public ILoopsCollection Search(IPdb Pdb)
{
ILoopsCollection LoopsCollection = new ILoopsCollection();
ILoopCriteria LoopCriteria = new ILoopCriteria(true);
//Try some LINQ on this beast
var Chains = (from p in Pdb.Atoms select p.Chain).Distinct();
foreach (string ChainName in Chains)
{
//Get all the atoms in the chain
var Chain = (from p in Pdb.Atoms where p.Chain == ChainName select p);
IResidue[] AtomsInChain = Chain.ToArray();
//Set the chain length if it is not explicitly set (full length here)
if (LoopCriteria.Start == -1)
{
LoopCriteria.Start = 0;
}
if (LoopCriteria.End == -1)
{
LoopCriteria.End = AtomsInChain.Count();
}
//Search for the loops
if (LoopCriteria.Length > AtomsInChain.Count())
{
continue;
}
for (int i = LoopCriteria.Start; i < LoopCriteria.End - LoopCriteria.Length - 1; i++)
{
//Check for the terminal distances
double eDistance = EuclideanDistance(AtomsInChain[i], AtomsInChain[i + LoopCriteria.Length]);
if (eDistance < LoopCriteria.TerminiDistanceThreshold)
{
//Passes the Distance Test
//Check for the solvent accessible regions
if (LoopCriteria.SolventAccessible)
{
List <IResidue> LoopResidues = new List <IResidue>();
for (int j = i; j < i + LoopCriteria.Length; j++)
{
LoopResidues.Add(AtomsInChain[j]);
}
IResidue ApexResidue = IsLoopVectorSolventAccessible(Pdb.Atoms, LoopResidues, 4, LoopCriteria);
//IResidue ApexResidue = IsLoopSolventAccessible(Pdb.Atoms, LoopResidues, LoopCriteria);
if (ApexResidue != null)
{
//Passes the solvent test, add it to the loop collection
ILoop Loop = new ILoop();
Loop.NTermAtom = AtomsInChain[i];
Loop.NTermAtomP1 = AtomsInChain[i + 1];
Loop.NTermAtomP2 = AtomsInChain[i + 2];
Loop.CTermAtom = AtomsInChain[i + LoopCriteria.Length];
Loop.CTermAtomP1 = AtomsInChain[i + LoopCriteria.Length - 1];
Loop.CTermAtomP2 = AtomsInChain[i + LoopCriteria.Length - 2];
Loop.LoopLength = LoopCriteria.Length;
Loop.PdbCode = Pdb.PdbCode;
Loop.Chain = ChainName;
Loop.ApexAtom = ApexResidue;
Loop.UnitVector = new Vector3D(((Loop.NTermAtomP2.X + Loop.CTermAtomP2.X) / 2) - ((Loop.NTermAtom.X + Loop.CTermAtom.X) / 2),
((Loop.NTermAtomP2.Y + Loop.CTermAtomP2.Y) / 2) - ((Loop.NTermAtom.Y + Loop.CTermAtom.Y) / 2),
((Loop.NTermAtomP2.X + Loop.CTermAtomP2.X) / 2) - ((Loop.NTermAtom.X + Loop.CTermAtom.X) / 2));
//Ensure Beta-Loop-Beta
if (EuclideanDistance(Loop.NTermAtom, Loop.CTermAtom) < LoopCriteria.TerminiDistanceThreshold &&
EuclideanDistance(Loop.NTermAtomP1, Loop.CTermAtomP1) < LoopCriteria.TerminiDistanceThreshold &&
EuclideanDistance(Loop.NTermAtomP2, Loop.CTermAtomP2) < LoopCriteria.TerminiDistanceThreshold)
{
LoopsCollection.Add(Loop);
}
}
}
}
}
LoopCriteria.End = -1;
LoopCriteria.Start = -1;
}
//Cleanup
LoopCriteria = null;
//Return
return(CleanupLoops(LoopsCollection));
}
static void Main(string[] args)
{
#region Argument Parameters
string OutputFilename;
string PdbDirectory = @"C:\PDB\Monomer";
string PdbMultimerDirectory = @"C:\Archives\Biounits";
string PdbFile = @"C:\PDB\Monomer\pdb1rcv.ent.gz";
bool useMultimeric = false;
#endregion
ILoopCriteria LoopCriteria = new ILoopCriteria();
Arguments CommandLine = new Arguments(args);
