/// <summary>
/// Gets GenomicBins by chromosome as an OrderedDictionary. The order of chromosomes in path is preserved.
/// Assumes that bins are sorted in ascending order by the start position within each chromosome.
/// </summary>
/// <param name="path"></param>
/// <returns>OrderedDictionary with string key and List<GenomicBin> value</returns>
public static OrderedDictionary <string, List <SampleGenomicBin> > GetGenomicBinsByChrom(string path)
{
OrderedDictionary <string, List <SampleGenomicBin> > binsByChrom = new OrderedDictionary <string, List <SampleGenomicBin> >();
SampleGenomicBin prevBin = null;
foreach (var bin in IterateThroughTextFile(path))
{
if (!binsByChrom.ContainsKey(bin.GenomicBin.Chromosome))
{
binsByChrom[bin.GenomicBin.Chromosome] = new List <SampleGenomicBin>();
prevBin = null;
}
if (prevBin != null && bin.Start < prevBin.Start)
{
throw new Exception("Bins are not sorted in ascending order by the start position." +
$" First offending bin: {bin.GenomicBin.Chromosome}\t{bin.Start}\t{bin.Stop}");
}
binsByChrom[bin.GenomicBin.Chromosome].Add(bin);
prevBin = bin;
}
return(binsByChrom);
}
public static Dictionary <string, List <SampleGenomicBin> > LoadBedFile(string bedPath, int?gcIndex = null)
{
Dictionary <string, List <SampleGenomicBin> > excludedIntervals = new Dictionary <string, List <SampleGenomicBin> >();
int count = 0;
using (FileStream stream = new FileStream(bedPath, FileMode.Open, FileAccess.Read))
using (StreamReader reader = new StreamReader(stream))
{
while (true)
{
string fileLine = reader.ReadLine();
if (fileLine == null)
{
break;
}
string[] bits = fileLine.Split('\t');
string chr = bits[0];
if (!excludedIntervals.ContainsKey(chr))
{
excludedIntervals[chr] = new List <SampleGenomicBin>();
}
SampleGenomicBin interval = new SampleGenomicBin();
interval.GenomicBin.Chromosome = chr;
interval.Start = int.Parse(bits[1]);
interval.Stop = int.Parse(bits[2]);
if (interval.Start < 0)
{
throw new Illumina.Common.IlluminaException(String.Format("Start must be non-negative in a BED file: {0}", fileLine));
}
if (interval.Start >= interval.Stop) // Do not allow empty intervals
{
throw new Illumina.Common.IlluminaException(String.Format("Start must be less than Stop in a BED file: {0}", fileLine));
}
if (gcIndex.HasValue && gcIndex.Value < bits.Length)
{
interval.GenomicBin.GC = int.Parse(bits[gcIndex.Value]);
}
excludedIntervals[chr].Add(interval);
count++;
}
}
Console.WriteLine(">>> Loaded {0} intervals for {1} sequences", count, excludedIntervals.Keys.Count);
return(excludedIntervals);
}
public static IEnumerable <SampleGenomicBin> IterateThroughTextFile(string infile)
{
using (GzipReader reader = new GzipReader(infile))
{
string row;
while ((row = reader.ReadLine()) != null)
{
string[] fields = row.Split('\t');
string chr = fields[0];
int start = Convert.ToInt32(fields[1]);
int stop = Convert.ToInt32(fields[2]);
float count = float.Parse(fields[3]);
int gc = Convert.ToInt32(fields[4]);
SampleGenomicBin bin = new SampleGenomicBin(chr, start, stop, gc, count);
yield return(bin);
}
}
}
public bool IsSameBin(SampleGenomicBin bin)
{
return(GenomicBin.Chromosome == bin.GenomicBin.Chromosome && GenomicBin.Interval.Start == bin.GenomicBin.Interval.Start && GenomicBin.Interval.End == bin.GenomicBin.Interval.End);
}
public bool SkipBin(SampleGenomicBin bin)
{
return(SkipBin(bin.GenomicBin.Chromosome, (uint)bin.Start, (uint)bin.Stop));
}