public static SNPInfo Read(BinaryReader reader)
{
var result = new SNPInfo();
result.id = reader.ReadString();
result.descriptionWikiText = reader.ReadString();
result.gene = reader.ReadString();
result.chromosome = reader.ReadString();
result.position = reader.ReadInt32();
result.orientation = (Orientation)reader.ReadByte();
result.updateTime = DateTime.FromBinary(reader.ReadInt64());
result.genotypes = new SNPGenotypeInfo[reader.ReadInt32()];
for (int genotypeIndex = 0; genotypeIndex < result.genotypes.Length; genotypeIndex++)
{
var genotypeInfo = new SNPGenotypeInfo();
genotypeInfo.genotype = DiploidGenotype.Read(reader);
genotypeInfo.trait = reader.ReadString();
genotypeInfo.populationFrequencies = new Dictionary <string, float>();
int numPopulations = reader.ReadInt32();
for (int populationIndex = 0; populationIndex < numPopulations; populationIndex++)
{
var populationTag = reader.ReadString();
var populationFrequency = reader.ReadSingle();
genotypeInfo.populationFrequencies.Add(populationTag, populationFrequency);
}
result.genotypes[genotypeIndex] = genotypeInfo;
}
return(result);
}
/** Determines the orientation of the genotypes of a SNP. */
public Orientation GetOrientation(SNPInfo snpInfo)
{
// Check whether any of the SNP's genotypes and their complements don't match the valid alleles for this orientation.
var matches = new bool[2] {
true, true
};
foreach (var genotypeInfo in snpInfo.genotypes)
{
// Determine whether this genotype or its complement matches the valid alleles for this orientation.
var orientedGenotypes = new DiploidGenotype[]
{
genotypeInfo.genotype,
genotypeInfo.genotype.GetComplement()
};
for (int tryIndex = 0; tryIndex < orientedGenotypes.Length; tryIndex++)
{
if (!DoesGenotypeMatch(orientedGenotypes[tryIndex].a) ||
!DoesGenotypeMatch(orientedGenotypes[tryIndex].b))
{
matches[tryIndex] = false;
}
}
}
if (matches[0] && !matches[1])
{
// If the SNP's genotypes all match this orientation's valid alleles, they have the same orientation.
return(orientation);
}
else if (matches[1] && !matches[0])
{
// If the SNP's genotypes' complements all match this orientation's valid alleles, the SNP's genotypes have the opposite orientation.
return(GetOppositeOrientation());
}
else
{
// If none of the SNP's genotypes or their complements mismatch this orientation's alleles, we can't determine the orientation of the SNP's genotypes.
return(Orientation.Unknown);
}
}
public SimpleDiploidTraitPage(
SNPInfo snpInfo,
DiploidGenotype personalGenotype
)
{
InitializeComponent();
// Setup the SNP information controls.
nameLabel.Content = snpInfo.id;
descriptionLabel.Text = Utilities.ConvertWikiTextToPlainText(snpInfo.descriptionWikiText);
snpediaLink.NavigateUri = new Uri(string.Format("http://www.snpedia.com/index.php?title={0}", snpInfo.id));
// Setup the list of genotypes for this SNP.
bHasMatchingGenotype = false;
foreach (var genotypeInfo in snpInfo.genotypes)
{
bool bGenotypeMatchesPersonalGenome = personalGenotype.Equals(genotypeInfo.genotype);
if (bGenotypeMatchesPersonalGenome)
{
bHasMatchingGenotype = true;
}
genotypeList.Items.Add(new SNPGenotypeUIAdapter(
genotypeInfo,
bGenotypeMatchesPersonalGenome
));
}
// If the genome doesn't match any of the genotypes, create a placeholder genotype for it.
if (!bHasMatchingGenotype)
{
var genotypeInfo = new SNPGenotypeInfo();
genotypeInfo.genotype = personalGenotype;
genotypeInfo.trait = "";
genotypeInfo.populationFrequencies = new Dictionary <string, float>();
genotypeList.Items.Add(new SNPGenotypeUIAdapter(
genotypeInfo,
true
));
}
}
public static bool Load(Stream stream, ref SNPDatabase outResult)
{
var reader = new BinaryReader(stream);
// Read the file magic and version.
char[] fileMagic = reader.ReadChars(referenceFileMagic.Length);
// If the file doesn't have the expected magic header, abort and return an error.
if (!Utilities.ArrayCompare(fileMagic, referenceFileMagic))
{
return(false);
}
// Create the SNP info database that's about to be loaded.
outResult = new SNPDatabase();
// Read the SNPs in the database.
int numSNPs = reader.ReadInt32();
for (int snpIndex = 0; snpIndex < numSNPs; snpIndex++)
{
// Read a SNP ID and value pair, and add them to the database.
var Key = reader.ReadString();
var Value = SNPInfo.Read(reader);
outResult.snpToInfoMap.Add(Key, Value);
}
// Read the traits in the database.
int numTraits = reader.ReadInt32();
for (int traitIndex = 0; traitIndex < numTraits; traitIndex++)
{
outResult.traits.Add(TraitInfo.Read(reader));
}
return(true);
}
private static void CreateTraitPage(string snpId, SNPInfo snpInfo, ref List <UIElement> genotypedPageList, ref List <UIElement> ungenotypedPageList)
{
// Check if the current genome database has a genotype for this SNP.
