public void add_topdown_theoreticals(ProteinWithGoTerms prot, string seq, string accession, double unmodified_mass, int decoy_number, int lysine_count, List <TheoreticalProteoform> new_theoreticals, int ptm_set_counter, Dictionary <double, int> mod_ranks, int added_ptm_penalty)
{
foreach (TopDownProteoform topdown in Sweet.lollipop.topdown_proteoforms.Where(p => prot.AccessionList.Select(a => a.Split('_')[0]).Contains(p.accession.Split('_')[0].Split('-')[0]) &&
p.sequence == seq).OrderBy(t => t.accession).ThenByDescending(t => t.sequence.Length)) //order by gene name then descending sequence length --> order matters for creating theoreticals.
{
if (!new_theoreticals.Any(t => t.ptm_set.same_ptmset(topdown.topdown_ptm_set, true)))
{
//match each td proteoform group to the closest theoretical w/ best explanation.... otherwise make new theoretical proteoform
PtmSet ptm_set = new PtmSet(topdown.topdown_ptm_set.ptm_combination, mod_ranks, added_ptm_penalty);
TheoreticalProteoform t =
new TheoreticalProteoform(
accession + "_P" + ptm_set_counter.ToString(),
prot.FullDescription + "_P" + ptm_set_counter.ToString() + (decoy_number < 0 ? "" : "_DECOY_" + decoy_number.ToString()),
seq,
(prot as ProteinSequenceGroup != null ? (prot as ProteinSequenceGroup).proteinWithGoTermList.ToArray() : new ProteinWithGoTerms[] { prot }),
unmodified_mass,
lysine_count,
ptm_set,
decoy_number < 0,
false,
theoretical_proteins);
t.topdown_theoretical = true;
new_theoreticals.Add(t);
ptm_set_counter++;
}
}
}
//if protein sequence doesn't exist, need to add...
public void add_topdown_sequences()
{
List <ProteinWithGoTerms> new_proteins = new List <ProteinWithGoTerms>();
foreach (TopDownProteoform topdown in Sweet.lollipop.topdown_proteoforms.OrderBy(t => t.modified_mass))
{
List <ProteinWithGoTerms> candidate_theoreticals = expanded_proteins.Where(p => p.AccessionList.Select(a => a.Split('_')[0].Split('-')[0]).Contains(topdown.accession.Split('_')[0].Split('-')[0])).ToList();
if (candidate_theoreticals.Count > 0)
{
topdown.gene_name = new GeneName(candidate_theoreticals.SelectMany(t => t.GeneNames));
topdown.geneID = string.Join("; ", candidate_theoreticals.SelectMany(p => p.DatabaseReferences.Where(r => r.Type == "GeneID").Select(r => r.Id)).Distinct());
if (!candidate_theoreticals.Any(p => p.BaseSequence == topdown.sequence) && !new_proteins.Any(p => p.AccessionList.Select(a => a.Split('_')[0]).Contains(topdown.accession.Split('_')[0].Split('-')[0]) && p.BaseSequence == topdown.sequence))
{
int old_proteins_with_same_begin_end_diff_sequence = candidate_theoreticals.Count(t => t.ProteolysisProducts.First().OneBasedBeginPosition == topdown.topdown_begin && t.ProteolysisProducts.First().OneBasedEndPosition == topdown.topdown_end && t.BaseSequence != topdown.sequence);
int new_proteins_with_same_being_end_diff_sequence = new_proteins.Count(t => t.AccessionList.Select(a => a.Split('_')[0].Split('-')[0]).Contains(topdown.accession.Split('_')[0]) && t.ProteolysisProducts.First().OneBasedBeginPosition == topdown.topdown_begin && t.ProteolysisProducts.First().OneBasedEndPosition == topdown.topdown_end && t.BaseSequence != topdown.sequence);
int count = old_proteins_with_same_begin_end_diff_sequence + new_proteins_with_same_being_end_diff_sequence;
ProteinWithGoTerms p = new ProteinWithGoTerms(topdown.sequence, topdown.accession.Split('_')[0].Split('-')[0] + "_" + topdown.topdown_begin + "frag" + topdown.topdown_end + (count > 0 ? "_" + count : ""), candidate_theoreticals.First().GeneNames.ToList(), candidate_theoreticals.First().OneBasedPossibleLocalizedModifications, new List <ProteolysisProduct>()
{
new ProteolysisProduct(topdown.topdown_begin, topdown.topdown_end, "full")
}, candidate_theoreticals.First().Name, candidate_theoreticals.First().FullName, false, false, candidate_theoreticals.First().DatabaseReferences, candidate_theoreticals.First().GoTerms);
p.topdown_protein = true;
new_proteins.Add(p);
}
}
else
{
topdown.accepted = false;
}
}
expanded_proteins = expanded_proteins.Concat(new_proteins).ToArray();
}
//if protein sequence doesn't exist, need to add...
