// The calculations for unlabeled and neucode components are the same, currently
public static List <Component> generate_neucode_components(double starter_mass)
{
List <Component> components = new List <Component>();
InputFile inFile = new ProteoformSuiteInternal.InputFile("somepath", Labeling.NeuCode, Purpose.Identification);
for (int i = 0; i < 2; i++)
{
Component light = new Component();
Component heavy = new Component();
light.input_file = inFile;
heavy.input_file = inFile;
light.id = 1.ToString();
heavy.id = 2.ToString();
light.weighted_monoisotopic_mass = starter_mass;
heavy.weighted_monoisotopic_mass = starter_mass + starter_lysine_count * Lollipop.NEUCODE_LYSINE_MASS_SHIFT;
light.intensity_sum_olcs = starter_intensity; //using the special intensity sum for overlapping charge states in a neucode pair
heavy.intensity_sum_olcs = starter_intensity / 2; //using the special intensity sum for overlapping charge states in a neucode pair
light.rt_apex = starter_rt;
heavy.rt_apex = starter_rt;
light.accepted = true;
heavy.accepted = true;
ChargeState light_charge_state = new ChargeState(1, light.intensity_sum_olcs, light.weighted_monoisotopic_mass, 1.00727645D);
ChargeState heavy_charge_state = new ChargeState(1, heavy.intensity_sum_olcs, heavy.weighted_monoisotopic_mass, 1.00727645D);
light.charge_states = new List <ChargeState> {
light_charge_state
};
heavy.charge_states = new List <ChargeState> {
heavy_charge_state
};
NeuCodePair n = new NeuCodePair(light, heavy);
n.lysine_count = starter_lysine_count;
n.calculate_properties();
components.Add(n);
}
return(components);
}
public bool Run_TdMzCal(InputFile raw_file, List <TopDownHit> topdown_hits)
{
all_topdown_hits = topdown_hits.Where(h => h.score > 0).ToList();
//need to reset m/z in case same td hits used for multiple calibration raw files...
Parallel.ForEach(all_topdown_hits, h => h.mz = h.reported_mass.ToMz(h.charge));
high_scoring_topdown_hits = all_topdown_hits.Where(h => h.score >= 40).ToList();
this.raw_file = raw_file;
if (high_scoring_topdown_hits.Count < 5)
{
return(false);
}
myMsDataFile = Path.GetExtension(raw_file.complete_path) == ".raw" ?
ThermoStaticData.LoadAllStaticData(raw_file.complete_path) :
null;
if (myMsDataFile == null)
{
myMsDataFile = Mzml.LoadAllStaticData(raw_file.complete_path);
}
if (myMsDataFile == null)
{
return(false);
}
DataPointAquisitionResults dataPointAcquisitionResult = GetDataPoints();
if (dataPointAcquisitionResult.Ms1List.Count < 10)
{
return(false);
}
if (Sweet.lollipop.mass_calibration)
{
var myMs1DataPoints = new List <(double[] xValues, double yValue)>();
for (int i = 0; i < dataPointAcquisitionResult.Ms1List.Count; i++)
{
//x values
var explanatoryVariables = new double[4];
explanatoryVariables[0] = dataPointAcquisitionResult.Ms1List[i].mz;
explanatoryVariables[1] = dataPointAcquisitionResult.Ms1List[i].retentionTime;
explanatoryVariables[2] = dataPointAcquisitionResult.Ms1List[i].logTotalIonCurrent;
explanatoryVariables[3] = dataPointAcquisitionResult.Ms1List[i].logInjectionTime;
//yvalue
double mzError = dataPointAcquisitionResult.Ms1List[i].massError;
myMs1DataPoints.Add((explanatoryVariables, mzError));
}
var ms1Model = GetRandomForestModel(myMs1DataPoints);
CalibrateHitsAndComponents(ms1Model);
if (Sweet.lollipop.calibrate_raw_files)
{
MzmlMethods.CreateAndWriteMyMzmlWithCalibratedSpectra(myMsDataFile,
raw_file.directory + "\\" + raw_file.filename + "_calibrated.mzML", false);
}
}
if (Sweet.lollipop.retention_time_calibration)
{
var myMs1DataPoints = new List <(double[] xValues, double yValue)>();
List <TopDownHit> firstElutingTopDownHit = new List <TopDownHit>();
List <string> PFRs = high_scoring_topdown_hits.Select(h => h.pfr_accession).Distinct().ToList();
foreach (var PFR in PFRs)
{
var firstHitWithPFR = high_scoring_topdown_hits
.Where(h => h.pfr_accession == PFR).OrderBy(h => h.ms2_retention_time).First();
firstElutingTopDownHit.Add(firstHitWithPFR);
}
for (int i = 0; i < dataPointAcquisitionResult.Ms1List.Count; i++)
{
if (firstElutingTopDownHit.Contains(dataPointAcquisitionResult.Ms1List[i].identification))
{
//x values
var explanatoryVariables = new double[1];
explanatoryVariables[0] = dataPointAcquisitionResult.Ms1List[i].retentionTime;
//yvalue
double RTError = dataPointAcquisitionResult.Ms1List[i].RTError;
myMs1DataPoints.Add((explanatoryVariables, RTError));
}
}
if (myMs1DataPoints.Count < 10)
{
return(false);
}
var ms1Model = GetRandomForestModel(myMs1DataPoints);
foreach (Component c in Sweet.lollipop.calibration_components.Where(h => h.input_file.lt_condition == raw_file.lt_condition && h.input_file.biological_replicate == raw_file.biological_replicate && h.input_file.fraction == raw_file.fraction && h.input_file.technical_replicate == raw_file.technical_replicate))
{
c.rt_apex = c.rt_apex - ms1Model.Predict(new double[] { c.rt_apex });
}
}
return(true);
}
public List <string> bad_topdown_ptms = new List <string>(); //PTMs not in theoretical database added to warning file.