if (CommandLine["directory"] != null)
{
PdbDirectory = CommandLine["directory"];
}
if (CommandLine["multimeric"] != null)
{
useMultimeric = true;
}
if (CommandLine["searchfortrimer"] != null)
{
LoopCriteria.SearchForTrimer = true;
LoopCriteria.SearchForLoopPairs = false;
}
if (CommandLine["file"] != null)
{
PdbFile = CommandLine["file"];
}
if (CommandLine["match"] != null)
{
LoopCriteria.SearchForLoopPairs = true;
if (CommandLine["lower"] != null)
{
LoopCriteria.LoopPairDistanceLowerLimit = Convert.ToDouble(CommandLine["lower"]);
}
if (CommandLine["upper"] != null)
{
LoopCriteria.LoopPairDistanceUpperLimit = Convert.ToDouble(CommandLine["upper"]);
}
}
else
{
LoopCriteria.SearchForLoopPairs = false;
}
if (CommandLine["out"] != null)
{
OutputFilename = CommandLine["out"];
}
else
{
OutputFilename = "ScaffoldFinderLog_" + DateTime.Now.ToString("yyyyMMddHHmmss") + ".txt";
}
string LogFilename = @"c:\Archives\Output\" + OutputFilename;
FileStream LogStream = File.Open(LogFilename,
FileMode.CreateNew,
FileAccess.Write);
StreamWriter LogFile = new StreamWriter(LogStream, Encoding.UTF8);
//Write the lof file header
if (LoopCriteria.SearchForLoopPairs)
{
LogFile.WriteLine("PDB,ChainA,NTermLoopA,CTermLoopA,Length,ChainB,NTermLoopB,CTermLoopB,LengthB,Size");
}
else if (LoopCriteria.SearchForTrimer)
{
LogFile.WriteLine("PDB,ChainA,NTermLoopA,CTermLoopA,Length,ChainB,NTermLoopB,CTermLoopB,LengthB,ChainC,NTermLoopC,CTermLoopC,LengthC");
}
else
{
LogFile.WriteLine("PDB,Chain,NTerm,CTerm,Length");
}
//Loop through the directory to get the PDBs
DirectoryInfo Directory;
if (useMultimeric)
{
Directory = new DirectoryInfo(PdbMultimerDirectory);
}
else
{
Directory = new DirectoryInfo(PdbDirectory);
}
int FileCounter = 0;
int FileTotals = Directory.GetFiles().Count();
foreach (FileInfo file in Directory.GetFiles())
{
FileCounter++;
if (FileCounter % 500 == 0)
{
Console.WriteLine("[" + FileCounter + "/" + FileTotals + "]: " + file.FullName);
}
//Console.SetCursorPosition(0, Console.CursorTop);
//Console.WriteLine("Filename: " + file.Name + " Size: " + file.Length);
if (file.Length > 500000)
{
continue;
}
IPdb Pdb = new IPdb(file.FullName, useMultimeric);
ILoopsCollection LoopsCollection = new ILoopsCollection();
if (Pdb.Atoms.Count() != 0) // Implies Nucleic Acid
{
FindLoops findLoops = new FindLoops();
LoopsCollection = findLoops.Search(Pdb);
//LoopsCollection = FindLoops.Search(Pdb);
findLoops = null;
}
else
{
continue;
}
//Search for three loops that match the trimer constraints
if (LoopCriteria.SearchForTrimer)
{
FindLoops findLoops = new FindLoops();
List <ILoopTrimer> LoopTrimer = findLoops.FindTrimerLoops(LoopsCollection, LoopCriteria);
foreach (ILoopTrimer Loops in LoopTrimer)
{
LogFile.WriteLine(Loops.LoopA.PdbCode + "," +
Loops.LoopA.Chain + "," +
Loops.LoopA.NTermAtom.ResidueNumber + Loops.LoopA.NTermAtom.ResidueExtension + "," +
Loops.LoopA.CTermAtom.ResidueNumber + Loops.LoopA.CTermAtom.ResidueExtension + "," +
Loops.LoopA.LoopLength + "," +
Loops.LoopB.Chain + "," +
Loops.LoopB.NTermAtom.ResidueNumber + Loops.LoopB.NTermAtom.ResidueExtension + "," +
