public void TestUnabeledProteoformCommunityRelateWithNotches_ET()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.neucode_labeled = false;
Sweet.lollipop.et_use_notch = true;
Sweet.lollipop.et_low_mass_difference = -200;
Sweet.lollipop.et_high_mass_difference = 200;
// One experimental one theoretical protoeform; mass difference < 500 -- return 1
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, -1, true);
TheoreticalProteoform pf2 = ConstructorsForTesting.make_a_theoretical();
pf2.modified_mass = 1010.0;
pf2.lysine_count = 1;
pf2.is_target = true;
pf2.ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
ExperimentalProteoform[] paE = new ExperimentalProteoform[1];
TheoreticalProteoform[] paT = new TheoreticalProteoform[1];
paE[0] = pf1;
paT[0] = pf2;
List <ProteoformRelation> prList = new List <ProteoformRelation>();
prepare_for_et(new List <double> {
pf1.modified_mass - pf2.modified_mass
});
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, prList.Count);
//Da instead of ppm -- return 1
Sweet.lollipop.et_notch_ppm = false;
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, prList.Count);
//outside ppm tolerance - return 0
pf1.modified_mass = 1000;
pf2.modified_mass = 1010.5;
Sweet.lollipop.et_notch_ppm = true;
Sweet.lollipop.notch_tolerance_et = 1;
paE[0] = pf1;
paT[0] = pf2;
prepare_for_et(new List <double> {
-10
});
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
//outside Da toelrance - return 0.
pf1.modified_mass = 1000;
pf2.modified_mass = 1010.5;
Sweet.lollipop.et_notch_ppm = false;
Sweet.lollipop.notch_tolerance_et = .5;
paE[0] = pf1;
paT[0] = pf2;
prepare_for_et(new List <double> {
-10
});
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
// One experimental one theoretical protoeform; mass difference > 500 -- return 0
pf1.modified_mass = 1000;
pf2.modified_mass = 2000;
paE[0] = pf1;
paT[0] = pf2;
prepare_for_et(new List <double> {
pf1.modified_mass - pf2.modified_mass
});
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
//Two experimental one theoretical proteoforms; mass difference < 500 Da -- return 2
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, -1, true);
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("A2", 1010.0, -1, true);
TheoreticalProteoform pf5 = ConstructorsForTesting.make_a_theoretical();
pf5.modified_mass = 1020.0;
pf5.lysine_count = 1;
pf5.is_target = true;
pf5.ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
ExperimentalProteoform[] paE2 = new ExperimentalProteoform[2];
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(2, prList.Count);
//Two experimental one theoretical proteoforms; one mass difference >500 Da -- return 0
pf3.modified_mass = 1000;
pf4.modified_mass = 1010;
pf5.modified_mass = 2000;
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
//Two experimental one theoretical proteoforms; mass difference > 500 Da -- return 0
pf3.modified_mass = 1000;
pf4.modified_mass = 2000;
pf5.modified_mass = 3000;
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
//test methionine retention
pf1.modified_mass = 2131.04;
pf1.lysine_count = 1;
pf2.modified_mass = 2000;
pf2.lysine_count = 1;
pf2.begin = 2;
ModificationMotif motif;
ModificationMotif.TryGetMotif("M", out motif);
Modification m = new Modification("Met retention", _modificationType: "AminoAcid", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 131.04);
Sweet.lollipop.theoretical_database.all_mods_with_mass.Add(m);
Sweet.lollipop.theoretical_database.all_possible_ptmsets.Add(new PtmSet(new List <Ptm> {
new Ptm(-1, m)
}));
Sweet.lollipop.modification_ranks.Add(131.04, 2);
Sweet.lollipop.theoretical_database.possible_ptmset_dictionary = Sweet.lollipop.theoretical_database.make_ptmset_dictionary();
paE[0] = pf1;
paT[0] = pf2;
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, prList.Count);
}
public void test_construct_target_and_decoy_families()
{
SaveState.lollipop = new Lollipop();
SaveState.lollipop.neucode_labeled = true;
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("E1", 1000, 5, true);
ExperimentalProteoform pf2 = ConstructorsForTesting.ExperimentalProteoform("E2", 1020, 5, true);
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E3", 1040, 8, true);
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E4", 1050, 5, true);
TheoreticalProteoform t1 = ConstructorsForTesting.make_a_theoretical("t1", 1000, 5);
TheoreticalProteoform decoy1 = ConstructorsForTesting.make_a_theoretical("Decoy1", 1020, 5);
TestProteoformCommunityRelate.prepare_for_et(new List <double>()
{
0
});
SaveState.lollipop.target_proteoform_community = new ProteoformCommunity();
SaveState.lollipop.target_proteoform_community.experimental_proteoforms = new ExperimentalProteoform[4] {
pf1, pf2, pf3, pf4
};
SaveState.lollipop.target_proteoform_community.theoretical_proteoforms = new TheoreticalProteoform[1] {
t1
};
SaveState.lollipop.decoy_proteoform_communities.Add(SaveState.lollipop.decoy_community_name_prefix + "0", new ProteoformCommunity());
SaveState.lollipop.decoy_proteoform_communities[SaveState.lollipop.decoy_community_name_prefix + "0"].experimental_proteoforms = SaveState.lollipop.target_proteoform_community.experimental_proteoforms.Select(e => new ExperimentalProteoform(e)).ToArray();
SaveState.lollipop.decoy_proteoform_communities[SaveState.lollipop.decoy_community_name_prefix + "0"].theoretical_proteoforms = new TheoreticalProteoform[1] {
decoy1
};