public void add_topdown_sequences()
{
List <ProteinWithGoTerms> new_proteins = new List <ProteinWithGoTerms>();
foreach (TopDownProteoform topdown in Sweet.lollipop.topdown_proteoforms.OrderBy(t => t.modified_mass))
{
List <ProteinWithGoTerms> candidate_theoreticals = expanded_proteins.Where(p => p.AccessionList.Select(a => a.Split('_')[0].Split('-')[0]).Contains(topdown.accession.Split('_')[0].Split('-')[0])).ToList();
bool accessions_in_database = true;
foreach (var hit in topdown.ambiguous_topdown_hits)
{
var ambiguous_hit_theoretical = expanded_proteins.Where(p => p.AccessionList.Select(a => a.Split('_')[0].Split('-')[0]).Contains(hit.accession.Split('_')[0].Split('-')[0])).ToList();
if (ambiguous_hit_theoretical.Count > 0)
{
hit.gene_name = new GeneName(ambiguous_hit_theoretical.SelectMany(t => t.GeneNames));
}
else
{
accessions_in_database = false;
}
}
if (candidate_theoreticals.Count > 0 && accessions_in_database)
{
topdown.topdown_geneName = new GeneName(candidate_theoreticals.SelectMany(t => t.GeneNames));
if (!candidate_theoreticals.Any(p => p.BaseSequence == topdown.sequence) && !new_proteins.Any(p => p.AccessionList.Select(a => a.Split('_')[0]).Contains(topdown.accession.Split('_')[0].Split('-')[0]) && p.BaseSequence == topdown.sequence))
{
int old_proteins_with_same_begin_end_diff_sequence = candidate_theoreticals.Count(t => t.ProteolysisProducts.First().OneBasedBeginPosition == topdown.topdown_begin && t.ProteolysisProducts.First().OneBasedEndPosition == topdown.topdown_end && t.BaseSequence != topdown.sequence);
int new_proteins_with_same_being_end_diff_sequence = new_proteins.Count(t => t.AccessionList.Select(a => a.Split('_')[0].Split('-')[0]).Contains(topdown.accession.Split('_')[0]) && t.ProteolysisProducts.First().OneBasedBeginPosition == topdown.topdown_begin && t.ProteolysisProducts.First().OneBasedEndPosition == topdown.topdown_end && t.BaseSequence != topdown.sequence);
int count = old_proteins_with_same_begin_end_diff_sequence + new_proteins_with_same_being_end_diff_sequence;
ProteinWithGoTerms p = new ProteinWithGoTerms(topdown.sequence, topdown.accession.Split('_')[0].Split('-')[0] + "_" + topdown.topdown_begin + "frag" + topdown.topdown_end + (count > 0 ? "_" + count : ""), candidate_theoreticals.First().GeneNames.ToList(), candidate_theoreticals.First().OneBasedPossibleLocalizedModifications, new List <ProteolysisProduct>()
{
new ProteolysisProduct(topdown.topdown_begin, topdown.topdown_end, "full")
}, candidate_theoreticals.First().Name, candidate_theoreticals.First().FullName, false, false, candidate_theoreticals.First().DatabaseReferences, candidate_theoreticals.First().GoTerms);
p.topdown_protein = true;
new_proteins.Add(p);
}
}
else
{
Sweet.lollipop.topdown_proteoforms_no_theoretical.Add(topdown);
}
}
