public static List <Sequence> LoadStructureFile(string atomsFilename, char[] chainIdWhiteList = null, bool padMissingBool = true, int[] startResSeq = null, int[] endResSeq = null, char outsidePaddingChar = ' ', char insidePaddingChar = 'X')
{
var pdb = ProteinBioClass.PdbAtomicChains(atomsFilename, chainIdWhiteList, -1, -1, true);
var pdbId = ProteinBioClass.PdbIdFromPdbFilename(atomsFilename);
return(LoadStructureFile(pdb, pdbId, chainIdWhiteList, padMissingBool, startResSeq, endResSeq,
outsidePaddingChar, insidePaddingChar));
}
public static List <SequenceIdentityClusterMember> ClusterSequenceByAlignedSequenceIdentity(List <Sequence> seqList, ProteinBioClass.AlignmentIdentityOption alignmentIdentityOption, decimal mininumClusterPairwiseSimilarity = 0.3m, decimal mininumEvoClusterPairwiseSimilarity = 0.3m)
{
var allsequences = seqList.Select(a => new Tuple <string, char, string>(new ProteinBioClass.SequenceId(a.Id).PdbId, new ProteinBioClass.SequenceId(a.Id).ChainId, Sequence.EscapeAminoAcidSequence(a.FullSequence))).ToList();
var sequences = allsequences.Select(a => a.Item3).Distinct().ToList();
var sequenceIds = sequences.Select(a => allsequences.Where(b => b.Item3 == a).ToList()).ToList();
var seqClusters = new List <List <string> >();
for (int x = 0; x < sequences.Count; x++)
{
var seq1 = sequences[x];
var newCluster = new List <string>();
newCluster.Add(seq1);
seqClusters.Add(newCluster);
}
for (int indexX = 0; indexX < sequences.Count; indexX++)
{
Console.WriteLine("Aligning sequence " + indexX);
var seqX = sequences[indexX];
//List<decimal> scoreList = new List<decimal>();
//List<decimal> scoreEvoList = new List<decimal>();
for (int indexY = 0; indexY < sequences.Count; indexY++)
{
if (indexY <= indexX)
{
continue;
}
var seqY = sequences[indexY];
if ((decimal)Math.Min(seqX.Length, seqY.Length) / (decimal)Math.Max(seqX.Length, seqY.Length) < mininumClusterPairwiseSimilarity)
{
continue;
}
var cluster1 = seqClusters.FirstOrDefault(a => a.Contains(seqX));
var cluster2 = seqClusters.FirstOrDefault(a => a.Contains(seqY));
if (cluster1 != null && cluster2 != null && cluster1 == cluster2)
{
continue;
}
var score = ProteinBioClass.AlignedSequenceSimilarityPercentage(seqX, seqY, ProteinBioClass.AlignmentType.NON, alignmentIdentityOption);
Console.WriteLine("1: " + seqX);
Console.WriteLine("2: " + seqY);
Console.WriteLine("Score1: " + score.Score);
Console.WriteLine("Score2: " + score.ScoreEvo);
if (score.Score < mininumClusterPairwiseSimilarity || score.ScoreEvo < mininumEvoClusterPairwiseSimilarity)
{
var x = ProteinBioClass.AlignedSequenceSimilarityPercentage(seqX, seqY, ProteinBioClass.AlignmentType.SIM, alignmentIdentityOption);
if (x.Score > score.Score)
{
score.Score = x.Score;
}
if (x.ScoreEvo > score.ScoreEvo)
{
score.ScoreEvo = x.ScoreEvo;
}
}
if (score.Score < mininumClusterPairwiseSimilarity || score.ScoreEvo < mininumEvoClusterPairwiseSimilarity)
{
var x = ProteinBioClass.AlignedSequenceSimilarityPercentage(seqX, seqY, ProteinBioClass.AlignmentType.NMW, alignmentIdentityOption);
if (x.Score > score.Score)
{
score = x;
}
if (x.ScoreEvo > score.ScoreEvo)
{
score.ScoreEvo = x.ScoreEvo;
}
}
if (score.Score >= mininumClusterPairwiseSimilarity && score.ScoreEvo >= mininumEvoClusterPairwiseSimilarity)
{
var newCluster = new List <string>();
newCluster.AddRange(cluster1);