SaveState.lollipop.et_relations = SaveState.lollipop.target_proteoform_community.relate(SaveState.lollipop.target_proteoform_community.experimental_proteoforms, SaveState.lollipop.target_proteoform_community.theoretical_proteoforms, ProteoformComparison.ExperimentalTheoretical, true, TestContext.CurrentContext.TestDirectory, true);
SaveState.lollipop.ee_relations = SaveState.lollipop.target_proteoform_community.relate(SaveState.lollipop.target_proteoform_community.experimental_proteoforms, SaveState.lollipop.target_proteoform_community.experimental_proteoforms, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, true);
SaveState.lollipop.relate_ed();
SaveState.lollipop.relate_ef();
foreach (ProteoformRelation pr in SaveState.lollipop.et_relations)
{
pr.set_nearby_group(SaveState.lollipop.et_relations, SaveState.lollipop.et_relations.Select(r => r.InstanceId).ToList());
}
foreach (ProteoformRelation pr in SaveState.lollipop.ee_relations)
{
pr.set_nearby_group(SaveState.lollipop.ee_relations, SaveState.lollipop.ee_relations.Select(r => r.InstanceId).ToList());
}
SaveState.lollipop.et_peaks = SaveState.lollipop.target_proteoform_community.accept_deltaMass_peaks(SaveState.lollipop.et_relations, SaveState.lollipop.ed_relations);
SaveState.lollipop.ee_peaks = SaveState.lollipop.target_proteoform_community.accept_deltaMass_peaks(SaveState.lollipop.ee_relations, SaveState.lollipop.ef_relations);
//one ED relation
Assert.AreEqual(1, SaveState.lollipop.ed_relations.Values.SelectMany(v => v).Count());
Assert.AreEqual(1, SaveState.lollipop.decoy_proteoform_communities[SaveState.lollipop.decoy_community_name_prefix + "0"].theoretical_proteoforms.SelectMany(t => t.relationships).Count());
//peak is unaccepted --> relation should be unaccepted, peak added to relation
Assert.IsFalse(SaveState.lollipop.ed_relations.First().Value.FirstOrDefault().Accepted); //should be false if peak unaccepted
Assert.AreEqual(1, SaveState.lollipop.et_peaks.Count);
Assert.IsNotNull(SaveState.lollipop.ed_relations.Values.SelectMany(v => v).First().peak);
//peak is accepted --> relation should be accepted, peak added to relation
SaveState.lollipop.clear_et();
SaveState.lollipop.min_peak_count_et = 1;
SaveState.lollipop.et_relations = SaveState.lollipop.target_proteoform_community.relate(SaveState.lollipop.target_proteoform_community.experimental_proteoforms, SaveState.lollipop.target_proteoform_community.theoretical_proteoforms, ProteoformComparison.ExperimentalTheoretical, true, TestContext.CurrentContext.TestDirectory, true);
SaveState.lollipop.relate_ed();
foreach (ProteoformRelation pr in SaveState.lollipop.et_relations)
{
pr.set_nearby_group(SaveState.lollipop.et_relations, SaveState.lollipop.et_relations.Select(r => r.InstanceId).ToList());
}
foreach (ProteoformRelation pr in SaveState.lollipop.ee_relations)
{
pr.set_nearby_group(SaveState.lollipop.ee_relations, SaveState.lollipop.ee_relations.Select(r => r.InstanceId).ToList());
}
SaveState.lollipop.et_peaks = SaveState.lollipop.target_proteoform_community.accept_deltaMass_peaks(SaveState.lollipop.et_relations, SaveState.lollipop.ed_relations);
SaveState.lollipop.et_peaks = SaveState.lollipop.target_proteoform_community.accept_deltaMass_peaks(SaveState.lollipop.et_relations, SaveState.lollipop.ed_relations);
Assert.IsTrue(SaveState.lollipop.ed_relations.First().Value.FirstOrDefault().Accepted); //should be true if peak accepted
Assert.IsNotNull(SaveState.lollipop.ed_relations.Values.SelectMany(v => v).First().peak);
Assert.AreEqual(1, SaveState.lollipop.ed_relations.Values.SelectMany(v => v).Count(r => r.Accepted)); //only 1 relation is in delta mass range of accepted peak - only 1 accepted
//2 ef relations, unaccepted because peak unaccepted
Assert.AreEqual(3, SaveState.lollipop.ef_relations.Values.SelectMany(v => v).Count()); //2 ef relations
Assert.AreEqual(0, SaveState.lollipop.ef_relations.Values.SelectMany(v => v).Count(r => r.Accepted)); //all peaks accepted are false, so relation accepted should be false
//one peak accepted --> one of the EF relations falls into range of peak and should be accepted
SaveState.lollipop.clear_ee();
SaveState.lollipop.min_peak_count_ee = 1;
SaveState.lollipop.ee_relations = SaveState.lollipop.target_proteoform_community.relate(SaveState.lollipop.target_proteoform_community.experimental_proteoforms, SaveState.lollipop.target_proteoform_community.experimental_proteoforms, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, true);
SaveState.lollipop.relate_ef();
SaveState.lollipop.ee_peaks = SaveState.lollipop.target_proteoform_community.accept_deltaMass_peaks(SaveState.lollipop.ee_relations, SaveState.lollipop.ef_relations);
Assert.AreEqual(3, SaveState.lollipop.ee_peaks.Count(p => p.Accepted));
Assert.AreEqual(3, SaveState.lollipop.ee_relations.Count);
Assert.AreEqual(1, SaveState.lollipop.ef_relations.Values.SelectMany(v => v).Count(r => r.Accepted)); //only 1 relation is in delta mass range of accepted peak - only 1 accepted
SaveState.lollipop.construct_target_and_decoy_families();
//should be 2 target families and 3 decoy families
//only make decoy relations out of accepted relations --> 1 family with accepted relations
Assert.AreEqual(2, SaveState.lollipop.target_proteoform_community.families.Count);
Assert.AreEqual(3, SaveState.lollipop.decoy_proteoform_communities[SaveState.lollipop.decoy_community_name_prefix + "0"].families.Count);
Assert.AreEqual(1, SaveState.lollipop.decoy_proteoform_communities[SaveState.lollipop.decoy_community_name_prefix + "0"].families.Count(f => f.relations.Count > 0));