//Reading in Top-down excel
public List <TopDownHit> ReadTDFile(InputFile file)
{
//if neucode labeled, calculate neucode light theoretical AND observed mass! --> better for matching up
//if carbamidomethylated, add 57 to theoretical mass (already in observed mass...)
aaIsotopeMassList = new AminoAcidMasses(Sweet.lollipop.carbamidomethylation, Sweet.lollipop.neucode_labeled).AA_Masses;
List <TopDownHit> td_hits = new List <TopDownHit>();
List <List <string> > cells = ExcelReader.get_cell_strings(file, true);//This returns the entire sheet except for the header. Each row of cells is one List<string>
//get ptms on proteoform -- check for mods. IF not in database, make new topdown mod, show Warning message.
Parallel.ForEach(cells, cellStrings =>
{
bool add_topdown_hit = true; //if PTM or accession not found, will not add (show warning)
if (cellStrings.Count == 24)
{
TopDownResultType tdResultType = (cellStrings[15] == "BioMarker") ? TopDownResultType.Biomarker : ((cellStrings[15] == "Tight Absolute Mass") ? TopDownResultType.TightAbsoluteMass : TopDownResultType.Unknown);
if (tdResultType != TopDownResultType.Unknown) //uknown result type!
{
List <Ptm> ptm_list = new List <Ptm>(); // if nothing gets added, an empty ptmlist is passed to the topdownhit constructor.
//N-term modifications
if (cellStrings[10].Length > 0) //N Terminal Modification Code
{
string[] ptms = cellStrings[10].Split('|');
foreach (string ptm in ptms)
{
int position = Int32.TryParse(cellStrings[5], out int i) ? i : 0;
if (position == 0)
{
add_topdown_hit = false;
continue;
}
if (cellStrings[10].Split(':')[1] == "1458")//PSI-MOD 1458 is supposed to be N-terminal acetylation
{
ptm_list.Add(new Ptm(position, Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.OriginalId == "N-terminal Acetyl").FirstOrDefault()));
}
else
{
string psimod = ptm.Split(':')[1].Split('@')[0];//The number after the @ is the position in the protein
while (psimod.Length < 5)
{
psimod = "0" + psimod; //short part should be the accession number, which is an integer
}
Modification mod = Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.DatabaseReference != null && m.DatabaseReference.ContainsKey("PSI-MOD") && m.DatabaseReference["PSI-MOD"].Contains(psimod)).FirstOrDefault();
if (mod == null)
{
psimod = "MOD:" + psimod;
mod = Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.DatabaseReference != null && m.DatabaseReference.ContainsKey("PSI-MOD") && m.DatabaseReference["PSI-MOD"].Contains(psimod)).FirstOrDefault();
}
if (mod != null)
{
ptm_list.Add(new Ptm(position, mod));
}
else
{
lock (bad_topdown_ptms)
{
bad_topdown_ptms.Add("PSI-MOD:" + psimod + " at " + position);
}
add_topdown_hit = false;
}
}
}
}
//don't have example of c-term modification to write code
//other mods
if (cellStrings[9].Length > 0)//Modification Codes
{
string[] ptms = cellStrings[9].Split('|');
foreach (string ptm in ptms)
{
Modification mod = null;
string id = "";
if (ptm.Split(':').Length < 2)
{
add_topdown_hit = false;