newCluster.AddRange(cluster2);
seqClusters.Remove(cluster1);
seqClusters.Remove(cluster2);
seqClusters.Add(newCluster);
}
//scoreList.Add(score.Score);
//scoreEvoList.Add(score.ScoreEvo);
}
//Console.WriteLine("[" + string.Join(", ", scoreList.Select(a => String.Format("{0:0.00}", a)).ToList()) + "]");
//Console.WriteLine("[" + string.Join(", ", scoreEvoList.Select(a => String.Format("{0:0.00}", a)).ToList()) + "]");
}
seqClusters = seqClusters.OrderBy(a => a.Count).ToList();
var output = new List <SequenceIdentityClusterMember>();
for (var index = 0; index < seqClusters.Count; index++)
{
var seqCluster = seqClusters[index];
foreach (var item in seqCluster)
{
var indexIds = sequences.IndexOf(item);
var ids = sequenceIds[indexIds];
foreach (var id in ids)
{
output.Add(new SequenceIdentityClusterMember(index + 1, ProteinBioClass.PdbIdFromPdbFilename(id.Item1), id.Item2, id.Item3));
}
}
}
return(output);
}
public static string LoadDsspStructureSequence(string pdbFilename, string chainId = null, int startResidueSequenceIndex = -1, int endResidueSequenceIndex = -1, bool reversedSequence = false)
{
if (string.IsNullOrWhiteSpace(pdbFilename))
{
return("");
}
var pdbId = ProteinBioClass.PdbIdFromPdbFilename(pdbFilename);
var dsspFilename = pdbFilename;
if (Path.GetExtension(dsspFilename) != ".dssp")
{
dsspFilename += ".dssp";
}
if (!File.Exists(dsspFilename))
{
return("");
}
var secondaryStructure = DsspFormatFile.LoadDsspFile(dsspFilename);
if (chainId != null && secondaryStructure.FirstOrDefault(a => a.FieldChain.FieldValue.ToUpperInvariant() == chainId.ToUpperInvariant()) == null)
{
return("");
}
if (startResidueSequenceIndex == -1)
{
startResidueSequenceIndex = secondaryStructure.Where(a => chainId == null || a.FieldChain.FieldValue.ToUpperInvariant() == chainId.ToUpperInvariant()).Min(a => int.Parse(a.FieldPdbResidueSequenceIndex.FieldValue));
}
if (endResidueSequenceIndex == -1)
{
endResidueSequenceIndex = secondaryStructure.Where(a => chainId == null || a.FieldChain.FieldValue.ToUpperInvariant() == chainId.ToUpperInvariant()).Max(a => int.Parse(a.FieldPdbResidueSequenceIndex.FieldValue));
}
// dssp specification says order may not be correct
secondaryStructure = secondaryStructure.Where(a => !string.IsNullOrWhiteSpace(a.FieldChain.FieldValue) && !string.IsNullOrWhiteSpace(a.FieldPdbResidueSequenceIndex.FieldValue)).OrderBy(a => a.FieldChain.FieldValue).ThenBy(a => NullableTryParseInt32(a.FieldPdbResidueSequenceIndex.FieldValue)).ToList();
var proteinInterfaceLen = CalculateProteinInterfaceLength(startResidueSequenceIndex, endResidueSequenceIndex);
char[] result = new char[proteinInterfaceLen];
for (int index = 0; index < result.Length; index++)
{
result[index] = '_';
}
foreach (var record in secondaryStructure.Where(a => chainId == null || a.FieldChain.FieldValue.ToUpperInvariant() == chainId.ToUpperInvariant()))
{
var resSeq = NullableTryParseInt32(record.FieldPdbResidueSequenceIndex.FieldValue);
if (resSeq == null || resSeq < startResidueSequenceIndex || resSeq > endResidueSequenceIndex)
{
continue;
}
var position = resSeq - startResidueSequenceIndex;
if (record.FieldSecondaryStructure.FieldValue.Length == 0)
{
continue;
}
result[position.Value] = record.FieldSecondaryStructure.FieldValue[0];
}
if (reversedSequence)
{
Array.Reverse(result);
}
return(new string(result));
}