var value = App.document.GetSNPValue(snpId);
var genotype = value != null?
value.Value.GetOrientedGenotype(snpInfo.orientation) :
new DiploidGenotype(Genotype.Unknown, Genotype.Unknown);
// Create the trait page for this SNP.
var page = new SimpleDiploidTraitPage(
snpInfo,
genotype
);
// Add the trait page to the appropriate list depending on whether there's a genotype for it.
if (page.bHasMatchingGenotype)
{
genotypedPageList.Add(page);
}
else
{
ungenotypedPageList.Add(page);
}
}
private static SNPInfo?ParseSNPPage(string pageText)
{
// Check if this page is a SNP.
string snpSubstring = ParsePageCategorySubstring(pageText, "rsnum");
if (snpSubstring != null)
{
// Parse this SNP's properties.
var result = new SNPInfo();
result.id = string.Format("rs{0}", ParseCategoryProperty(snpSubstring, "rsid"));
result.gene = ParseCategoryProperty(snpSubstring, "gene");
result.chromosome = ParseCategoryProperty(snpSubstring, "chromosome");
if (!int.TryParse(ParseCategoryProperty(snpSubstring, "position"), out result.position))
{
result.position = -1;
}
result.orientation = Orientation.Unknown;
result.updateTime = DateTime.Today;
// If the page is empty, don't add it to the database.
result.descriptionWikiText = pageText;
if (Utilities.ConvertWikiTextToPlainText(pageText).Trim().Length == 0)
{
return(null);
}
// Parse the SNP's genotypes.
var tempGenotypes = new List <SNPGenotypeInfo>();
for (int genotypeIndex = 0;; genotypeIndex++)
{
string genotypeString = ParseCategoryProperty(snpSubstring, string.Format("geno{0}", genotypeIndex + 1));
if (genotypeString == "")
{
break;
}
else
{
// SNPedia represents deletion genotypes as a '-', but we use D.
genotypeString = genotypeString.Replace('-', 'D');
// Parse the genotype info.
var newGenotype = new SNPGenotypeInfo();
newGenotype.genotype = DNA.StringToDiploidGenotype(genotypeString);
newGenotype.trait = "";
newGenotype.populationFrequencies = new Dictionary <string, float>();
tempGenotypes.Add(newGenotype);
// If any of the genotypes are deletions or unknown, don't add this snp to the database as we can't correctly parse the inserted genotype yet.
if (newGenotype.genotype.a == Genotype.Unknown || newGenotype.genotype.a == Genotype.Deletion ||
newGenotype.genotype.b == Genotype.Unknown || newGenotype.genotype.b == Genotype.Deletion)
{
return(null);
}
}
}
result.genotypes = tempGenotypes.ToArray();
// Check if the page has SNP population frequency data.
var populationFrequencySubstring = ParsePageCategorySubstring(pageText, "population diversity");
if (populationFrequencySubstring != null)
{
// Remap the genotype indices to match the genotypes declared above.
var genotypeIndexRemap = new int?[result.genotypes.Length];
for (int genotypeIndex = 0; genotypeIndex < result.genotypes.Length; genotypeIndex++)
{
genotypeIndexRemap[genotypeIndex] = null;
var genotype = DNA.StringToDiploidGenotype(ParseCategoryProperty(populationFrequencySubstring, string.Format("geno{0}", genotypeIndex + 1)));
for (int genotypeInfoIndex = 0; genotypeInfoIndex < result.genotypes.Length; genotypeInfoIndex++)
{
if (result.genotypes[genotypeInfoIndex].genotype.Equals(genotype))
{
genotypeIndexRemap[genotypeIndex] = genotypeInfoIndex;
break;
}
}
}
// Ignore population frequencies if they are redundantly defined or not defined for any genotypes.
bool bValidPopulationData = true;
for (int genotypeIndex = 0; genotypeIndex < result.genotypes.Length; genotypeIndex++)
{
int numRemaps = 0;
for (int remapIndex = 0; remapIndex < genotypeIndexRemap.Length; remapIndex++)
{
if (genotypeIndexRemap[remapIndex] == genotypeIndex)
{
numRemaps++;
}
}
if (numRemaps != 1)
{
bValidPopulationData = false;
}
}
if (bValidPopulationData)
{
// Parse each population's frequencies for this SNP.
string[] populations = new string[] { "CEU", "HCB", "JPT", "YRI" };
foreach (var population in populations)
{
var frequencies = ParseList(ParseCategoryProperty(populationFrequencySubstring, population));
if (frequencies.Count == result.genotypes.Length)
{
for (int genotypeIndex = 0; genotypeIndex < genotypeIndexRemap.Length; genotypeIndex++)
{
if (genotypeIndexRemap[genotypeIndex] != null)
{
float frequency;
if (float.TryParse(frequencies[genotypeIndex], out frequency))
{
result.genotypes[genotypeIndexRemap[genotypeIndex].Value].populationFrequencies.Add(population, frequency / 100.0f);
}
}
}
}
}
}
}
return(result);
}
else
{
return(null);
}
}