continue;
}
if (ptm.Split(':')[1].Split('@').Length < 2)
{
add_topdown_hit = false;
continue;
}
int position_after_begin = (Int32.TryParse(ptm.Split(':')[1].Split('@')[1], out int j) ? j : -1) + 1; //one based sequence
//they give position # as from begin site -> want to report in terms of overall sequence #'s
//begin + position from begin - 1 => position in overall sequence
if (position_after_begin == 0)
{
add_topdown_hit = false;
continue;
}
int begin = Int32.TryParse(cellStrings[5], out int k) ? k : 0;
if (begin == 0)
{
add_topdown_hit = false;
continue;
}
int position = begin + position_after_begin - 1;
if (ptm.Split(':')[0] == "RESID")
{
string resid = ptm.Split(':')[1].Split('@')[0];//The number after the @ is the position in the protein
while (resid.Length < 4)
{
resid = "0" + resid; //short part should be the accession number, which is an integer
}
resid = "AA" + resid;
id = "RESID:" + resid;
mod = Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.DatabaseReference != null && m.DatabaseReference.ContainsKey("RESID") && m.DatabaseReference["RESID"].Contains(resid)).FirstOrDefault();
}
else if (ptm.Split(':')[0] == "PSI-MOD")
{
string psimod = ptm.Split(':')[1].Split('@')[0];//The number after the @ is the position in the protein
while (psimod.Length < 5)
{
psimod = "0" + psimod; //short part should be the accession number, which is an integer
}
mod = Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.DatabaseReference != null && m.DatabaseReference.ContainsKey("PSI-MOD") && m.DatabaseReference["PSI-MOD"].Contains(psimod)).FirstOrDefault();
if (mod == null)
{
psimod = "MOD:" + psimod;
mod = Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.DatabaseReference != null && m.DatabaseReference.ContainsKey("PSI-MOD") && m.DatabaseReference["PSI-MOD"].Contains(psimod)).FirstOrDefault();
}
id = "PSI-MOD:" + psimod;
}
if (mod != null)
{
ptm_list.Add(new Ptm(position, mod));
}
else
{
lock (bad_topdown_ptms)
{
bad_topdown_ptms.Add(id + " at " + cellStrings[4][position_after_begin - 1]);
}
add_topdown_hit = false;
}
}
//Reading in metamopheus excel
public List <TopDownHit> ReadMetamopheusFile(InputFile file)
{
//if neucode labeled, calculate neucode light theoretical AND observed mass! --> better for matching up
//if carbamidomethylated, add 57 to theoretical mass (already in observed mass...)
aaIsotopeMassList = new AminoAcidMasses(Sweet.lollipop.carbamidomethylation, Sweet.lollipop.neucode_labeled).AA_Masses;
List <TopDownHit> td_hits = new List <TopDownHit>();//for one line in excel file
//creates dictionary to find mods
Dictionary <string, Modification> mods = Sweet.lollipop.theoretical_database.all_mods_with_mass.ToDictionary(kv => kv.IdWithMotif, kv => kv);
List <List <string> > cells = ExcelReader.get_cell_strings(file, true);//This returns the entire sheet except for the header. Each row of cells is one List<string>
//get ptms on proteoform -- check for mods. IF not in database, make new topdown mod, show Warning message.
