public static bool ImportDatabase(Stream stream)
{
SNPDatabase newDatabase = null;
if (SNPDatabase.Load(stream, ref newDatabase))
{
localDatabase = newDatabase;
return(true);
}
else
{
return(false);
}
}
public static async Task <SNPDatabase> CreateSNPDatabaseAsync(UpdateProgressDelegate updateProgressDelegate, CancellationToken cancellationToken)
{
var result = new SNPDatabase();
try
{
await UpdateSNPDatabaseAsync(result, updateProgressDelegate, cancellationToken);
return(result);
}
catch (TaskCanceledException)
{
return(null);
}
}
private static SNPDatabase InternalLoadDatabase()
{
// Load the default database.
using (var defaultDatabaseStream = new MemoryStream(Properties.Resources.DefaultSNPDatabase, false))
{
SNPDatabase result = null;
if (SNPDatabase.Load(defaultDatabaseStream, ref result))
{
return(result);
}
else
{
return(new SNPDatabase());
}
}
}
private static async Task UpdatePages(SNPDatabase database, IEnumerable <string> pageTitles, CancellationToken cancellationToken)
{
var queryResponse = await CallMediaWikiAPI(string.Format(
"action=query&prop=revisions&rvprop=content&titles={0}&format=xml",
HttpUtility.UrlEncode(string.Join("|", pageTitles))
), cancellationToken);
if (!cancellationToken.IsCancellationRequested && queryResponse.bSuccess)
{
// Process each page received in the response.
var pageElements = queryResponse.resultXML.SelectNodes("query/pages/page");
foreach (XmlNode pageElement in pageElements)
{
var pageTitle = pageElement.Attributes["title"].Value;
var pageRevision = pageElement.SelectSingleNode("revisions/rev");
ProcessPage(database, pageTitle, pageRevision.InnerText);
}
}
}
public static async Task UpdateSNPDatabaseAsync(SNPDatabase database, UpdateProgressDelegate updateProgressDelegate, CancellationToken cancellationToken)
{
// Read the page list.
updateProgressDelegate("Reading SNPedia page list", 0);
var pageList = await ReadPageListAsync(updateProgressDelegate, cancellationToken);
// Sort the page list to ensure that genotype pages (e.g. rs1234(a;a)) are processed after the corresponding snp page (i.e. rs1234)
pageList.Sort(StringComparer.OrdinalIgnoreCase);
var pagesPerBatch = 50;
for (int pageIndex = 0; pageIndex < pageList.Count; pageIndex += pagesPerBatch)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
await UpdatePages(database, new ArraySegment <string>(pageList.ToArray(), pageIndex, Math.Min(pageList.Count - pageIndex, pagesPerBatch)), cancellationToken);
updateProgressDelegate(string.Format("Processed {0}/{1} pages", pageIndex, pageList.Count), (double)pageIndex / (double)pageList.Count);
}
}
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 ProcessPage(SNPDatabase database, string pageTitle, string pageText)
{
if (pageText != null)
{
// Try to parse the page's SNP category properties.
var snpInfo = ParseSNPPage(pageText);
if (snpInfo != null)
{
// Skip pages with mismatched SNP ID and title.
if (snpInfo.Value.id == pageTitle.ToLowerInvariant())
{
database.snpToInfoMap.Add(snpInfo.Value.id, snpInfo.Value);
}
}
else
{
var genotypeCategoryText = ParsePageCategorySubstring(pageText, "Genotype");
if (genotypeCategoryText != null)
{
// Parse the trait summary for a SNP genotype and add it to the previously created SNPInfo.
var rsid = string.Format("rs{0}", ParseCategoryProperty(genotypeCategoryText, "rsid"));
var genotype = DNA.StringToDiploidGenotype(string.Format("{0};{1}", ParseCategoryProperty(genotypeCategoryText, "allele1"), ParseCategoryProperty(genotypeCategoryText, "allele2")));
var genotypeTraitSummary = ParseCategoryProperty(genotypeCategoryText, "summary");
if (database.snpToInfoMap.ContainsKey(rsid))
{
var snpGenotypes = database.snpToInfoMap[rsid].genotypes;
for (var genotypeIndex = 0; genotypeIndex < snpGenotypes.Length; ++genotypeIndex)
{
if (snpGenotypes[genotypeIndex].genotype.Equals(genotype))
{
snpGenotypes[genotypeIndex].trait = genotypeTraitSummary;
break;
}
}
}
}
else if (ParsePageCategorySubstring(pageText, "is a \\| medical condition") != null ||
ParsePageCategorySubstring(pageText, "is a \\| medicine") != null ||
ParsePageCategorySubstring(pageText, "is a \\| gene") != null)
{
// If the page isn't a SNP, add it to the database as a trait with associated SNPs.
var traitInfo = new TraitInfo();
traitInfo.title = pageTitle;
// Parse links to associated SNPs.
var associatedSNPs = new List <string>();
Utilities.ProcessDelimitedItems(pageText, "[[", "]]", delegate(string itemText)
{
var snpId = itemText.ToLowerInvariant();
if (snpId.StartsWith("rs"))
{
if (!associatedSNPs.Contains(snpId))
{
associatedSNPs.Add(snpId);
}
}
return(itemText);
});
traitInfo.associatedSNPs = associatedSNPs.ToArray();
if (traitInfo.associatedSNPs.Length > 0)
{
// Add the trait to the database.
database.traits.Add(traitInfo);
}
}
}
}
}
public static void RevertDatabase()
{
localDatabase = InternalLoadDatabase();
}