Loops.LoopB.CTermAtom.ResidueNumber + Loops.LoopB.CTermAtom.ResidueExtension + "," +
Loops.LoopB.LoopLength + "," +
Loops.LoopC.Chain + "," +
Loops.LoopC.NTermAtom.ResidueNumber + Loops.LoopC.NTermAtom.ResidueExtension + "," +
Loops.LoopC.CTermAtom.ResidueNumber + Loops.LoopC.CTermAtom.ResidueExtension + "," +
Loops.LoopC.LoopLength);
}
LoopTrimer = null;
findLoops = null;
}
else if (LoopCriteria.SearchForLoopPairs)
{
FindLoops findLoops = new FindLoops();
List <ILoopPair> LoopPairs = findLoops.FindMatchedLoops(LoopsCollection, LoopCriteria);
foreach (ILoopPair LoopPair in LoopPairs)
{
Console.WriteLine("Found: " + LoopPair.LoopA.PdbCode);
LogFile.WriteLine(LoopPair.LoopA.PdbCode + "," +
LoopPair.LoopA.Chain + "," +
LoopPair.LoopA.NTermAtom.ResidueNumber + LoopPair.LoopA.NTermAtom.ResidueExtension + "," +
LoopPair.LoopA.CTermAtom.ResidueNumber + LoopPair.LoopA.CTermAtom.ResidueExtension + "," +
LoopPair.LoopA.LoopLength + "," +
LoopPair.LoopB.Chain + "," +
LoopPair.LoopB.NTermAtom.ResidueNumber + LoopPair.LoopB.NTermAtom.ResidueExtension + "," +
LoopPair.LoopB.CTermAtom.ResidueNumber + LoopPair.LoopB.CTermAtom.ResidueExtension + "," +
LoopPair.LoopB.LoopLength + "," +
Pdb.Atoms.Count());
}
LoopPairs = null;
findLoops = null;
}
else
{
foreach (ILoop Loop in LoopsCollection.Loops)
{
LogFile.WriteLine(Loop.PdbCode + "," +
Loop.Chain + "," +
Loop.NTermAtom.ResidueNumber + Loop.NTermAtom.ResidueExtension + "," +
Loop.CTermAtom.ResidueNumber + Loop.CTermAtom.ResidueExtension + "," +
Loop.LoopLength);
}
//Cleanup
}
}
LogFile.Close();
LogStream.Close();
// Console.ReadLine();
}
static void Main(string[] args)
{
#region Argument Parameters
string OutputFilename;
string PdbDirectory = @"C:\PDB\Monomer";
string PdbMultimerDirectory = @"C:\Archives\Biounits";
string PdbFile = @"C:\PDB\Monomer\pdb1rcv.ent.gz";
bool useMultimeric = false;
#endregion
ILoopCriteria LoopCriteria = new ILoopCriteria();
Arguments CommandLine = new Arguments(args);
if (CommandLine["directory"] != null) {
PdbDirectory = CommandLine["directory"];
}
if (CommandLine["multimeric"] != null) {
useMultimeric = true;
}
if (CommandLine["searchfortrimer"] != null)
{
LoopCriteria.SearchForTrimer = true;
LoopCriteria.SearchForLoopPairs = false;
}
if (CommandLine["file"] != null) {
PdbFile = CommandLine["file"];
}
if (CommandLine["match"] != null) {
LoopCriteria.SearchForLoopPairs = true;
if (CommandLine["lower"] != null) {
LoopCriteria.LoopPairDistanceLowerLimit = Convert.ToDouble(CommandLine["lower"]);
}
if (CommandLine["upper"] != null) {
LoopCriteria.LoopPairDistanceUpperLimit = Convert.ToDouble(CommandLine["upper"]);
}
}
else {
LoopCriteria.SearchForLoopPairs = false;
}
if (CommandLine["out"] != null) {
OutputFilename = CommandLine["out"];
}
else {
OutputFilename = "ScaffoldFinderLog_" + DateTime.Now.ToString("yyyyMMddHHmmss") + ".txt";
}
string LogFilename = @"c:\Archives\Output\" + OutputFilename;
FileStream LogStream = File.Open(LogFilename,
FileMode.CreateNew,
FileAccess.Write);
StreamWriter LogFile = new StreamWriter(LogStream, Encoding.UTF8);
//Write the lof file header
if (LoopCriteria.SearchForLoopPairs) {