Sweet.lollipop.topdown_proteoforms = Sweet.lollipop.topdown_proteoforms.Except(Sweet.lollipop.topdown_proteoforms_no_theoretical).ToList();
expanded_proteins = expanded_proteins.Concat(new_proteins).ToArray();
}
public void EnterTheoreticalProteformFamily(string seq, ProteinWithGoTerms prot, IDictionary <int, List <Modification> > modifications, string accession, List <TheoreticalProteoform> theoretical_proteoforms, int decoy_number, IEnumerable <Modification> variableModifications)
{
List <TheoreticalProteoform> new_theoreticals = new List <TheoreticalProteoform>();
if (seq.Length > 3000 || seq.Any(s => !aaIsotopeMassList.ContainsKey(s)))
{
return;
}
//Calculate the properties of this sequence
double unmodified_mass = TheoreticalProteoform.CalculateProteoformMass(seq, new List <Ptm>());
int lysine_count = seq.Split('K').Length - 1;
bool check_contaminants = theoretical_proteins.Any(item => item.Key.ContaminantDB);
//Figure out the possible ptm sets
Dictionary <int, List <Modification> > possibleLocalizedMods = modifications.ToDictionary(kv => kv.Key, kv => new List <Modification>(kv.Value));
foreach (Modification m in variableModifications)
{
for (int i = 1; i <= prot.BaseSequence.Length; i++)
{
if (prot.BaseSequence[i - 1].ToString() == m.Target.ToString())
{
if (!possibleLocalizedMods.TryGetValue(i, out List <Modification> a))
{
possibleLocalizedMods.Add(i, new List <Modification> {
m
});
}
else
{
a.Add(m);
}
}
}
}
int ptm_set_counter = 1;
//if top-down protein sequence, only add PTMs from that top-down proteoforms (will happen in add_topdown_theoreticals method)
if (!prot.topdown_protein)
{
List <PtmSet> unique_ptm_groups = PtmCombos.get_combinations(possibleLocalizedMods, Sweet.lollipop.max_ptms, Sweet.lollipop.modification_ranks, Sweet.lollipop.mod_rank_first_quartile / 2, limit_triples_and_greater);
//Enumerate the ptm combinations with _P# to distinguish from the counts in ProteinSequenceGroups (_#G) and TheoreticalPfGps (_#T)
foreach (PtmSet ptm_set in unique_ptm_groups)
{
TheoreticalProteoform t =
new TheoreticalProteoform(
accession + "_P" + ptm_set_counter.ToString(),
prot.FullDescription + "_P" + ptm_set_counter.ToString() + (decoy_number < 0 ? "" : "_DECOY_" + decoy_number.ToString()),
seq,
(prot as ProteinSequenceGroup != null ? (prot as ProteinSequenceGroup).proteinWithGoTermList.ToArray() : new ProteinWithGoTerms[] { prot }),
unmodified_mass,
lysine_count,
ptm_set,
decoy_number < 0,
check_contaminants,
theoretical_proteins);
t.topdown_theoretical = prot.topdown_protein;
new_theoreticals.Add(t);
ptm_set_counter++;
}
}
add_topdown_theoreticals(prot, seq, accession, unmodified_mass, decoy_number, lysine_count, new_theoreticals, ptm_set_counter, Sweet.lollipop.modification_ranks, Sweet.lollipop.mod_rank_first_quartile / 2);
lock (theoretical_proteoforms) theoretical_proteoforms.AddRange(new_theoreticals);
}