public void CalculateProteinCoverage(List <FastaItem> fastaItems)
{
Parallel.ForEach(MyProteinList, myProtein =>
//foreach (MyProtein myProtein in MyProteinList)
{
//Locate the corrsponding fasta item
myProtein.Locus = myProtein.Locus.TrimEnd('\r', '\n');
PatternTools.FastaParser.FastaItem item = fastaItems.Find(a => a.SequenceIdentifier.Equals(myProtein.Locus));
if (item != null)
{
myProtein.Coverage = Math.Round(item.Coverage(myProtein.DistinctPeptides), 4);
myProtein.Length = item.Sequence.Length;
myProtein.MolWt = Math.Round(item.MonoisotopicMass, 1);
myProtein.Description = item.Description;
myProtein.Sequence = item.Sequence;
}
else
{
//For some bizarre reason the key was not found
Console.WriteLine("ERROR:: + Protein must be found in the database: " + myProtein.Locus);
myProtein.Coverage = 0;
myProtein.Length = 0;
myProtein.MolWt = 0;
myProtein.Description = "Description not found in database";
}
}
);
}
public void IncludeScrambled()
{
List <FastaItem> myScrambled = new List <FastaItem>(myItems.Count);
for (int i = 0; i < myItems.Count; i++)
{
FastaItem f = new FastaItem();
f.SequenceIdentifier = "Scrambled_" + myItems[i].SequenceIdentifier;
f.Sequence = ScrambleString(myItems[i].Sequence);
f.Description = "";
myScrambled.Add(f);
}
myItems.AddRange(myScrambled);
}
/// <summary>
///
/// </summary>
/// <param name="fileName"></param>
/// <param name="removeReverse"></param>
/// <param name="removeIPIVersionNumber"></param>
/// <param name="considerOnlyFirstName"></param>
/// <param name="useRegexForTitleOrDescription">If you wish to only consider items that have certain words in their descriptions or names. You should also set the Regex parameter</param>
/// <param name="theRegex"></param>
public void ParseFile(
StreamReader sr,
bool removeReverse,
DBTypes dbType,
bool useRegexForTitleOrDescription = false,
Regex theRegex = null
)
{
//Prepare the stream Reader variables
string read;
Regex ipiVersion = new Regex(@"\.[0-9]+", RegexOptions.Compiled);
Regex ipiRemove = new Regex(@">IPI:", RegexOptions.Compiled);
Regex NextProtIDRegex = new Regex(@">nxp:([A-Z|0-9|\-|_]+) ", RegexOptions.Compiled);
Regex correctIPI = new Regex(@"^\|>", RegexOptions.Compiled);
Regex spaceSplitter = new Regex(@" ", RegexOptions.Compiled);
Regex SequestFriendlyGI = new Regex(@">gi\|([0-9]*)", RegexOptions.Compiled);
Regex SequestFriendlyGIDescription = new Regex(@">[^ ]* (.*)", RegexOptions.Compiled);
Regex SequestFriendlyIPI = new Regex(@"IPI:(IPI[0-9|\.]+)\|", RegexOptions.Compiled);
Regex SequestFriendlyIPIDescription = new Regex(@"Tax_Id=[0-9]+ Gene_Symbol=(.*)", RegexOptions.Compiled);
Regex SequestFriendUniProt = new Regex(@"[a-z]+\|([A-Z|0-9|\-]+)\|", RegexOptions.Compiled);
Regex SequestFriendlyUniProtDescription = new Regex(@" (.*)");
Regex TitleSpaceDescriptionID = new Regex(@">(\S+)", RegexOptions.Compiled);
Regex TitleSpaceDescriptionDescription = new Regex(@" (.*)", RegexOptions.Compiled);
//Declare the PathCandidates we will use
//Parse and load to Ram
FastaItem item = new FastaItem();
int SequenceCounter = 0;
int lineCounter = 0;
int problematicCounter = 0;
while ((read = sr.ReadLine()) != null)
{
lineCounter++;
if (read.StartsWith(">"))
{
if (SequenceCounter > 0)
{
if (useRegexForTitleOrDescription)
{
if (theRegex.IsMatch(item.SequenceIdentifier) || theRegex.IsMatch(item.Description))
{
myItems.Add(item);
}
}
else
{
myItems.Add(item);
}
}
else
{
SequenceCounter++;
}
//---Parse search engine friendly
item = new FastaItem();
Match id;
Match description;
if (dbType.Equals(DBTypes.NCBInr))
{
id = SequestFriendlyGI.Match(read);
description = SequestFriendlyGIDescription.Match(read);
item.Description = description.Groups[1].Value;
}
else if (dbType.Equals(DBTypes.IPI))
{
id = SequestFriendlyIPI.Match(read);
description = SequestFriendlyIPIDescription.Match(read);
item.Description = description.Groups[1].Value;
}
else if (dbType.Equals(DBTypes.IDSpaceDescription))
{
id = TitleSpaceDescriptionID.Match(read);
description = TitleSpaceDescriptionDescription.Match(read);
item.Description = description.Groups[1].Value;
}
else if (dbType.Equals(DBTypes.UniProt))
{
id = SequestFriendUniProt.Match(read);
description = SequestFriendlyUniProtDescription.Match(read);
item.Description = description.Groups[1].Value;
}
else if (dbType.Equals(DBTypes.NeXtProt))
{
id = NextProtIDRegex.Match(read);
List <string> cols = Regex.Split(read, Regex.Escape(@"\")).ToList();
string desc = cols.Find(a => a.StartsWith("Pname="));
item.Description = desc.Remove(0, 6);
}
else if (dbType.Equals(DBTypes.Contaminant))
{
//we are working with dbtypes = all
id = Regex.Match(read, @">(.*)");
//we do not want a match in the description
description = Regex.Match(read, @"cvbcvbfgz");
item.Description = description.Groups[1].Value;
}
else
{
string[] cols = Regex.Split(read, " ");
id = Regex.Match(cols[0], @">(.*)");
try
{
string desc = read.Substring(id.Length + 1);
description = Regex.Match(desc, @"(.*)");
item.Description = description.Groups[1].Value;
}
catch
{
item.Description = "No Description";
}
}
if (id.Groups[1].Value.Length == 0)
{
item.SequenceIdentifier = "ProblematicID" + problematicCounter++;
Console.WriteLine("Problem parsing identifier: " + lineCounter + " problem:" + problematicCounter);
}
else
{
item.SequenceIdentifier = id.Groups[1].Value;
}
//------------
}
else
{
item.Sequence += read;
}
}
Console.WriteLine(lineCounter + "lines parsed in the fasta DB.");
//And now for the last item
if (useRegexForTitleOrDescription)
{
if (theRegex.IsMatch(item.SequenceIdentifier) || theRegex.IsMatch(item.Description))
{
myItems.Add(item);
}
}
else
{
myItems.Add(item);
}
if (removeReverse)
{
myItems.RemoveAll(a => a.SequenceIdentifier.Contains("Reverse_"));
}
sr.Close();
}