Assert.AreEqual(2, SaveState.lollipop.decoy_proteoform_communities[SaveState.lollipop.decoy_community_name_prefix + "0"].families.SelectMany(f => f.relations).Count());
}
public void TestNeuCodeLabeledProteoformCommunityRelate_ET()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.neucode_labeled = true;
// One experimental one theoretical proteoforms; lysine count equal; mass difference < 500 -- return 1
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true);
TheoreticalProteoform pf2 = ConstructorsForTesting.make_a_theoretical();
pf2.modified_mass = 1010.0;
pf2.lysine_count = 1;
pf2.is_target = true;
pf2.ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
prepare_for_et(new List <double> {
pf1.modified_mass - pf2.modified_mass
});
ExperimentalProteoform[] paE = new ExperimentalProteoform[1];
TheoreticalProteoform[] paT = new TheoreticalProteoform[1];
paE[0] = pf1;
paT[0] = pf2;
List <ProteoformRelation> prList = new List <ProteoformRelation>();
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, prList.Count);
// One experimental one theoretical proteoforms; lysine count equal; mass difference > 500 -- return 0
pf1.modified_mass = 1000;
pf1.lysine_count = 1;
pf2.modified_mass = 2000;
pf2.lysine_count = 1;
prepare_for_et(new List <double> {
pf1.modified_mass - pf2.modified_mass
});
paE[0] = pf1;
paT[0] = pf2;
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
// One experimental one theoretical proteoforms; lysine count NOT equal; mass difference < 500 -- return 0
pf1.modified_mass = 1000;
pf1.lysine_count = 1;
pf2.modified_mass = 1100;
pf2.lysine_count = 2;
prepare_for_et(new List <double> {
pf1.modified_mass - pf2.modified_mass
});
paE[0] = pf1;
paT[0] = pf2;
prList = community.relate(paE, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
//Two experimental one theoretical proteoforms; lysine count equal; mass difference < 500 Da -- return 2
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true);
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("A2", 1010.0, 1, true);
TheoreticalProteoform pf5 = ConstructorsForTesting.make_a_theoretical();
pf5.modified_mass = 1020.0;
pf5.lysine_count = 1;
pf5.is_target = true;
pf5.ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
ExperimentalProteoform[] paE2 = new ExperimentalProteoform[2];
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(2, prList.Count);
//Two experimental one theoretical proteoforms; lysine count equal; one mass difference < 500 Da; one mass difference > 500 -- return 1
pf3.modified_mass = 1000;
pf3.lysine_count = 1;
pf4.modified_mass = 1500;
pf4.lysine_count = 1;
pf5.modified_mass = 1510;
pf5.lysine_count = 1;
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, prList.Count);
//Two experimental one theoretical proteoforms; lysine count NOT equal; mass difference < 500 Da -- return 0
pf3.modified_mass = 1000;
pf3.lysine_count = 1;
pf4.modified_mass = 1010;
pf4.lysine_count = 2;
pf5.modified_mass = 1020;
pf5.lysine_count = 3;
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
//Two experimental one theoretical proteoforms; lysine count equal; mass difference > 500 Da -- return 0
pf3.lysine_count = 1;
pf3.modified_mass = 1000;
pf4.lysine_count = 1;
pf4.modified_mass = 1600;
pf5.lysine_count = 1;
pf5.modified_mass = 2500;
paE2[0] = pf3;
paE2[1] = pf4;
paT[0] = pf5;
prepare_for_et(new List <double> {
pf3.modified_mass - pf5.modified_mass,
pf4.modified_mass - pf5.modified_mass,
});
prList = community.relate(paE2, paT, ProteoformComparison.ExperimentalTheoretical, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
}
public void get_interesting_goterm_families()
{
SaveState.lollipop = new Lollipop();
DatabaseReference d1 = new DatabaseReference("GO", "GO:1", new List <Tuple <string, string> > {
new Tuple <string, string>("term", "P:1")
});
DatabaseReference d2 = new DatabaseReference("GO", "GO:2", new List <Tuple <string, string> > {
new Tuple <string, string>("term", "P:2")
});
DatabaseReference d3 = new DatabaseReference("GO", "GO:1", new List <Tuple <string, string> > {
new Tuple <string, string>("term", "P:1")
});
GoTerm g1 = new GoTerm(d1);
GoTerm g2 = new GoTerm(d2);
GoTerm g3 = new GoTerm(d3);
ProteinWithGoTerms p1 = new ProteinWithGoTerms("", "T1", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, "T2", "T3", true, false, new List <DatabaseReference> {
d1
}, new List <GoTerm> {
g1
});
ProteinWithGoTerms p2 = new ProteinWithGoTerms("", "T2", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, "T2", "T3", true, false, new List <DatabaseReference> {
d2
}, new List <GoTerm> {
g2
});
ProteinWithGoTerms p3 = new ProteinWithGoTerms("", "T3", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, "T2", "T3", true, false, new List <DatabaseReference> {
d3
}, new List <GoTerm> {
g3
});
Dictionary <InputFile, Protein[]> dict = new Dictionary <InputFile, Protein[]>
{
{ new InputFile("fake.txt", Purpose.ProteinDatabase), new Protein[] { p1 } },
{ new InputFile("fake.txt", Purpose.ProteinDatabase), new Protein[] { p2 } },
{ new InputFile("fake.txt", Purpose.ProteinDatabase), new Protein[] { p3 } },
};
ExperimentalProteoform e1 = ConstructorsForTesting.ExperimentalProteoform("E");
ExperimentalProteoform e2 = ConstructorsForTesting.ExperimentalProteoform("E");
e1.quant.intensitySum = 1;
e1.quant.FDR = 0;
e1.quant.logFoldChange = 1;
e2.quant.intensitySum = 1;
e2.quant.FDR = 0;
e2.quant.logFoldChange = 1;
