence, new ProtIdNames(protein.Id.Value, protein.Names));
}
}
int proteinCount = 0;
using (var transaction = session.BeginTransaction())
using (IDbCommand insertProtein = session.Connection.CreateCommand())
using (IDbCommand insertName = session.Connection.CreateCommand())
{
WebEnabledFastaImporter fastaImporter = new WebEnabledFastaImporter(new WebEnabledFastaImporter.DelayedWebSearchProvider()); // just parse, no search for now
insertProtein.CommandText =
"INSERT INTO ProteomeDbProtein (Version, Sequence) Values (1,?);select last_insert_rowid();"; // Not L10N
insertProtein.Parameters.Add(new SQLiteParameter());
insertName.CommandText =
"INSERT INTO ProteomeDbProteinName (Version, Protein, IsPrimary, Name, Description, PreferredName, Accession, Gene, Species, WebSearchStatus) Values(1,?,?,?,?,?,?,?,?,?)"; // Not L10N
insertName.Parameters.Add(new SQLiteParameter()); // Id
insertName.Parameters.Add(new SQLiteParameter()); // IsPrimary
insertName.Parameters.Add(new SQLiteParameter()); // Name
insertName.Parameters.Add(new SQLiteParameter()); // Description
insertName.Parameters.Add(new SQLiteParameter()); // PreferredName
insertName.Parameters.Add(new SQLiteParameter()); // Accession
insertName.Parameters.Add(new SQLiteParameter()); // Gene
insertName.Parameters.Add(new SQLiteParameter()); // Species
insertName.Parameters.Add(new SQLiteParameter()); // WebSearchInfo
foreach (DbProtein protein in fastaImporter.Import(reader))
{
int iProgress = (int)(reader.BaseStream.Position * 100 / (reader.BaseStream.Length + 1));
if (!progressMonitor.Invoke(string.Format(Resources.ProteomeDb_AddFastaFile_Added__0__proteins, proteinCount), iProgress))
{
return;
}
bool existingProtein = false;
ProtIdNames proteinIdNames;
if (proteinIds.TryGetValue(protein.Sequence, out proteinIdNames))
{
existingProtein = true;
}
else
{
((SQLiteParameter)insertProtein.Parameters[0]).Value = protein.Sequence;
proteinIdNames = new ProtIdNames(Convert.ToInt64(insertProtein.ExecuteScalar()), new DbProteinName[0]);
proteinIds.Add(protein.Sequence, proteinIdNames);
proteinCount++;
}
foreach (var proteinName in protein.Names)
{
// Skip any names that already exist
if (proteinIdNames.Names.Any(dbProteinName => Equals(dbProteinName.Name, proteinName.Name)))
{
continue;
}
try
{
((SQLiteParameter)insertName.Parameters[0]).Value = proteinIdNames.Id;
((SQLiteParameter)insertName.Parameters[1]).Value = proteinName.IsPrimary && !existingProtein;
((SQLiteParameter)insertName.Parameters[2]).Value = proteinName.Name;
((SQLiteParameter)insertName.Parameters[3]).Value = proteinName.Description;
((SQLiteParameter)insertName.Parameters[4]).Value = proteinName.PreferredName;
((SQLiteParameter)insertName.Parameters[5]).Value = proteinName.Accession;
((SQLiteParameter)insertName.Parameters[6]).Value = proteinName.Gene;
((SQLiteParameter)insertName.Parameters[7]).Value = proteinName.Species;
((SQLiteParameter)insertName.Parameters[8]).Value = proteinName.WebSearchStatus; // represent as a string for ease of serialization
insertName.ExecuteNonQuery();
}
catch (Exception exception)
{
Console.Out.WriteLine(exception);
}
}
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_AddFastaFile_Saving_changes, 99))
{
return;
}
transaction.Commit();
}
AnalyzeDb(session);
progressMonitor.Invoke(
string.Format(Resources.ProteomeDb_AddFastaFile_Finished_importing__0__proteins, proteinCount), 100);
}
}
/// <summary>
/// Access the web to resolve protein metadata not directly found in fasta file.
/// The fasta text importer will have left search hints in ProteinMetadata.
