public TopDownProteoform(string accession, List <TopDownHit> hits) : base(accession, null, true)
{
TopDownHit root = hits[0];
this.name = root.name;
this.pfr_accession = root.pfr_accession;
this.topdown_ptm_set = new PtmSet(root.ptm_list);
this.uniprot_id = root.uniprot_id;
this.sequence = root.sequence;
this.topdown_begin = root.begin;
this.theoretical_mass = root.theoretical_mass;
this.topdown_end = root.end;
this.topdown_hits = hits;
this.calculate_td_properties();
this.accession = accession + "_" + topdown_begin + "to" + topdown_end + "_1TD";
this.lysine_count = sequence.Count(s => s == 'K');
this.topdown_id = true;
}
private IEnumerable <LabeledMs1DataPoint> SearchMS1Spectra(double[] originalMasses, double[] originalIntensities, int ms2spectrumIndex, int direction, List <int> scansAdded, HashSet <Tuple <double, int> > peaksAddedHashSet, int proteinCharge, TopDownHit identification)
{
int theIndex = direction == 1 ? ms2spectrumIndex : ms2spectrumIndex - 1;
bool addedAscan = true;
int highestKnownChargeForThisPeptide = proteinCharge;
while (theIndex >= 1 && theIndex <= myMsDataFile.NumSpectra && addedAscan)
{
if (scansAdded.Contains(theIndex))
{
theIndex += direction;
continue;
}
scansAdded.Add(theIndex);
int countForThisScan = 0;
if (myMsDataFile.GetOneBasedScan(theIndex).MsnOrder > 1)
{
theIndex += direction;
continue;
}
addedAscan = false;
var fullMS1scan = myMsDataFile.GetOneBasedScan(theIndex);
int ms1ScanNumber = fullMS1scan.OneBasedScanNumber;
var scanWindowRange = fullMS1scan.ScanWindowRange;
var fullMS1spectrum = fullMS1scan.MassSpectrum;
if (fullMS1spectrum.Size == 0)
{
break;
}
//look in both charge state directions for proteins -- likely to have multiple charge states.
for (int i = -1; i <= 1; i += 2)
{
bool startingToAddCharges = false;
bool continueAddingCharges = false;
int chargeToLookAt = i == 1 ? proteinCharge : (proteinCharge - 1);
do
{
//if m/z too big or small and going in correct direction, increase or decrease charge state. otherwise break.
if (originalMasses[0].ToMz(chargeToLookAt) > scanWindowRange.Maximum)
{
if (i == 1)
{
chargeToLookAt++;
continue;
}
else
{
break;
}
}
if (originalMasses[0].ToMz(chargeToLookAt) < scanWindowRange.Minimum)
{
if (i == -1)
{
chargeToLookAt--;
continue;
}
else
{
break;
}
}
var trainingPointsToAverage = new List <LabeledMs1DataPoint>();
foreach (double a in originalMasses)
{
double theMZ = a.ToMz(chargeToLookAt);
if (Sweet.lollipop.neucode_labeled)
{
theMZ = Sweet.lollipop.get_neucode_mass(theMZ.ToMass(chargeToLookAt), identification.sequence.Count(s => s == 'K')).ToMz(chargeToLookAt);
}
//10 ppm
double mass_tolerance = theMZ / 1e6 * 10;
var npwr = fullMS1spectrum.NumPeaksWithinRange(theMZ - mass_tolerance, theMZ + mass_tolerance);
if (npwr == 0)
{
break;
}
if (npwr > 1)
{
continue;
}
var closestPeak = fullMS1spectrum.GetClosestPeakXvalue(theMZ);
var closestPeakMZ = closestPeak;
var theTuple = closestPeakMZ != null?Tuple.Create((double)closestPeakMZ, ms1ScanNumber) : null;
if (theTuple == null)
{
continue;
}
if (!peaksAddedHashSet.Contains(theTuple))
{
peaksAddedHashSet.Add(theTuple);
highestKnownChargeForThisPeptide = Math.Max(highestKnownChargeForThisPeptide, chargeToLookAt);
trainingPointsToAverage.Add(new LabeledMs1DataPoint((double)closestPeakMZ, double.NaN, double.NaN, double.NaN, (double)closestPeakMZ - theMZ, null));
}
else
{
break;
}
}
// If started adding and suddnely stopped, go to next one, no need to look at higher charges
if (trainingPointsToAverage.Count == 0 && startingToAddCharges == true)
{
break;
}
if ((trainingPointsToAverage.Count == 0 || (trainingPointsToAverage.Count == 1 && originalIntensities[0] < 0.65)) && (proteinCharge <= chargeToLookAt))
{
break;
}
if (trainingPointsToAverage.Count < Math.Min(5, originalIntensities.Count()))
{
}
else
{
continueAddingCharges = true;
addedAscan = true;
startingToAddCharges = true;
countForThisScan++;
yield return(new LabeledMs1DataPoint(trainingPointsToAverage.Select(b => b.mz).Average(),
fullMS1scan.RetentionTime,
Math.Log(fullMS1scan.TotalIonCurrent),
fullMS1scan.InjectionTime.HasValue ? Math.Log(fullMS1scan.InjectionTime.Value) : double.NaN,
trainingPointsToAverage.Select(b => b.massError).Median(),
identification));
}
chargeToLookAt += i;
} while (continueAddingCharges);
}
theIndex += direction;
}
}