TheoreticalProteoform t = ConstructorsForTesting.make_a_theoretical("T1_T1_asdf", p1, dict);
TheoreticalProteoform u = ConstructorsForTesting.make_a_theoretical("T2_T1_asdf_asdf", p2, dict);
TheoreticalProteoform v = ConstructorsForTesting.make_a_theoretical("T3_T1_asdf_Asdf_Asdf", p3, dict);
t.ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
u.ExpandedProteinList = new List <ProteinWithGoTerms> {
p2
};
v.ExpandedProteinList = new List <ProteinWithGoTerms> {
p3
};
make_relation(e1, t);
//make_relation(e1, v); // we don't allow this to happen anymore... we only allow one ET conntection per E
make_relation(e2, u);
ProteoformFamily f = new ProteoformFamily(e1); // two theoreticals with the same GoTerms... expecting one GoTerm number but two theoretical proteins (now only one)
ProteoformFamily h = new ProteoformFamily(e2);
f.construct_family();
f.identify_experimentals();
h.construct_family();
h.identify_experimentals();
List <ProteoformFamily> families = new List <ProteoformFamily> {
f, h
};
t.family = f;
v.family = f;
e1.family = f;
u.family = h;
e2.family = h;
List <ExperimentalProteoform> fake_significant = new List <ExperimentalProteoform> {
e1
};
List <ProteinWithGoTerms> significant_proteins = SaveState.lollipop.getInducedOrRepressedProteins(fake_significant, 0, 1, 0);
List <GoTermNumber> gtn = SaveState.lollipop.getGoTermNumbers(significant_proteins, new List <ProteinWithGoTerms> {
p1, p2, p3
});
Assert.AreEqual(1, significant_proteins.Count);
Assert.AreEqual(1, gtn.Count);
Assert.AreEqual("1", gtn.First().Id);
Assert.AreEqual(0 - (decimal)Math.Log(2d / 3d, 2), gtn.First().log_odds_ratio);
List <ProteoformFamily> fams = SaveState.lollipop.getInterestingFamilies(gtn, families);
Assert.AreEqual(1, fams.Count);
Assert.AreEqual(1, fams[0].theoretical_proteoforms.Count);
}
public static ProteoformCommunity construct_two_families_with_potentially_colliding_theoreticals()
{
//Five experimental proteoforms, four relations (linear), second on not accepted into a peak, one peak; should give 2 families
ProteoformCommunity community = new ProteoformCommunity();
SaveState.lollipop = new Lollipop();
SaveState.lollipop.target_proteoform_community = community;
SaveState.lollipop.theoretical_database.uniprotModifications = new Dictionary <string, List <Modification> >
{
{ "unmodified", new List <Modification> {
ConstructorsForTesting.get_modWithMass("unmodified", 0)
} },
{ "fake", new List <Modification> {
ConstructorsForTesting.get_modWithMass("fake", 19)
} },
};
SaveState.lollipop.modification_ranks = new Dictionary <double, int> {
{ 0, 1 }, { 19, 2 }
};
SaveState.lollipop.mod_rank_sum_threshold = 2;
SaveState.lollipop.theoretical_database.all_possible_ptmsets = PtmCombos.generate_all_ptmsets(1, SaveState.lollipop.theoretical_database.uniprotModifications.SelectMany(kv => kv.Value).OfType <ModificationWithMass>().ToList(), SaveState.lollipop.modification_ranks, 1);
SaveState.lollipop.theoretical_database.all_mods_with_mass = SaveState.lollipop.theoretical_database.uniprotModifications.SelectMany(kv => kv.Value).OfType <ModificationWithMass>().ToList();
SaveState.lollipop.theoretical_database.possible_ptmset_dictionary = SaveState.lollipop.theoretical_database.make_ptmset_dictionary();
SaveState.lollipop.ee_max_mass_difference = 20;
SaveState.lollipop.peak_width_base_ee = 0.015;
SaveState.lollipop.min_peak_count_ee = 3; //needs to be high so that 0 peak accepted, other peak isn't....
SaveState.lollipop.min_peak_count_et = 2; //needs to be lower so the 2 ET relations are accepted
//TheoreticalProteoformGroup
InputFile f = new InputFile("fake.txt", Purpose.ProteinDatabase);
ProteinWithGoTerms p1 = new ProteinWithGoTerms("", p1_accession, new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, p1_name, p1_fullName, true, false, new List <DatabaseReference> {
p1_dbRef
}, new List <GoTerm> {
p1_goterm
});
Dictionary <InputFile, Protein[]> dict = new Dictionary <InputFile, Protein[]> {
{ f, new Protein[] { p1 } }
};
TheoreticalProteoform t = ConstructorsForTesting.make_a_theoretical("T1_asdf", "T1_asdf", 1234.56, p1, dict);
TheoreticalProteoformGroup pf1 = new TheoreticalProteoformGroup(new List <TheoreticalProteoform> {
t
});
TheoreticalProteoform pf2 = ConstructorsForTesting.make_a_theoretical("T1_asdf_pf2", "T1_asdf_1", 1234.56, dict);
//ExperimentalProteoforms
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E1", 0, 0, true);
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E2", 0, 0, true);
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("E3", 0, 0, true);
ExperimentalProteoform pf6 = ConstructorsForTesting.ExperimentalProteoform("E4", 0, 0, true);
ExperimentalProteoform pf7 = ConstructorsForTesting.ExperimentalProteoform("E5", 0, 0, true);
ExperimentalProteoform pf8 = ConstructorsForTesting.ExperimentalProteoform("E6", 0, 0, true);
TheoreticalProteoform pf9 = ConstructorsForTesting.make_a_theoretical("T1_asdf_pf9", "T1_asdf_1", 1253.56, dict);
community.theoretical_proteoforms = new TheoreticalProteoform[] { pf1, pf2, pf9 };
community.experimental_proteoforms = new ExperimentalProteoform[] { pf3, pf4, pf5, pf6, pf7, pf8 };
pf3.agg_mass = 1234.56;
pf4.agg_mass = 1234.56;
pf5.agg_mass = 1234.56;
pf6.agg_mass = 1253.56;
pf7.agg_mass = 1253.56;
pf8.agg_mass = 1253.56;
ProteoformComparison comparison13 = ProteoformComparison.ExperimentalTheoretical;
ProteoformComparison comparison23 = ProteoformComparison.ExperimentalTheoretical;