/// </summary>
/// <param name="progressMonitor"></param>
/// <param name="fastaImporter">object that accesses the web, or pretends to if in a test</param>
/// <param name="polite">if true, don't try to resolve everything in one go, assume we can come back later</param>
/// <returns>true on success</returns>
public bool LookupProteinMetadata(ProgressMonitor progressMonitor, WebEnabledFastaImporter fastaImporter, bool polite = false)
{
var unsearchedProteins = new List<ProteinSearchInfo>();
List<DbProteinName> untaggedProteins;
using (ISession session = OpenSession())
{
if (!progressMonitor.Invoke(Resources.ProteomeDb_LookupProteinMetadata_looking_for_unresolved_protein_details, 0))
{
return false;
}
// get a list of proteins with unresolved metadata websearches
var proteinNames = session.CreateCriteria(typeof (DbProteinName)).List<DbProteinName>();
var proteinsToSearch =
proteinNames.Where(proteinName => (proteinName.GetProteinMetadata().GetPendingSearchTerm().Length > 0))
.ToList();
// and a list of proteins which have never been considered for metadata search
untaggedProteins =
proteinNames.Where(proteinName => proteinName.WebSearchInfo.IsEmpty()).ToList();
foreach (var untaggedProtein in untaggedProteins)
{
untaggedProtein.SetWebSearchCompleted(); // by default take this out of consideration for next time
var metadata = untaggedProtein.GetProteinMetadata();
if (metadata.HasMissingMetadata())
{
var search = fastaImporter.ParseProteinMetaData(metadata);
if (search!=null)
{
metadata = untaggedProtein.ChangeProteinMetadata(metadata.Merge(search)); // don't stomp name by accident
metadata = untaggedProtein.ChangeProteinMetadata(metadata.ChangeWebSearchInfo(search.WebSearchInfo));
}
}
if (metadata.NeedsSearch())
proteinsToSearch.Add(untaggedProtein); // add to the list of things to commit back to the db
}
// Get the lengths of the sequences without getting the sequences themselves, for best speed
var proteinIds = proteinsToSearch.Select(name => name.Protein.Id.Value).Distinct().ToArray();
var proteinLengths = new Dictionary<long, int>();
using (var cmd = session.Connection.CreateCommand())
{
string sql = "SELECT Id, LENGTH(Sequence) AS SequenceLength FROM ProteomeDbProtein P"; // Not L10N
if (proteinIds.Length < 1000)
{
sql += " WHERE P.Id IN (" + // Not L10N
string.Join(",", proteinIds) + ")"; // Not L10N
}
cmd.CommandText = sql;
using (var reader = cmd.ExecuteReader())
{
while (reader.Read())
{
var id = reader.GetValue(0);
var len = reader.GetValue(1);
proteinLengths.Add(Convert.ToInt64(id), Convert.ToInt32(len));
}
}
}
foreach (var p in proteinsToSearch)
{
int length;
proteinLengths.TryGetValue(p.Protein.Id.GetValueOrDefault(), out length);
unsearchedProteins.Add(new ProteinSearchInfo(p, length));
}
}
if (untaggedProteins.Any(untagged => !untagged.GetProteinMetadata().NeedsSearch())) // did any get set as unsearchable?
{
// Write back the ones that were formerly without search terms, but which now indicate no search is possible
using (ISession session = OpenWriteSession())
{
using (var transaction = session.BeginTransaction())
{
foreach (var untagged in untaggedProteins.Where(untagged => !untagged.GetProteinMetadata().NeedsSearch()))
session.SaveOrUpdate(untagged); // update the metadata
transaction.Commit();
}
}
}
if (unsearchedProteins.Any())
{
int resultsCount = 0;
int unsearchedCount = unsearchedProteins.Count;
for (bool success = true; success;)
{
success = false; // Until we see at least one succeed this round
var results = new List<DbProteinName>();
// The "true" arg means "do just one batch then return"
foreach (var result in fastaImporter.DoWebserviceLookup(unsearchedProteins, null, true))
{
if (result != null)
{
if (
!progressMonitor.Invoke(
string.Format(
Resources.ProteomeDb_LookupProteinMetadata_Retrieving_details_for__0__proteins,
unsearchedProteins.Count), 100 * resultsCount++ / unsearchedCount))
{
return false;
}
success = true;
results.Add(result.ProteinDbInfo);
}
}
if (results.Any()) // save this batch
{
