static PhasedGenomeFile SimplePhase(string filename, GenomeFile gfile, SnpCollection refSnps) {
Console.Write("Creating simple phased genome...");
PhasedGenome phased = new PhasedGenome(gfile.Genome);
Console.WriteLine("completed");
PhasedGenomeFile pfile = new PhasedGenomeFile(GetPhasedFilename(filename));
pfile.SetStandardComments();
pfile.AddComment("## history");
pfile.AddComment("## " + DateTime.Now.ToString() + " SimplePhase of " + phased.Count.ToString("#,##0")
+ " homozygous SNPs (of " + gfile.Genome.Count.ToString("#,##0") + " total).");
pfile.PhasedGenome = phased;
return pfile;
}
public static void AddHistory(PhasedGenomeFile pfile, GenomeFile gfile, int count, IList<Tuple<SegmentMatch, SegmentMatch>> segs) {
DateTime now = DateTime.Now;
pfile.AddComment("## " + DateTime.Now.ToString() + " MatchPhase added " + count.ToString("#,##0") + " phased heterozygous SNPs using " + gfile.Name + ".");
if ((segs != null) && (segs.Count > 0)) {
string filename = Path.Combine(Path.GetDirectoryName(pfile.Filename), pfile.Name + "HISTORY.csv");
List<string> lines = new List<string>();
if (!File.Exists(filename)) {
lines.Add("date,time,source,chromosome,origStart,origEnd,origLen(cM),sides,usedStart,usedEnd,usedLen(cM),newlyPhasedCount");
}
foreach (var tup in segs) {
var match = tup.Item1;
var phased = tup.Item2;
string line = "\"" + now.ToShortDateString() + "\",\"" + now.ToShortTimeString() + "\",\"" + gfile.Name + "\","
+ Snp.ChromosomeToString(match.StartSnp.Chromosome) + "," + match.StartSnp.Position.ToString() + ","
+ match.EndSnp.Position.ToString() + "," + match.CmLength.ToString("#0.00") + ","
+ match.MatchKind.ToString() + "," + phased.StartSnp.Position.ToString() + ","
+ phased.EndSnp.Position.ToString() + "," + phased.CmLength.ToString("#0.00") + ","
+ phased.PhasedSnpCount.ToString();
lines.Add(line);
}
File.AppendAllLines(filename, lines);
}
}
public static void WriteGenone(GenomeFile genomeFile) {
genomeFile.Extension = "txt";
Console.Write("Writing genome " + genomeFile.Name + "...");
genomeFile.Write(true, Configuration.UnphasedChar, new System.Threading.CancellationToken(), null);
Console.WriteLine("completed");
}
public static GenomeFile ReadGenome(SnpCollection refSnps, string filename) {
GenomeFile result = new GenomeFile(filename);
result.Genome = new Genome(1, 23);
Console.Write("Reading genome " + result.Name + "...");
result.Read(refSnps, new System.Threading.CancellationToken(), null);
Console.WriteLine("completed");
return result;
}
public static void SplitClone(string[] args) {
if (args.Length != 2) {
Console.WriteLine("invalid arguments");
ShowHelpSplitClone();
return;
}
SnpCollection refSnps = ReadRefSnps();
string path = Configuration.GenomePath;
string file1 = Path.Combine(path, args[1]);
var full = ReadGenome(refSnps, file1);
var phased1 = ReadPhasedGenome(refSnps, file1);
if (phased1 == null) {
phased1 = SimplePhase(file1, full, refSnps);
WritePhasedGenome(phased1);
}
Console.Write("Splitting and cloning phased genome...");
var result = phased1.PhasedGenome.Split(true);
Console.WriteLine("completed");
var split = new GenomeFile(phased1.Filename.Replace("phased-", "phasedAA-"));
split.Genome = result.Item1;
split.AddComment("# this file is the phased A side split and cloned from " + phased1.Name);
WriteGenone(split);
split = new GenomeFile(phased1.Filename.Replace("phased-", "phasedBB-"));
split.Genome = result.Item2;
split.AddComment("# this file is the phased B side split and cloned from " + phased1.Name);
WriteGenone(split);
}
static void Match(string[] args) {
if (args.Length < 4) {
Console.WriteLine("invalid arguments");
ShowHelpMatch();
return;
}
SnpCollection refSnps = ReadRefSnps();
string path = Configuration.GenomePath;
string lastFilename = Path.Combine(path, args[args.Length - 1]);
var testGenomeFile = new GenomeFile(lastFilename);
if (File.Exists(testGenomeFile.Filename)) {
try {
testGenomeFile.Read(refSnps, new System.Threading.CancellationToken(), null);
if (testGenomeFile.Genome.Count > 0) {
int i = 1;
while (File.Exists(Path.Combine(path, "MatchResults(" + i + ").csv"))) i++;
string filename = Path.Combine(path, "MatchResults(" + i + ").csv");
Console.WriteLine("** no filename was specified for the results");
Console.WriteLine("** results will be written to " + filename);
string[] args2 = new string[args.Length + 1];
for (int k = 0; k < args.Length; k++) args2[k] = args[k];
args = args2;
args[args.Length - 1] = filename;
}
}
catch { }
}
string[] filenames = new string[args.Length - 2];
for (int i = 0; i < filenames.Length; i++) {
filenames[i] = args[i + 1];
}
List<Tuple<string, string, SegmentMatch>> results = new List<Tuple<string, string, SegmentMatch>>();
for (int i = 0; i < filenames.Length -1; i++) {
string file1 = Path.Combine(path, filenames[i]);
var phased1 = ReadPhasedGenome(refSnps, file1);
if (phased1 == null) {
var full = ReadGenome(refSnps, file1);
phased1 = SimplePhase(file1, full, refSnps);
WritePhasedGenome(phased1);
}
for (int j = i + 1; j < filenames.Length; j++) {
string file2 = Path.Combine(path, filenames[j]);
var phased2 = ReadPhasedGenome(refSnps, file2);
if (phased2 == null) {
var full = ReadGenome(refSnps, file2);
phased2 = SimplePhase(file2, full, refSnps);
WritePhasedGenome(phased2);
}
Console.Write("Finding matching segments...");
var result = phased1.PhasedGenome.MatchingSegments(phased2.PhasedGenome, Configuration.MaxErrorsToStitch,
Configuration.SegmentMinPhasedSnpCount, Configuration.SegmentMinCmLength,
Configuration.StitchMinPhasedSnpCount, Configuration.StitchMinCmLength);
if (result.Count > 0) {
foreach (var seg in result) results.Add(new Tuple<string, string, SegmentMatch>(phased1.Name, phased2.Name, seg));
} else {
results.Add(new Tuple<string, string, SegmentMatch>(phased1.Name, phased2.Name, default(SegmentMatch)));
}
Console.WriteLine("completed");
}
}
Console.Write("Writing match results to " + args[3] + "...");
WriteSegmentMatches(results, Path.Combine(path, args[args.Length - 1]));
Console.WriteLine("completed");
}
