public List<ChromosomeCountSlimItem> Build(string fileName)
{
if (File.Exists(options.CategoryMapFile))
{
Progress.SetMessage("Reading name map file " + options.CategoryMapFile + " ...");
nameMap = new MapItemReader(0, 1).ReadFromFile(options.CategoryMapFile).ToDictionary(m => m.Key, m => m.Value.Value);
}
var result = new List<ChromosomeCountSlimItem>();
var queries = new Dictionary<string, SAMChromosomeItem>();
var chromosomes = new Dictionary<string, ChromosomeCountSlimItem>();
Progress.SetMessage("Parsing alignment file " + fileName + " ...");
using (var sr = SAMFactory.GetReader(fileName, true))
{
int count = 0;
int waitingcount = 0;
string line;
while ((line = sr.ReadLine()) != null)
{
if (count % 1000 == 0)
{
if (Progress.IsCancellationPending())
{
throw new UserTerminatedException();
}
}
if (count % 100000 == 0 && count > 0)
{
Progress.SetMessage("{0} candidates from {1} reads", waitingcount, count);
}
count++;
var parts = line.Split('\t');
SAMFlags flag = (SAMFlags)int.Parse(parts[SAMFormatConst.FLAG_INDEX]);
//unmatched
if (flag.HasFlag(SAMFlags.UnmappedQuery))
{
continue;
}
var qname = parts[SAMFormatConst.QNAME_INDEX];
SAMChromosomeItem query;
if (!queries.TryGetValue(qname, out query))
{
query = new SAMChromosomeItem();
query.Qname = qname;
queries[qname] = query;
if (options.KeepSequence)
{
query.Sequence = parts[SAMFormatConst.SEQ_INDEX];
if (flag.HasFlag(SAMFlags.QueryOnReverseStrand))
{
query.Sequence = SequenceUtils.GetReverseComplementedSequence(query.Sequence);
}
}
}
var seqname = GetName(parts[SAMFormatConst.RNAME_INDEX]);
query.Chromosomes.Add(seqname);
ChromosomeCountSlimItem item;
if (!chromosomes.TryGetValue(seqname, out item))
{
item = new ChromosomeCountSlimItem();
item.Names.Add(seqname);
chromosomes[seqname] = item;
result.Add(item);
}
item.Queries.Add(query);
waitingcount++;
}
Progress.SetMessage("Finally, there are {0} candidates from {1} reads", waitingcount, count);
}
foreach (var query in queries.Values)
{
query.Chromosomes = query.Chromosomes.Distinct().OrderBy(m => m).ToList();
}
foreach (var sam in chromosomes.Values)
{
sam.Queries = sam.Queries.Distinct().OrderBy(m => m.Qname).ToList();
}
if (!string.IsNullOrEmpty(options.PreferPrefix))
{
foreach (var query in queries.Values)
{
if (query.Chromosomes.Any(l => l.StartsWith(options.PreferPrefix)))
{
var chroms = query.Chromosomes.Where(l => l.StartsWith(options.PreferPrefix)).ToArray();
foreach (var chrom in chroms)
{
chromosomes[chrom].Queries.Remove(query);
query.Chromosomes.Remove(chrom);
}
}
}
result.RemoveAll(l => l.Queries.Count == 0);
}
return result;
}
public override IEnumerable<string> Process()
{
var result = new List<string>();
var countFiles = options.GetCountFiles();
countFiles.Sort((m1, m2) => m1.Name.CompareTo(m2.Name));
var format = new ChromosomeCountSlimItemXmlFormat(outputSample: true);
var countMap = new Dictionary<string, ChromosomeCountSlimItem>();
int fileIndex = 0;
foreach (var file in countFiles)
{
fileIndex++;
Progress.SetMessage("Reading {0}/{1}: {2} ...", fileIndex, countFiles.Count, file.File);
var curcounts = format.ReadFromFile(file.File);
if (curcounts.Count > 0 && string.IsNullOrEmpty(curcounts[0].Queries[0].Sequence))
{
Console.WriteLine("Didn't read in the sequence of query " + curcounts[0].Queries[0].Qname);
}
curcounts.ForEach(m =>
{
foreach (var q in m.Queries)
{
q.Sample = file.Name;
}
});
foreach (var c in curcounts)
{
var name = c.Names.First();
ChromosomeCountSlimItem item;
if (countMap.TryGetValue(name, out item))
{
item.Queries.AddRange(c.Queries);
}
else
{
countMap[name] = c;
}
}
}
var counts = countMap.Values.ToList();
WriteOutput(options.OutputFile, countFiles, format, counts);
result.Add(options.OutputFile);
if (File.Exists(options.CategoryMapFile))
{
Progress.SetMessage("Reading category map ...");
var categoryMap = new MapItemReader(0, 1).ReadFromFile(options.CategoryMapFile);
var queries = new HashSet<SAMChromosomeItem>(from c in counts
from q in c.Queries
select q);
var dic = new Dictionary<string, ChromosomeCountSlimItem>();