using (var session = OpenWriteSession())
{
using (var transaction = session.BeginTransaction())
{
foreach (var result in results)
session.SaveOrUpdate(result);
transaction.Commit();
session.Close();
}
}
}
// Edit this list rather than rederive with database access
var hits = unsearchedProteins.Where(p => !p.GetProteinMetadata().NeedsSearch()).ToList();
foreach (var hit in hits)
{
unsearchedProteins.Remove(hit);
}
}
}
return true;
}
public Digestion Digest(IProtease protease, ProgressMonitor progressMonitor)
{
using (ISession session = OpenWriteSession())
{
DbDigestion dbDigestion = GetDbDigestion(protease.Name);
HashSet<string> existingSequences = new HashSet<string>();
using (var transaction = session.BeginTransaction())
{
if (dbDigestion != null)
{
if (dbDigestion.MaxSequenceLength >= MAX_SEQUENCE_LENGTH)
{
return new Digestion(this, dbDigestion);
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_Digest_Listing_existing_peptides, 0))
{
return null;
}
IQuery query = session.CreateQuery("SELECT P.Sequence FROM " // Not L10N
+ typeof(DbDigestedPeptide) + " P WHERE P.Digestion = :Digestion") // Not L10N
.SetParameter("Digestion", dbDigestion); // Not L10N
List<String> listSequences = new List<string>();
query.List(listSequences);
existingSequences.UnionWith(listSequences);
dbDigestion.MaxSequenceLength = MAX_SEQUENCE_LENGTH;
session.Update(dbDigestion);
}
else
{
dbDigestion = new DbDigestion
{
Name = protease.Name,
MinSequenceLength = MIN_SEQUENCE_LENGTH,
MaxSequenceLength = MAX_SEQUENCE_LENGTH,
};
session.Save(dbDigestion);
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_Digest_Listing_proteins, 0))
{
return null;
}
List<DbProtein> proteins = new List<DbProtein>();
session.CreateCriteria(typeof(DbProtein)).List(proteins);
Dictionary<String, long> digestedPeptideIds
= new Dictionary<string, long>();
const String sqlPeptide =
"INSERT INTO ProteomeDbDigestedPeptide (Digestion, Sequence) VALUES(?,?);select last_insert_rowid();"; // Not L10N
using (var commandPeptide = session.Connection.CreateCommand())
using (var commandProtein = session.Connection.CreateCommand())
{
commandPeptide.CommandText = sqlPeptide;
commandPeptide.Parameters.Add(new SQLiteParameter());
commandPeptide.Parameters.Add(new SQLiteParameter());
const String sqlPeptideProtein =
"INSERT INTO ProteomeDbDigestedPeptideProtein (Peptide, Protein) VALUES(?,?);"; // Not L10N
commandProtein.CommandText = sqlPeptideProtein;
commandProtein.Parameters.Add(new SQLiteParameter());
commandProtein.Parameters.Add(new SQLiteParameter());
commandProtein.Parameters.Add(new SQLiteParameter());
for (int i = 0; i < proteins.Count; i++)
{
var proteinSequences = new HashSet<string>();
if (!progressMonitor.Invoke(string.Format(Resources.ProteomeDb_Digest_Digesting__0__proteins,proteins.Count), 100 * i / proteins.Count))
{
return null;
}
Protein protein = new Protein(ProteomeDbPath, proteins[i]);
foreach (DigestedPeptide digestedPeptide in protease.Digest(protein))
{
if (digestedPeptide.Sequence.Length < dbDigestion.MinSequenceLength)
{
continue;
}
String truncatedSequence = digestedPeptide.Sequence.Substring(
0, Math.Min(digestedPeptide.Sequence.Length, dbDigestion.MaxSequenceLength));
if (existingSequences.Contains(truncatedSequence))
{
continue;
}
if (proteinSequences.Contains(truncatedSequence))
{
continue;
}
proteinSequences.Add(truncatedSequence);
long digestedPeptideId;
if (!digestedPeptideIds.TryGetValue(truncatedSequence, out digestedPeptideId))
{
((SQLiteParameter)commandPeptide.Parameters[0]).Value = dbDigestion.Id;
((SQLiteParameter)commandPeptide.Parameters[1]).Value = truncatedSequence;
digestedPeptideId = Convert.ToInt64(commandPeptide.ExecuteScalar());
digestedPeptideIds.Add(truncatedSequence, digestedPeptideId);
}
((SQLiteParameter)commandProtein.Parameters[0]).Value = digestedPeptideId;
((SQLiteParameter)commandProtein.Parameters[1]).Value = protein.Id;
commandProtein.ExecuteNonQuery();
}
}
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_AddFastaFile_Saving_changes, 99))