ProteoformComparison comparison34 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison45 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison56 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison67 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison78 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison89 = ProteoformComparison.ExperimentalTheoretical;
ConstructorsForTesting.make_relation(pf3, pf1, comparison13, 0);
ConstructorsForTesting.make_relation(pf3, pf2, comparison23, 0);
ConstructorsForTesting.make_relation(pf3, pf4, comparison34, 0);
ConstructorsForTesting.make_relation(pf4, pf5, comparison45, 0);
ConstructorsForTesting.make_relation(pf5, pf6, comparison56, 19); //not accepted
ConstructorsForTesting.make_relation(pf6, pf7, comparison67, 0);
ConstructorsForTesting.make_relation(pf7, pf8, comparison78, 0);
ConstructorsForTesting.make_relation(pf8, pf9, comparison89, 0);
List <ProteoformRelation> prs = new HashSet <ProteoformRelation>(community.experimental_proteoforms.SelectMany(p => p.relationships).Concat(community.theoretical_proteoforms.SelectMany(p => p.relationships))).ToList();
foreach (Proteoform p in prs.SelectMany(r => r.connected_proteoforms))
{
Assert.IsNotNull(p);
}
List <ProteoformRelation> prs_et = prs.Where(r => r.RelationType == ProteoformComparison.ExperimentalTheoretical).OrderBy(r => r.DeltaMass).ToList();
SaveState.lollipop.et_relations = prs_et;
List <ProteoformRelation> prs_ee = prs.Where(r => r.RelationType == ProteoformComparison.ExperimentalExperimental).OrderBy(r => r.DeltaMass).ToList();
SaveState.lollipop.ee_relations = prs_ee;
foreach (ProteoformRelation pr in prs_et)
{
pr.set_nearby_group(prs_et, prs_et.Select(r => r.InstanceId).ToList());
}
foreach (ProteoformRelation pr in prs_ee)
{
pr.set_nearby_group(prs_ee, prs_ee.Select(r => r.InstanceId).ToList());
}
Assert.AreEqual(3, pf1.relationships.First().nearby_relations_count); // 2 ET relations at 0 delta mass
Assert.AreEqual(3, pf2.relationships.First().nearby_relations_count);
Assert.AreEqual(4, pf4.relationships.First().nearby_relations_count); // 4 EE relations at 0 delta mass
Assert.AreEqual(4, pf5.relationships.First().nearby_relations_count);
Assert.AreEqual(1, pf6.relationships.First().nearby_relations_count); // 1 EE relation at 19 delta mass
Assert.AreEqual(4, pf7.relationships.First().nearby_relations_count);
Assert.AreEqual(4, pf8.relationships.First().nearby_relations_count);
community.accept_deltaMass_peaks(prs_et, new List <ProteoformRelation>());
community.accept_deltaMass_peaks(prs_ee, new List <ProteoformRelation>());
Assert.AreEqual(3, SaveState.lollipop.et_peaks.Count + SaveState.lollipop.ee_peaks.Count);
Assert.AreEqual(1, SaveState.lollipop.et_peaks.Where(peak => peak.Accepted).Count()); // 1 ET peak
Assert.AreEqual(1, SaveState.lollipop.ee_peaks.Where(peak => peak.Accepted).Count()); // 1 EE peak accepted
Assert.AreEqual(4, SaveState.lollipop.ee_peaks.Where(peak => peak.Accepted && peak.RelationType == ProteoformComparison.ExperimentalExperimental).First().grouped_relations.Count());
Assert.AreEqual(3, SaveState.lollipop.et_peaks.Where(peak => peak.Accepted && peak.RelationType == ProteoformComparison.ExperimentalTheoretical).First().grouped_relations.Count());
community.construct_families();
//Testing the identification of experimentals
//test with a modificationwithmass that's 0 mass, and then see that it crawls around and labels them each with growing ptm sets with that modification
//test that the relation.represented_modification gets set
Assert.True(SaveState.lollipop.et_relations.All(r => r.peak.DeltaMass != 19 || r.represented_ptmset == null));
Assert.True(SaveState.lollipop.et_relations.All(r => r.peak.DeltaMass != 0 || r.represented_ptmset.ptm_combination.First().modification.id == "unmodified"));
Assert.True(pf1 == pf3.linked_proteoform_references.First() || pf2 == pf3.linked_proteoform_references.First());
//test I don't get re-reassignments
Assert.AreEqual(pf3, pf4.linked_proteoform_references.Last()); //test that the proteoform.theoretical_reference gets set to each successive PF base
Assert.AreEqual((pf3.linked_proteoform_references.First() as TheoreticalProteoform).accession, (pf4.linked_proteoform_references.First() as TheoreticalProteoform).accession);
Assert.AreEqual((pf3.linked_proteoform_references.First() as TheoreticalProteoform).fragment, (pf4.linked_proteoform_references.First() as TheoreticalProteoform).fragment);
Assert.AreEqual(pf4, pf5.linked_proteoform_references.Last());
Assert.AreEqual((pf3.linked_proteoform_references.First() as TheoreticalProteoform).accession, (pf5.linked_proteoform_references.First() as TheoreticalProteoform).accession); //test that the accession gets carried all the way through the depth of connections
Assert.AreEqual((pf3.linked_proteoform_references.First() as TheoreticalProteoform).fragment, (pf5.linked_proteoform_references.First() as TheoreticalProteoform).fragment);
Assert.AreEqual(pf9, pf8.linked_proteoform_references.Last());
return(community);
}
public void test_goterm_analysis_with_custom_list()
{
SaveState.lollipop = new Lollipop();
DatabaseReference d1 = new DatabaseReference("GO", "GO:1", new List <Tuple <string, string> > {
new Tuple <string, string>("term", "P:1")
});
DatabaseReference d2 = new DatabaseReference("GO", "GO:2", new List <Tuple <string, string> > {
new Tuple <string, string>("term", "P:2")
});
DatabaseReference d3 = new DatabaseReference("GO", "GO:1", new List <Tuple <string, string> > {