Parallel.ForEach(cells, cellStrings =>
{
bool add_topdown_hit = true; //if PTM or accession not found, will not add (show warning)
if (cellStrings.Count == 55)
{
List <Ptm> new_ptm_list = new List <Ptm>();
//if bad mod itll catch it to add to bad_topdown_ptms
try
{
PeptideWithSetModifications modsIdentifier = new PeptideWithSetModifications(cellStrings[14].Split('|')[0], mods);
var ptm_list = modsIdentifier.AllModsOneIsNterminus;
//for each entry in ptm_list make a new Ptm and add it to the new_ptm_list
foreach (KeyValuePair <int, Proteomics.Modification> entry in ptm_list)
{
Modification mod = Sweet.lollipop.theoretical_database.uniprotModifications.Values.SelectMany(m => m).Where(m => m.IdWithMotif == entry.Value.IdWithMotif).FirstOrDefault();
var Ptm = new Ptm();
if (mod != null)
{
new_ptm_list.Add(new Ptm(entry.Key, entry.Value));
}
else
{
lock (bad_topdown_ptms)
{
//error is somewahre in sequece
bad_topdown_ptms.Add("Mod Name:" + entry.Value.IdWithMotif + " at " + entry.Key);
add_topdown_hit = false;
}
}
}
}
catch (MzLibUtil.MzLibException)
{
lock (bad_topdown_ptms)
{
//error is somewahre in sequece
bad_topdown_ptms.Add("Bad mod at " + cellStrings[0] + " scan " + cellStrings[1]);
add_topdown_hit = false;
}
}
//This is the excel file header:
//cellStrings[0]=File Name
//cellStrings[1]=Scan Number
//cellStrings[2]=Scan Retention Time
//cellStrings[3]=Num Experimental Peaks
//cellStrings[4]=Total Ion Current
//cellStrings[5]=Precursor Scan Number
//cellStrings[6]=Precursor Charge
//cellStrings[7]=Precursor MZ
//cellStrings[8]=Precursor Mass
//cellStrings[9]=Score
//cellStrings[10]=Delta Score
//cellStrings[11]=Notch
//cellStrings[12]=Different Peak Matches
//cellStrings[13]=Base Sequence
//cellStrings[14]=Full Sequence
//cellStrings[15]=Essential Sequence
//cellStrings[16]=PSM Count
//cellStrings[17]=Mods
//cellStrings[18]=Mods Chemical Formulas
//cellStrings[19]=Mods Combined Chemical Formula
//cellStrings[20]=Num Variable Mods
//cellStrings[21]=Missed Cleavages
//cellStrings[22]=Peptide Monoisotopic Mass
//cellStrings[23]=Mass Diff (Da)
//cellStrings[24]=Mass Diff (ppm)
//cellStrings[25]=Protein Accession
//cellStrings[26]=Protein Name
//cellStrings[27]=Gene Name
//cellStrings[28]=Organism Name
//cellStrings[29]=Intersecting Sequence Variations
//cellStrings[30]=Identified Sequence Variations
//cellStrings[31]=Splice Sites
//cellStrings[32]=Contaminant
//cellStrings[33]=Decoy
//cellStrings[34]=Peptide Description
//cellStrings[35]=Start and End Residues In Protein
//cellStrings[36]=Previous Amino Acid
//cellStrings[37]=Next Amino Acid
//cellStrings[38]=All Scores
//cellStrings[39]=Theoreticals Searched
//cellStrings[40]=Decoy/Contaminant/Target
//cellStrings[41]=Matched Ion Series
//cellStrings[42]=Matched Ion Mass-To-Charge Ratios
//cellStrings[43]=Matched Ion Mass Diff (Da)
//cellStrings[44]=Matched Ion Mass Diff (Ppm)
//cellStrings[45]=Matched Ion Intensities
//cellStrings[46]=Matched Ion Counts
//cellStrings[47]=Localized Scores
//cellStrings[48]=Improvement Possible
//cellStrings[49]=Cumulative Target
//cellStrings[50]=Cumulative Decoy
//cellStrings[51]=QValue
//cellStrings[52]=Cumulative Target Notch
//cellStrings[53]=Cumulative Decoy Notch
//cellStrings[54]=QValue Notch
//cellStrings[55]=eValue
//cellStrings[56]=eScore
if (cellStrings[35].Length > 0)
{
string[] ids = cellStrings[35].Split('|');
//splits the string to get the value of starting index
string[] index = ids[0].Split(' ');
string[] startIndexValue = index[0].Split('[');
string startResidues = startIndexValue[1];
//splits string to get value of ending index
string[] endIndexValue = index[2].Split(']');
string endResidues = endIndexValue[0];
if (add_topdown_hit)
{
//if bad mod u want td hit to be false
TopDownHit td_hit = new TopDownHit(aaIsotopeMassList, file, TopDownResultType.TightAbsoluteMass, cellStrings[25], cellStrings[14], cellStrings[25], cellStrings[26], cellStrings[13],
Int32.TryParse(startResidues, out int j) ? j : 0, Int32.TryParse(endResidues, out int i) ? i : 0, new_ptm_list, Double.TryParse(cellStrings[8], out double d) ? d : 0, Double.TryParse(cellStrings[22], out d) ? d : 0,
Int32.TryParse(cellStrings[1], out i) ? i : 0, Double.TryParse(cellStrings[2], out d) ? d : 0, cellStrings[0].Split('.')[0], Double.TryParse(cellStrings[8], out d) ? d : 0, Sweet.lollipop.min_score_td + 1);
if (td_hit.begin > 0 && td_hit.end > 0 && td_hit.theoretical_mass > 0 && td_hit.pscore > 0 && td_hit.reported_mass > 0 && td_hit.score > 0 &&
td_hit.ms2ScanNumber > 0 && td_hit.ms2_retention_time > 0)
{
lock (td_hits) td_hits.Add(td_hit);
}
}
}
}
});
return(td_hits);
}