{
return null;
}
transaction.Commit();
AnalyzeDb(session);
progressMonitor.Invoke(
string.Format(Resources.ProteomeDb_Digest_Digested__0__proteins_into__1__unique_peptides,
proteins.Count, digestedPeptideIds.Count),
100);
}
return new Digestion(this, dbDigestion);
}
}
public void AddFastaFile(StreamReader reader, ProgressMonitor progressMonitor)
{
Dictionary<string, ProtIdNames> proteinIds = new Dictionary<string, ProtIdNames>();
using (ISession session = OpenWriteSession())
{
foreach (DbProtein protein in session.CreateCriteria(typeof(DbProtein)).List())
{
if (protein.Id.HasValue)
proteinIds.Add(protein.Sequence, new ProtIdNames(protein.Id.Value, protein.Names));
}
int proteinCount = 0;
using (var transaction = session.BeginTransaction())
using (IDbCommand insertProtein = session.Connection.CreateCommand())
using (IDbCommand insertName = session.Connection.CreateCommand())
{
WebEnabledFastaImporter fastaImporter = new WebEnabledFastaImporter(new WebEnabledFastaImporter.DelayedWebSearchProvider()); // just parse, no search for now
insertProtein.CommandText =
"INSERT INTO ProteomeDbProtein (Version, Sequence) Values (1,?);select last_insert_rowid();"; // Not L10N
insertProtein.Parameters.Add(new SQLiteParameter());
insertName.CommandText =
"INSERT INTO ProteomeDbProteinName (Version, Protein, IsPrimary, Name, Description, PreferredName, Accession, Gene, Species, WebSearchStatus) Values(1,?,?,?,?,?,?,?,?,?)"; // Not L10N
insertName.Parameters.Add(new SQLiteParameter()); // Id
insertName.Parameters.Add(new SQLiteParameter()); // IsPrimary
insertName.Parameters.Add(new SQLiteParameter()); // Name
insertName.Parameters.Add(new SQLiteParameter()); // Description
insertName.Parameters.Add(new SQLiteParameter()); // PreferredName
insertName.Parameters.Add(new SQLiteParameter()); // Accession
insertName.Parameters.Add(new SQLiteParameter()); // Gene
insertName.Parameters.Add(new SQLiteParameter()); // Species
insertName.Parameters.Add(new SQLiteParameter()); // WebSearchInfo
foreach (DbProtein protein in fastaImporter.Import(reader))
{
int iProgress = (int)(reader.BaseStream.Position * 100 / (reader.BaseStream.Length + 1));
if (!progressMonitor.Invoke(string.Format(Resources.ProteomeDb_AddFastaFile_Added__0__proteins,proteinCount), iProgress))
{
return;
}
bool existingProtein = false;
ProtIdNames proteinIdNames;
if (proteinIds.TryGetValue(protein.Sequence, out proteinIdNames))
{
existingProtein = true;
}
else
{
((SQLiteParameter)insertProtein.Parameters[0]).Value = protein.Sequence;
proteinIdNames = new ProtIdNames(Convert.ToInt64(insertProtein.ExecuteScalar()), new DbProteinName[0]);
proteinIds.Add(protein.Sequence, proteinIdNames);
proteinCount++;
}
foreach (var proteinName in protein.Names)
{
// Skip any names that already exist
if (proteinIdNames.Names.Any(dbProteinName => Equals(dbProteinName.Name, proteinName.Name)))
continue;
try
{
((SQLiteParameter)insertName.Parameters[0]).Value = proteinIdNames.Id;
((SQLiteParameter)insertName.Parameters[1]).Value = proteinName.IsPrimary && !existingProtein;
((SQLiteParameter)insertName.Parameters[2]).Value = proteinName.Name;
((SQLiteParameter)insertName.Parameters[3]).Value = proteinName.Description;
((SQLiteParameter)insertName.Parameters[4]).Value = proteinName.PreferredName;
((SQLiteParameter)insertName.Parameters[5]).Value = proteinName.Accession;
((SQLiteParameter)insertName.Parameters[6]).Value = proteinName.Gene;
((SQLiteParameter)insertName.Parameters[7]).Value = proteinName.Species;
((SQLiteParameter)insertName.Parameters[8]).Value = proteinName.WebSearchStatus; // represent as a string for ease of serialization
insertName.ExecuteNonQuery();
}
catch (Exception exception)
{
Console.Out.WriteLine(exception);
}
}
}
if (!progressMonitor.Invoke(Resources.ProteomeDb_AddFastaFile_Saving_changes, 99))
{
return;
}
transaction.Commit();
}
AnalyzeDb(session);
progressMonitor.Invoke(
string.Format(Resources.ProteomeDb_AddFastaFile_Finished_importing__0__proteins, proteinCount), 100);
}
}