new Tuple <string, string>("term", "P:1")
});
GoTerm g1 = new GoTerm(d1);
GoTerm g2 = new GoTerm(d2);
GoTerm g3 = new GoTerm(d3);
ProteinWithGoTerms p1 = new ProteinWithGoTerms("", "T1", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, "T2", "T3", true, false, new List <DatabaseReference> {
d1
}, new List <GoTerm> {
g1
});
ProteinWithGoTerms p2 = new ProteinWithGoTerms("", "T2", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, "T2", "T3", true, false, new List <DatabaseReference> {
d2
}, new List <GoTerm> {
g2
});
ProteinWithGoTerms p3 = new ProteinWithGoTerms("", "T3", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new int?[] { 0 }, new int?[] { 0 }, new string[] { "" }, "T2", "T3", true, false, new List <DatabaseReference> {
d3
}, new List <GoTerm> {
g3
});
Dictionary <InputFile, Protein[]> dict = new Dictionary <InputFile, Protein[]> {
{ new InputFile("fake.txt", Purpose.ProteinDatabase), new Protein[] { p1 } },
{ new InputFile("fake.txt", Purpose.ProteinDatabase), new Protein[] { p2 } },
{ new InputFile("fake.txt", Purpose.ProteinDatabase), new Protein[] { p3 } },
};
ExperimentalProteoform e1 = ConstructorsForTesting.ExperimentalProteoform("E");
ExperimentalProteoform e2 = ConstructorsForTesting.ExperimentalProteoform("E");
e1.quant.intensitySum = 1;
e1.quant.FDR = 0;
e1.quant.logFoldChange = 1;
e2.quant.intensitySum = 1;
e2.quant.FDR = 0;
e2.quant.logFoldChange = 1;
TheoreticalProteoform t = ConstructorsForTesting.make_a_theoretical("T1_T1_asdf", p1, dict);
TheoreticalProteoform u = ConstructorsForTesting.make_a_theoretical("T2_T1_asdf_asdf", p2, dict);
TheoreticalProteoform v = ConstructorsForTesting.make_a_theoretical("T3_T1_asdf_Asdf_Asdf", p3, dict);
t.ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
u.ExpandedProteinList = new List <ProteinWithGoTerms> {
p2
};
v.ExpandedProteinList = new List <ProteinWithGoTerms> {
p3
};
make_relation(e1, t);
make_relation(e1, v);
make_relation(e2, u);
ProteoformFamily f = new ProteoformFamily(e1); // two theoreticals with the same GoTerms... expecting one GoTerm number but two theoretical proteins
ProteoformFamily h = new ProteoformFamily(e2);
f.construct_family();
f.identify_experimentals();
h.construct_family();
h.identify_experimentals();
List <ProteoformFamily> families = new List <ProteoformFamily> {
f, h
};
t.family = f;
v.family = f;
e1.family = f;
u.family = h;
e2.family = h;
SaveState.lollipop.inducedOrRepressedProteins = SaveState.lollipop.getInducedOrRepressedProteins(new List <ExperimentalProteoform> {
e1
}, 0, 1, 0);
SaveState.lollipop.allTheoreticalProteins = true;
SaveState.lollipop.theoretical_database.expanded_proteins = new ProteinWithGoTerms[] { p1, p2, p3 };
SaveState.lollipop.backgroundProteinsList = Path.Combine(TestContext.CurrentContext.TestDirectory, "test_protein_list.txt");
SaveState.lollipop.GO_analysis();
Assert.AreEqual(1, SaveState.lollipop.inducedOrRepressedProteins.Count); // only taking one ET connection by definition in forming ET relations; only one is used in identify theoreticals
Assert.AreEqual(1, SaveState.lollipop.goTermNumbers.Count);
Assert.AreEqual("1", SaveState.lollipop.goTermNumbers.First().Id);
Assert.AreEqual(0 - (decimal)Math.Log(2d / 3d, 2), SaveState.lollipop.goTermNumbers.First().log_odds_ratio);
List <ProteoformFamily> fams = SaveState.lollipop.getInterestingFamilies(SaveState.lollipop.goTermNumbers, families);
Assert.AreEqual(1, fams.Count);
Assert.AreEqual(2, fams[0].theoretical_proteoforms.Count);
}
public void shift_et_peak_unlabeled()
{
SaveState.lollipop = new Lollipop();
ProteoformCommunity test_community = new ProteoformCommunity();
SaveState.lollipop.target_proteoform_community = test_community;
//Make a few experimental proteoforms
List <Component> n1 = TestExperimentalProteoform.generate_neucode_components(100);
List <Component> n2 = TestExperimentalProteoform.generate_neucode_components(200);
List <Component> n3 = TestExperimentalProteoform.generate_neucode_components(200);
List <Component> n4 = TestExperimentalProteoform.generate_neucode_components(200);
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("E1");
pf1.aggregated_components = n1;
ExperimentalProteoform pf2 = ConstructorsForTesting.ExperimentalProteoform("E2");
pf2.aggregated_components = n2;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E3");
pf3.aggregated_components = n3;
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E4");
pf4.aggregated_components = n4;
SaveState.lollipop.target_proteoform_community.experimental_proteoforms = new List <ExperimentalProteoform> {
pf1, pf2, pf3, pf4
}.ToArray();
//Connect them to theoreticals to form two peaks
ProteoformComparison comparison14 = ProteoformComparison.ExperimentalTheoretical;
ProteoformComparison comparison25 = ProteoformComparison.ExperimentalTheoretical;
ProteoformComparison comparison36 = ProteoformComparison.ExperimentalTheoretical;
ProteoformComparison comparison47 = ProteoformComparison.ExperimentalTheoretical;
TheoreticalProteoform pf5 = ConstructorsForTesting.make_a_theoretical();
TheoreticalProteoform pf6 = ConstructorsForTesting.make_a_theoretical();
TheoreticalProteoform pf7 = ConstructorsForTesting.make_a_theoretical();
TheoreticalProteoform pf8 = ConstructorsForTesting.make_a_theoretical();