LogFile.WriteLine("PDB,ChainA,NTermLoopA,CTermLoopA,Length,ChainB,NTermLoopB,CTermLoopB,LengthB,Size");
}
else if (LoopCriteria.SearchForTrimer)
{
LogFile.WriteLine("PDB,ChainA,NTermLoopA,CTermLoopA,Length,ChainB,NTermLoopB,CTermLoopB,LengthB,ChainC,NTermLoopC,CTermLoopC,LengthC");
}
else
{
LogFile.WriteLine("PDB,Chain,NTerm,CTerm,Length");
}
//Loop through the directory to get the PDBs
DirectoryInfo Directory;
if (useMultimeric) {
Directory = new DirectoryInfo(PdbMultimerDirectory);
}
else {
Directory = new DirectoryInfo(PdbDirectory);
}
int FileCounter = 0;
int FileTotals = Directory.GetFiles().Count();
foreach (FileInfo file in Directory.GetFiles())
{
FileCounter++;
if (FileCounter % 500 == 0) {
Console.WriteLine("[" + FileCounter + "/" + FileTotals + "]: " + file.FullName);
}
//Console.SetCursorPosition(0, Console.CursorTop);
//Console.WriteLine("Filename: " + file.Name + " Size: " + file.Length);
if (file.Length > 500000)
{
continue;
}
IPdb Pdb = new IPdb(file.FullName, useMultimeric);
ILoopsCollection LoopsCollection = new ILoopsCollection();
if (Pdb.Atoms.Count() != 0) { // Implies Nucleic Acid
FindLoops findLoops = new FindLoops();
LoopsCollection = findLoops.Search(Pdb);
//LoopsCollection = FindLoops.Search(Pdb);
findLoops = null;
}
else {
continue;
}
//Search for three loops that match the trimer constraints
if(LoopCriteria.SearchForTrimer) {
FindLoops findLoops = new FindLoops();
List<ILoopTrimer> LoopTrimer = findLoops.FindTrimerLoops(LoopsCollection, LoopCriteria);
foreach (ILoopTrimer Loops in LoopTrimer)
{
LogFile.WriteLine(Loops.LoopA.PdbCode + "," +
Loops.LoopA.Chain + "," +
Loops.LoopA.NTermAtom.ResidueNumber + Loops.LoopA.NTermAtom.ResidueExtension + "," +
Loops.LoopA.CTermAtom.ResidueNumber + Loops.LoopA.CTermAtom.ResidueExtension + "," +
Loops.LoopA.LoopLength + "," +
Loops.LoopB.Chain + "," +
Loops.LoopB.NTermAtom.ResidueNumber + Loops.LoopB.NTermAtom.ResidueExtension + "," +
Loops.LoopB.CTermAtom.ResidueNumber + Loops.LoopB.CTermAtom.ResidueExtension + "," +
Loops.LoopB.LoopLength + "," +
Loops.LoopC.Chain + "," +
Loops.LoopC.NTermAtom.ResidueNumber + Loops.LoopC.NTermAtom.ResidueExtension + "," +
Loops.LoopC.CTermAtom.ResidueNumber + Loops.LoopC.CTermAtom.ResidueExtension + "," +
Loops.LoopC.LoopLength);
}
LoopTrimer = null;
findLoops = null;
}
else if (LoopCriteria.SearchForLoopPairs) {
FindLoops findLoops = new FindLoops();
List<ILoopPair> LoopPairs = findLoops.FindMatchedLoops(LoopsCollection, LoopCriteria);
foreach (ILoopPair LoopPair in LoopPairs)
{
Console.WriteLine("Found: " + LoopPair.LoopA.PdbCode);
LogFile.WriteLine(LoopPair.LoopA.PdbCode + "," +
LoopPair.LoopA.Chain + "," +
LoopPair.LoopA.NTermAtom.ResidueNumber + LoopPair.LoopA.NTermAtom.ResidueExtension + "," +
LoopPair.LoopA.CTermAtom.ResidueNumber + LoopPair.LoopA.CTermAtom.ResidueExtension + "," +
LoopPair.LoopA.LoopLength + "," +
LoopPair.LoopB.Chain + "," +
LoopPair.LoopB.NTermAtom.ResidueNumber + LoopPair.LoopB.NTermAtom.ResidueExtension + "," +
LoopPair.LoopB.CTermAtom.ResidueNumber + LoopPair.LoopB.CTermAtom.ResidueExtension + "," +
LoopPair.LoopB.LoopLength + "," +
Pdb.Atoms.Count()) ;
}
LoopPairs = null;
