public static List<CoverageRegion> GetSmallRNACoverageRegion(string mappedFeatureXmlFile, string[] includeSmallRNATags = null, string[] excudeSmallRNATags = null)
{
var smallRNAGroups = new FeatureItemGroupXmlFormat().ReadFromFile(mappedFeatureXmlFile);
if (includeSmallRNATags != null && includeSmallRNATags.Length > 0)
{
smallRNAGroups.ForEach(m => m.RemoveAll(l => includeSmallRNATags.All(k => !m.Name.StartsWith(k))));
smallRNAGroups.RemoveAll(m => m.Count == 0);
}
if (excudeSmallRNATags != null && excudeSmallRNATags.Length > 0)
{
smallRNAGroups.ForEach(m => m.RemoveAll(l => excudeSmallRNATags.Any(k => m.Name.StartsWith(k))));
smallRNAGroups.RemoveAll(m => m.Count == 0);
}
var result = new List<CoverageRegion>();
foreach (var sg in smallRNAGroups)
{
//since the items in same group shared same reads, only the first one will be used.
var smallRNA = sg[0];
smallRNA.Name = (from g in sg select g.Name).Merge("/");
smallRNA.Locations.RemoveAll(m => m.SamLocations.Count == 0);
smallRNA.CombineLocationByMappedReads();
//only first location will be used.
var loc = smallRNA.Locations[0];
//coverage in all position will be set as same as total query count
var rg = new CoverageRegion();
rg.Name = smallRNA.Name;
rg.Seqname = loc.Seqname;
rg.Start = loc.Start;
rg.End = loc.End;
rg.Strand = loc.Strand;
rg.Sequence = loc.Sequence;
var coverage = (from sloc in loc.SamLocations select sloc.SamLocation.Parent.QueryCount).Sum();
for (int i = 0; i < loc.Length; i++)
{
rg.Coverages.Add(coverage);
}
result.Add(rg);
}
return result;
}
public List<FeatureItemGroup> Build(string countXmlFile)
{
var result = new FeatureItemGroupXmlFormat().ReadFromFile(countXmlFile);
Progress.SetMessage("There are {0} groups in {1}", result.Count, countXmlFile);
result.ForEach(g => g.ForEach(smallRNA => smallRNA.Locations.ForEach(region => region.QueryCountBeforeFilter = region.QueryCount)));
//no number of no penalty mutation defined, check the T2C
if (result.All(m => m.All(l => l.Locations.All(k => k.SamLocations.All(s => s.NumberOfNoPenaltyMutation == 0)))))
{
foreach (var group in result)
{
foreach (var smallRNA in group)
{
smallRNA.Locations.RemoveAll(m => m.SamLocations.Count == 0);
foreach (var region in smallRNA.Locations)
{
region.SamLocations.ForEach(q =>
{
var snp = q.SamLocation.GetNotGsnapMismatch(q.SamLocation.Parent.Sequence);
if (null != snp && snp.IsMutation('T', 'C'))
{
q.NumberOfMismatch = q.SamLocation.NumberOfMismatch - 1;
q.NumberOfNoPenaltyMutation = 1;
}
else
{
q.NumberOfMismatch = q.SamLocation.NumberOfMismatch;
q.NumberOfNoPenaltyMutation = 0;
}
});
}
}
}
}
result.RemoveAll(m =>
{
m.RemoveAll(l =>
{
l.Locations.RemoveAll(k =>
{
k.SamLocations.RemoveAll(s => s.NumberOfNoPenaltyMutation == 0);
return k.SamLocations.Count == 0;
});
return l.Locations.Count == 0;
});
return m.Count == 0;
});
Progress.SetMessage("There are {0} groups having T2C mutation", result.Count);
foreach (var group in result)
{
foreach (var smallRNA in group)
{
foreach (var region in smallRNA.Locations)
{
region.PValue = CalculateT2CPvalue(region.QueryCountBeforeFilter, region.QueryCount, this.t2cRate);
}
}
}
return result;
}