ProteoformRelation pr1 = new ProteoformRelation(pf1, pf5, comparison14, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr2 = new ProteoformRelation(pf2, pf6, comparison25, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr3 = new ProteoformRelation(pf3, pf7, comparison36, 1, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr4 = new ProteoformRelation(pf4, pf8, comparison47, 1, TestContext.CurrentContext.TestDirectory);
List <ProteoformRelation> prs = new List <ProteoformRelation> {
pr1, pr2, pr3, pr4
};
foreach (ProteoformRelation pr in prs)
{
pr.set_nearby_group(prs, prs.Select(r => r.InstanceId).ToList());
}
test_community.accept_deltaMass_peaks(prs, new List <ProteoformRelation>());
Assert.AreEqual(2, SaveState.lollipop.et_peaks.Count);
//Shift the peaks, which shifts all of the proteoforms
DeltaMassPeak d2 = SaveState.lollipop.et_peaks[1];
d2.shift_experimental_masses(-1, false);
Assert.IsTrue(pf3.mass_shifted);
Assert.IsTrue(pf4.mass_shifted);
foreach (Component c in n3.Concat(n4))
{
Assert.AreEqual(-1.0 * Lollipop.MONOISOTOPIC_UNIT_MASS, c.manual_mass_shift);
Assert.AreEqual(200 - 1.0 * Lollipop.MONOISOTOPIC_UNIT_MASS, c.weighted_monoisotopic_mass);
}
}
public void TestDeltaMassPeakConstructor()
{
SaveState.lollipop.enter_input_files(new string[] { Path.Combine(TestContext.CurrentContext.TestDirectory, "ptmlist.txt") }, Lollipop.acceptable_extensions[2], Lollipop.file_types[2], SaveState.lollipop.input_files);
ConstructorsForTesting.read_mods();
SaveState.lollipop.et_high_mass_difference = 250;
SaveState.lollipop.et_low_mass_difference = -250;
SaveState.lollipop.peak_width_base_ee = 0.015;
SaveState.lollipop.peak_width_base_et = 0.015;
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("acession1");
TheoreticalProteoform pf2 = ConstructorsForTesting.make_a_theoretical();
ProteoformComparison relation_type = ProteoformComparison.ExperimentalTheoretical;
double delta_mass = 1 - 1e-7;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("acession3");
TheoreticalProteoform pf4 = ConstructorsForTesting.make_a_theoretical();
ProteoformComparison relation_type2 = ProteoformComparison.ExperimentalTheoretical;
double delta_mass2 = 1;
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("acession5");
TheoreticalProteoform pf6 = ConstructorsForTesting.make_a_theoretical();
ProteoformComparison relation_type3 = ProteoformComparison.ExperimentalTheoretical;
double delta_mass3 = 1 + 1e-7;
ExperimentalProteoform pf55 = ConstructorsForTesting.ExperimentalProteoform("acession5");
TheoreticalProteoform pf65 = ConstructorsForTesting.make_a_theoretical();
ProteoformComparison relation_type35 = ProteoformComparison.ExperimentalTheoretical;
double delta_mass35 = 1 + 2e-7;
List <ProteoformRelation> theList = new List <ProteoformRelation>();
theList.Add(new ProteoformRelation(pf1, pf2, relation_type, delta_mass, TestContext.CurrentContext.TestDirectory));
theList.Add(new ProteoformRelation(pf3, pf4, relation_type2, delta_mass2, TestContext.CurrentContext.TestDirectory));
theList.Add(new ProteoformRelation(pf5, pf6, relation_type3, delta_mass3, TestContext.CurrentContext.TestDirectory));
theList.Add(new ProteoformRelation(pf55, pf65, relation_type35, delta_mass35, TestContext.CurrentContext.TestDirectory));
ProteoformRelation base_relation = new ProteoformRelation(pf3, pf4, relation_type2, delta_mass2, TestContext.CurrentContext.TestDirectory);
//base_relation.nearby_relations = base_relation.set_nearby_group(theList, theList.Select(r => r.InstanceId).ToList());
Console.WriteLine("Creating deltaMassPeak");
DeltaMassPeak deltaMassPeak = new DeltaMassPeak(base_relation, theList);
Console.WriteLine("Created deltaMassPeak");
Assert.AreEqual(0, deltaMassPeak.peak_group_fdr);
Dictionary <string, List <ProteoformRelation> > decoy_relations = new Dictionary <string, List <ProteoformRelation> >();
decoy_relations["decoyDatabase1"] = new List <ProteoformRelation>();
ExperimentalProteoform pf7 = ConstructorsForTesting.ExperimentalProteoform("experimental1");
TheoreticalProteoform pf8 = ConstructorsForTesting.make_a_theoretical();
ProteoformComparison relation_type4 = ProteoformComparison.ExperimentalDecoy;
double delta_mass4 = 1;
ProteoformRelation decoy_relation = new ProteoformRelation(pf7, pf8, relation_type4, delta_mass4, TestContext.CurrentContext.TestDirectory);
decoy_relations["decoyDatabase1"].Add(decoy_relation);
deltaMassPeak.calculate_fdr(decoy_relations);
Assert.AreEqual(0.25, deltaMassPeak.peak_group_fdr); // 1 decoy database, (1 decoy relation, median=1), 4 target relations
decoy_relations["decoyDatabase2"] = new List <ProteoformRelation>();
decoy_relations["decoyDatabase2"].Add(decoy_relation);
decoy_relations["decoyDatabase2"].Add(decoy_relation);
deltaMassPeak.calculate_fdr(decoy_relations);
Assert.AreEqual(0.375, deltaMassPeak.peak_group_fdr); // 2 decoy databases (1 & 2 decoy relations, median=1.5), 4 target relations
}
public void TestCorrectTopDownID()
{
TopDownProteoform td = ConstructorsForTesting.TopDownProteoform("TD1", 1000, 40);
//linked reference null - should be false
td.set_correct_id();
Assert.IsFalse(td.correct_id);
TheoreticalProteoform t = ConstructorsForTesting.make_a_theoretical();
t.ExpandedProteinList.First().AccessionList.Add("TD1");
td.linked_proteoform_references = new List <Proteoform>()
{
t
};
//no PTMs diff begin fail
td.begin = 10;
td.topdown_begin = 10;
td.end = 20;
td.topdown_end = 30;
td.ptm_set = new PtmSet(new List <Ptm>());
td.topdown_ptm_set = new PtmSet(new List <Ptm>());
td.set_correct_id();
Assert.IsFalse(td.correct_id);
//no PTMs diff end fail
td.begin = 10;