findLoops = null;
}
else
{
foreach (ILoop Loop in LoopsCollection.Loops)
{
LogFile.WriteLine(Loop.PdbCode + "," +
Loop.Chain + "," +
Loop.NTermAtom.ResidueNumber + Loop.NTermAtom.ResidueExtension + "," +
Loop.CTermAtom.ResidueNumber + Loop.CTermAtom.ResidueExtension + "," +
Loop.LoopLength);
}
//Cleanup
}
}
LogFile.Close();
LogStream.Close();
// Console.ReadLine();
}
public ILoopsCollection Search(IPdb Pdb)
{
ILoopsCollection LoopsCollection = new ILoopsCollection();
ILoopCriteria LoopCriteria = new ILoopCriteria(true);
//Try some LINQ on this beast
var Chains = (from p in Pdb.Atoms select p.Chain).Distinct();
foreach (string ChainName in Chains)
{
//Get all the atoms in the chain
var Chain = (from p in Pdb.Atoms where p.Chain == ChainName select p);
IResidue[] AtomsInChain = Chain.ToArray();
//Set the chain length if it is not explicitly set (full length here)
if(LoopCriteria.Start == -1) { LoopCriteria.Start = 0; }
if(LoopCriteria.End == -1) { LoopCriteria.End = AtomsInChain.Count(); }
//Search for the loops
if (LoopCriteria.Length > AtomsInChain.Count()) { continue; }
for (int i = LoopCriteria.Start; i < LoopCriteria.End - LoopCriteria.Length-1; i++)
{
//Check for the terminal distances
double eDistance = EuclideanDistance(AtomsInChain[i], AtomsInChain[i + LoopCriteria.Length]);
if (eDistance < LoopCriteria.TerminiDistanceThreshold) {
//Passes the Distance Test
//Check for the solvent accessible regions
if (LoopCriteria.SolventAccessible) {
List<IResidue> LoopResidues = new List<IResidue>();
for (int j = i; j < i + LoopCriteria.Length; j++) { LoopResidues.Add(AtomsInChain[j]); }
IResidue ApexResidue = IsLoopVectorSolventAccessible(Pdb.Atoms, LoopResidues, 4, LoopCriteria);
//IResidue ApexResidue = IsLoopSolventAccessible(Pdb.Atoms, LoopResidues, LoopCriteria);
if (ApexResidue!=null) {
//Passes the solvent test, add it to the loop collection
ILoop Loop = new ILoop();
Loop.NTermAtom = AtomsInChain[i];
Loop.NTermAtomP1 = AtomsInChain[i + 1];
Loop.NTermAtomP2 = AtomsInChain[i + 2];
Loop.CTermAtom = AtomsInChain[i + LoopCriteria.Length];
Loop.CTermAtomP1 = AtomsInChain[i + LoopCriteria.Length - 1];
Loop.CTermAtomP2 = AtomsInChain[i + LoopCriteria.Length - 2];
Loop.LoopLength = LoopCriteria.Length;
Loop.PdbCode = Pdb.PdbCode;
Loop.Chain = ChainName;
Loop.ApexAtom = ApexResidue;
Loop.UnitVector = new Vector3D(((Loop.NTermAtomP2.X + Loop.CTermAtomP2.X) / 2) - ((Loop.NTermAtom.X + Loop.CTermAtom.X) / 2),
((Loop.NTermAtomP2.Y + Loop.CTermAtomP2.Y) / 2) - ((Loop.NTermAtom.Y + Loop.CTermAtom.Y) / 2),
((Loop.NTermAtomP2.X + Loop.CTermAtomP2.X) / 2) - ((Loop.NTermAtom.X + Loop.CTermAtom.X) / 2));
//Ensure Beta-Loop-Beta
if (EuclideanDistance(Loop.NTermAtom, Loop.CTermAtom) < LoopCriteria.TerminiDistanceThreshold &&
EuclideanDistance(Loop.NTermAtomP1, Loop.CTermAtomP1) < LoopCriteria.TerminiDistanceThreshold &&
EuclideanDistance(Loop.NTermAtomP2, Loop.CTermAtomP2) < LoopCriteria.TerminiDistanceThreshold)
{
LoopsCollection.Add(Loop);
}
}
}
}
}
LoopCriteria.End = -1;
LoopCriteria.Start = -1;
}
//Cleanup
LoopCriteria = null;
//Return
return CleanupLoops(LoopsCollection);
}