private static void SequenceAlignment()
{
if (!Directory.Exists(OutputDirectory))
{
Directory.CreateDirectory(OutputDirectory);
}
var peptides = PeptideFileReader.ReadPeptides(@"G:\Projects\HumanGenome\Homo_sapiens.GRCh38.pep.all.fa");
string lastChromosome = null;
string chromosomeData = null;
foreach (var peptide in peptides.OrderBy(pep => pep.Chromosome))
{
Console.WriteLine(peptide.GeneSymbol + ", chromosome " + peptide.Chromosome);
var chromosomeFile = Directory.GetFiles(ChromosomeDataDirectory, $"*chromosome_{peptide.Chromosome}.*").Single();
if (chromosomeData == null || peptide.Chromosome != lastChromosome)
{
chromosomeData = File.ReadAllText(chromosomeFile);
lastChromosome = peptide.Chromosome;
}
var nucleotides = chromosomeData.Substring(peptide.StartBase - 1, peptide.EndBase - peptide.StartBase + 1);
var peptideDescription = $"{peptide.GeneSymbol}:{peptide.Chromosome}:{peptide.StartBase}:{peptide.EndBase}";
try
{
var exons = ExonExtractor.ExtractExons(nucleotides, new string(peptide.Sequence.ToArray()));
var alignedSequence = BuildAlignedSequence(nucleotides, exons);
var statistics = Environment.NewLine
+ "Statistics:" + Environment.NewLine
+ "-------------------------------" + Environment.NewLine
+ "Exon count: " + exons.Count + Environment.NewLine
+ "Shortest exon: " + exons.Min(e => e.AminoAcids.Count) + Environment.NewLine
+ "Longest exon: " + exons.Max(e => e.AminoAcids.Count) + Environment.NewLine
+ "Median exon length: " + exons.Median(e => e.AminoAcids.Count);
File.AppendAllLines(Path.Combine(OutputDirectory, peptide.GeneSymbol + ".txt"), new[]
{
peptideDescription,
InterleaveLines(SpliceText(nucleotides, 120), SpliceText(alignedSequence, 120))
.Aggregate((line1, line2) => line1 + Environment.NewLine + line2),
statistics,
Environment.NewLine
});
}
catch (Exception ex)
{
File.AppendAllLines(Path.Combine(OutputDirectory, peptide.GeneSymbol + ".txt"), new[]
{
peptideDescription,
ex.Message,
Environment.NewLine
});
}
}
}
private static void WellAlignedSequences()
{
var peptides = PeptideFileReader.ReadPeptides(@"G:\Projects\HumanGenome\Homo_sapiens.GRCh38.pep.all.fa");
string lastChromosome = null;
string chromosomeData = null;
const int longSequenceThreshold = 10;
foreach (var peptide in peptides.OrderBy(pep => pep.Chromosome))
{
Console.WriteLine(peptide.GeneSymbol + ", chromosome " + peptide.Chromosome);
var chromosomeFile = Directory.GetFiles(ChromosomeDataDirectory, $"*chromosome_{peptide.Chromosome}.*").Single();
if (chromosomeData == null || peptide.Chromosome != lastChromosome)
{
chromosomeData = File.ReadAllText(chromosomeFile);
lastChromosome = peptide.Chromosome;
}
var nucleotides = chromosomeData.Substring(peptide.StartBase - 1, peptide.EndBase - peptide.StartBase + 1);
var peptideDescription = $"{peptide.GeneSymbol}:{peptide.Chromosome}:{peptide.StartBase}:{peptide.EndBase}";
try
{
if (!peptide.Sequence.Any())
{
continue;
}
if (peptide.Sequence.First() != 'M')
{
continue;
}
var exons = ExonExtractor.ExtractExons(nucleotides, new string(peptide.Sequence.ToArray()));
var sequenceLengths = exons.Select(e => e.AminoAcids.Count).ToList();
var longSequenceMarkers = sequenceLengths.Select(x => x >= longSequenceThreshold).ToList();
var hasContiguousLongSequences = Enumerable.Range(0, longSequenceMarkers.Count - 1)
.Select(idx => longSequenceMarkers[idx] && longSequenceMarkers[idx + 1])
.Any(x => x);
if (!hasContiguousLongSequences)
{
continue;
}
var exonData = peptideDescription + ":" + exons
.Select(e =>
e.StartNucelotideIndex
+ ";"
+ (e.StartNucelotideIndex + 3 * e.AminoAcids.Count - 1))
.Aggregate((a, b) => a + ";" + b);
File.AppendAllText(@"well_matched_peptides.csv", exonData + Environment.NewLine);
}
catch
{
// ignored
}
}
}
private static void TestExonExtraction()
{
var testPeptideSequence = "MAGVRTTGKLTT";
var testNucleotides = NucleotideFromPeptideSequence("IIMAGVLI#ARTTIIVLRTTGKLRTT#ITT");
var testExons = ExonExtractor.ExtractExons(testNucleotides, testPeptideSequence);
var testAlignedSequence = BuildAlignedSequence(testNucleotides, testExons);
File.WriteAllLines(@"peptide_alignment.txt", new[]
{
testNucleotides,
testAlignedSequence
});
}