foreach (var q in queries)
{
q.Chromosomes = (from chrom in q.Chromosomes
select categoryMap[chrom].Value).Distinct().OrderBy(m => m).ToList();
foreach (var chrom in q.Chromosomes)
{
ChromosomeCountSlimItem item;
if (!dic.TryGetValue(chrom, out item))
{
item = new ChromosomeCountSlimItem();
item.Names.Add(chrom);
dic[chrom] = item;
}
item.Queries.Add(q);
}
}
var catFile = Path.ChangeExtension(options.OutputFile, ".category" + Path.GetExtension(options.OutputFile));
WriteOutput(catFile, countFiles, format, dic.Values.ToList());
result.Add(catFile);
}
if (options.OutputReadTable || options.OutputReadContigTable)
{
Progress.SetMessage("Building sequence map...");
var reads = SmallRNASequenceUtils.ConvertFrom(counts);
if (options.OutputReadTable)
{
Progress.SetMessage("Saving read file...");
var readOutput = Path.ChangeExtension(options.OutputFile, ".read" + Path.GetExtension(options.OutputFile));
new SmallRNASequenceFormat(int.MaxValue, false).WriteToFile(readOutput, reads);
result.Add(readOutput);
}
if (options.OutputReadContigTable)
{
Progress.SetMessage("Building sequence contig by similarity ...");
var contigs = SmallRNASequenceUtils.BuildContigByIdenticalSimilarity(reads, options.MinimumOverlapRate, options.MaximumExtensionBase, progress: Progress);
Progress.SetMessage("Contig number = {0}", contigs.Count);
Progress.SetMessage("Saving contig file...");
var contigOutput = Path.ChangeExtension(options.OutputFile, ".contig" + Path.GetExtension(options.OutputFile));
new SmallRNASequenceContigFormat().WriteToFile(contigOutput, contigs);
result.Add(contigOutput);
Progress.SetMessage("Saving sequence contig details...");
new SmallRNASequenceContigDetailFormat().WriteToFile(contigOutput + ".details", contigs);
result.Add(contigOutput + ".details");
}
}
Progress.End();
return result;
}
public List <ChromosomeCountSlimItem> Build(string fileName)
{
if (File.Exists(options.CategoryMapFile))
{
Progress.SetMessage("Reading name map file " + options.CategoryMapFile + " ...");
nameMap = new MapItemReader(0, 1).ReadFromFile(options.CategoryMapFile).ToDictionary(m => m.Key, m => m.Value.Value);
}
var result = new List <ChromosomeCountSlimItem>();
var queries = new Dictionary <string, SAMChromosomeItem>();
var chromosomes = new Dictionary <string, ChromosomeCountSlimItem>();
Regex chromosomeRegex = null;
Func <string, bool> acceptChromosome;
if (string.IsNullOrEmpty(options.ChromosomePattern))
{
acceptChromosome = m => true;
}
else
{
chromosomeRegex = new Regex(options.ChromosomePattern);
acceptChromosome = m => chromosomeRegex.Match(m).Success;
}
Progress.SetMessage("Parsing alignment file " + fileName + " ...");
using (var sr = SAMFactory.GetReader(fileName, true))
{
int count = 0;
int waitingcount = 0;
string line;
while ((line = sr.ReadLine()) != null)
{
if (count % 1000 == 0)
{
if (Progress.IsCancellationPending())
{
throw new UserTerminatedException();
}
}
if (count % 100000 == 0 && count > 0)
{
Progress.SetMessage("{0} candidates from {1} reads", waitingcount, count);
}
count++;
var parts = line.Split('\t');
SAMFlags flag = (SAMFlags)int.Parse(parts[SAMFormatConst.FLAG_INDEX]);
//unmatched
if (flag.HasFlag(SAMFlags.UnmappedQuery))
{
continue;
}
var seqname = GetName(parts[SAMFormatConst.RNAME_INDEX]);
if (!acceptChromosome(seqname))
{
continue;
}
var qname = parts[SAMFormatConst.QNAME_INDEX];
SAMChromosomeItem query;
if (!queries.TryGetValue(qname, out query))
{
query = new SAMChromosomeItem();
query.Qname = qname;
queries[qname] = query;
if (options.KeepSequence)
{
query.Sequence = parts[SAMFormatConst.SEQ_INDEX];
if (flag.HasFlag(SAMFlags.QueryOnReverseStrand))
{
query.Sequence = SequenceUtils.GetReverseComplementedSequence(query.Sequence);
}
}
}
query.Chromosomes.Add(seqname);
ChromosomeCountSlimItem item;
if (!chromosomes.TryGetValue(seqname, out item))
{
item = new ChromosomeCountSlimItem();
item.Names.Add(seqname);
chromosomes[seqname] = item;
result.Add(item);
}
item.Queries.Add(query);
waitingcount++;
}
