private List <SpectrumMatch> TDPortalReader(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 <SpectrumMatch> td_hits = new List <SpectrumMatch>();
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)
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('|');
//for bottom-up, don't read in ambiguous PSMs
if (file.purpose == Purpose.BottomUp && ptms.Length > 1)
{
add_topdown_hit = false;
}
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_ptms)
{
bad_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_ptms)
{
bad_ptms.Add(id + " at " + cellStrings[4][position_after_begin - 1]);
}
add_topdown_hit = false;
}
}
public TopDownHit(Dictionary <char, double> aaIsotopeMassList, InputFile file, TopDownResultType tdResultType, string accession, string pfr, string uniprot_id, string name, string sequence, int begin, int end, List <Ptm> modifications, double reported_mass, double theoretical_mass, int scan, double retention_time, string filename, double pscore, double score)
{
this.pfr_accession = pfr;
this.file = file;
this.tdResultType = tdResultType;
this.accession = accession;
this.uniprot_id = uniprot_id;
this.name = name;
this.sequence = sequence;
this.begin = begin;
this.end = end;
this.ptm_list = modifications;
//if neucode labeled, calculate neucode mass....
this.reported_mass = Sweet.lollipop.neucode_labeled ? Sweet.lollipop.get_neucode_mass(reported_mass, sequence.Count(s => s == 'K')) : reported_mass;
this.theoretical_mass = CalculateProteoformMass(sequence, aaIsotopeMassList) + ptm_list.Where(p => p.modification != null).Sum(p => Math.Round((double)p.modification.MonoisotopicMass, 5));
this.ms2ScanNumber = scan;
this.ms2_retention_time = retention_time;
this.filename = filename;
this.score = score;
this.pscore = pscore;
}