Progress.SetMessage("Finally, there are {0} candidates from {1} reads", waitingcount, count);
}
foreach (var query in queries.Values)
{
query.Chromosomes = query.Chromosomes.Distinct().OrderBy(m => m).ToList();
}
foreach (var sam in chromosomes.Values)
{
sam.Queries = sam.Queries.Distinct().OrderBy(m => m.Qname).ToList();
}
if (!string.IsNullOrEmpty(options.PreferPrefix))
{
foreach (var query in queries.Values)
{
if (query.Chromosomes.Any(l => l.StartsWith(options.PreferPrefix)))
{
var chroms = query.Chromosomes.Where(l => l.StartsWith(options.PreferPrefix)).ToArray();
foreach (var chrom in chroms)
{
chromosomes[chrom].Queries.Remove(query);
query.Chromosomes.Remove(chrom);
}
}
}
result.RemoveAll(l => l.Queries.Count == 0);
}
return(result);
}
public override IEnumerable <string> Process()
{
var result = new List <string>();
var countFiles = options.GetCountFiles();
countFiles.Sort((m1, m2) => m1.Name.CompareTo(m2.Name));
var format = new ChromosomeCountSlimItemXmlFormat(outputSample: true);
var countMap = new Dictionary <string, ChromosomeCountSlimItem>();
int fileIndex = 0;
foreach (var file in countFiles)
{
fileIndex++;
Progress.SetMessage("Reading {0}/{1}: {2} ...", fileIndex, countFiles.Count, file.File);
var curcounts = format.ReadFromFile(file.File);
if (curcounts.Count > 0 && string.IsNullOrEmpty(curcounts[0].Queries[0].Sequence))
{
Console.WriteLine("Didn't read in the sequence of query " + curcounts[0].Queries[0].Qname);
}
curcounts.ForEach(m =>
{
foreach (var q in m.Queries)
{
q.Sample = file.Name;
}
});
foreach (var c in curcounts)
{
var name = c.Names.First();
ChromosomeCountSlimItem item;
if (countMap.TryGetValue(name, out item))
{
item.Queries.AddRange(c.Queries);
}
else
{
countMap[name] = c;
}
}
}
var counts = countMap.Values.ToList();
WriteOutput(options.OutputFile, countFiles, format, counts);
result.Add(options.OutputFile);
if (File.Exists(options.CategoryMapFile))
{
Progress.SetMessage("Reading category map ...");
var categoryMap = new MapItemReader(0, 1).ReadFromFile(options.CategoryMapFile);
var queries = new HashSet <SAMChromosomeItem>(from c in counts
from q in c.Queries
select q);
var dic = new Dictionary <string, ChromosomeCountSlimItem>();
foreach (var q in queries)
{
q.Chromosomes = (from chrom in q.Chromosomes
select categoryMap[chrom].Value).Distinct().OrderBy(m => m).ToList();
foreach (var chrom in q.Chromosomes)
{
ChromosomeCountSlimItem item;
if (!dic.TryGetValue(chrom, out item))
{
item = new ChromosomeCountSlimItem();
item.Names.Add(chrom);
dic[chrom] = item;
}
item.Queries.Add(q);
}
}
var catFile = Path.ChangeExtension(options.OutputFile, ".category" + Path.GetExtension(options.OutputFile));
WriteOutput(catFile, countFiles, format, dic.Values.ToList());
result.Add(catFile);
}
if (options.OutputReadTable || options.OutputReadContigTable)
{
Progress.SetMessage("Building sequence map...");
var reads = SmallRNASequenceUtils.ConvertFrom(counts);
if (options.OutputReadTable)
{
Progress.SetMessage("Saving read file...");
var readOutput = Path.ChangeExtension(options.OutputFile, ".read" + Path.GetExtension(options.OutputFile));
new SmallRNASequenceFormat(int.MaxValue, false).WriteToFile(readOutput, reads);
result.Add(readOutput);
}
if (options.OutputReadContigTable)
{
Progress.SetMessage("Building sequence contig by similarity ...");
var contigs = SmallRNASequenceUtils.BuildContigByIdenticalSimilarity(reads, options.MinimumOverlapRate, options.MaximumExtensionBase, progress: Progress);
Progress.SetMessage("Contig number = {0}", contigs.Count);
Progress.SetMessage("Saving contig file...");
var contigOutput = Path.ChangeExtension(options.OutputFile, ".contig" + Path.GetExtension(options.OutputFile));
new SmallRNASequenceContigFormat().WriteToFile(contigOutput, contigs);
result.Add(contigOutput);
Progress.SetMessage("Saving sequence contig details...");
new SmallRNASequenceContigDetailFormat().WriteToFile(contigOutput + ".details", contigs);
result.Add(contigOutput + ".details");
}
}
Progress.End();
return(result);
}