td.topdown_begin = 20;
td.end = 30;
td.topdown_end = 30;
td.set_correct_id();
Assert.IsFalse(td.correct_id);
//no PTMs same pass
td.begin = 10;
td.topdown_begin = 10;
td.end = 30;
td.topdown_end = 30;
td.set_correct_id();
Assert.IsTrue(td.correct_id);
//same begin and end, T has more PTMs
ModificationMotif motif;
ModificationMotif.TryGetMotif("K", out motif);
td.ptm_set = new PtmSet(new List <Ptm>()
{
new Ptm(15, new Modification("Acetylation", _modificationType: "type", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 42.02))
});
td.set_correct_id();
Assert.IsFalse(td.correct_id);
//same begin and end TD has more of a PTM type
td.ptm_set = new PtmSet(new List <Ptm>());
td.topdown_ptm_set = new PtmSet(new List <Ptm>()
{
new Ptm(15, new Modification("Acetylation", _modificationType: "type", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 42.02))
});
td.set_correct_id();
Assert.IsFalse(td.correct_id);
//same begin and end and PTMs
td.ptm_set = new PtmSet(new List <Ptm>()
{
new Ptm(15, new Modification("Acetylation", _modificationType: "type", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 42.02))
});
td.topdown_ptm_set = new PtmSet(new List <Ptm>()
{
new Ptm(15, new Modification("Acetylation", _modificationType: "type", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 42.02))
});
td.set_correct_id();
Assert.IsTrue(td.correct_id);
}
public void TestRelateTD()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.neucode_labeled = false;
Sweet.lollipop.maximum_missed_monos = 1;
Sweet.lollipop.agg_minBiorepsWithObservations = 0;
InputFile f = new InputFile("path", Purpose.Identification);
Sweet.lollipop.input_files.Add(f);
// Two proteoforms; lysine count equal; mass difference < 250 -- return 1
Component c1 = new Component();
c1.weighted_monoisotopic_mass = 1000.0;
c1.rt_apex = 45;
c1.accepted = true;
c1.id = 1.ToString();
c1.intensity_sum = 1e6;
c1.input_file = f;
c1.charge_states = new List <ChargeState>()
{
new ChargeState(1, c1.intensity_sum, c1.weighted_monoisotopic_mass)
};
Component c2 = new Component();
c2.weighted_monoisotopic_mass = 1000.0;
c2.rt_apex = 85;
c2.accepted = true;
c2.input_file = f;
c2.intensity_sum = 1e6;
c2.charge_states = new List <ChargeState>()
{
new ChargeState(1, c2.intensity_sum, c2.weighted_monoisotopic_mass)
};
c2.id = 2.ToString();
Component c3 = new Component();
c3.weighted_monoisotopic_mass = 1131.04;
c3.rt_apex = 45;
c3.accepted = true;
c3.input_file = f;
c3.intensity_sum = 1e6;
c3.charge_states = new List <ChargeState>()
{
new ChargeState(1, c3.intensity_sum, c3.weighted_monoisotopic_mass)
};
c3.id = 3.ToString();
Component c4 = new Component();
c4.weighted_monoisotopic_mass = 2000.00;
c4.rt_apex = 45;
c4.accepted = true;
c4.input_file = new InputFile("path", Purpose.Identification);
c4.intensity_sum = 1e6;
c4.charge_states = new List <ChargeState>()
{
new ChargeState(1, c4.intensity_sum, c4.weighted_monoisotopic_mass)
};
c4.id = 4.ToString();
Component c5 = new Component();
c5.weighted_monoisotopic_mass = 1001.0;
c5.rt_apex = 45;
c5.accepted = true;
c5.input_file = f;
c5.intensity_sum = 1e6;
c5.charge_states = new List <ChargeState>()
{
new ChargeState(1, c5.intensity_sum, c5.weighted_monoisotopic_mass)
};
c5.id = 2.ToString();
List <IAggregatable> components = new List <IAggregatable>()
{
c1, c2, c3, c4, c5
};
Sweet.lollipop.raw_experimental_components = components.OfType <Component>().ToList();
TopDownProteoform td1 = ConstructorsForTesting.TopDownProteoform("ACCESSION_1", 1000.0, 45);
TopDownProteoform td2 = ConstructorsForTesting.TopDownProteoform("ACCESSION_2", 1001.0, 85);
TopDownProteoform td3 = ConstructorsForTesting.TopDownProteoform("ACCESSION_3", 1131.04, 45);
TheoreticalProteoform t1 = ConstructorsForTesting.make_a_theoretical("ACCESSION", 1000.0, 1);
//need to make theoretical accession database
TestProteoformCommunityRelate.prepare_for_et(new List <double>()
{
0
});
Sweet.lollipop.target_proteoform_community.community_number = -100;
Sweet.lollipop.theoretical_database.theoreticals_by_accession = new Dictionary <int, Dictionary <string, List <TheoreticalProteoform> > >();
Sweet.lollipop.theoretical_database.theoreticals_by_accession.Add(-100, new Dictionary <string, List <TheoreticalProteoform> >());
Sweet.lollipop.theoretical_database.theoreticals_by_accession[-100].Add(t1.accession, new List <TheoreticalProteoform>()
{
t1
});
//need to make decon error top "deconvolution error"
ModificationMotif motif;
ModificationMotif.TryGetMotif("S", out motif);
Modification m = new Modification("id", _modificationType: "modtype", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 1);
Sweet.lollipop.theoretical_database.all_mods_with_mass.Add(m);
PtmSet set = new PtmSet(new List <Ptm> {
new Ptm(-1, m)
});
Sweet.lollipop.theoretical_database.all_possible_ptmsets.Add(set);
Sweet.lollipop.modification_ranks.Add(-1.0023, 2);
Sweet.lollipop.theoretical_database.possible_ptmset_dictionary.Add(-1.0, new List <PtmSet>()
{
set
});
//need missing error
Modification m2 = new Modification("id", _modificationType: "modtype", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 1);
Sweet.lollipop.theoretical_database.all_mods_with_mass.Add(m2);
PtmSet set2 = new PtmSet(new List <Ptm> {
new Ptm(-1, m2)
});
Sweet.lollipop.theoretical_database.all_possible_ptmsets.Add(set2);
Sweet.lollipop.modification_ranks.Add(-87.03, 2);
Sweet.lollipop.theoretical_database.possible_ptmset_dictionary.Add(-87.0, new List <PtmSet>()
{
set2
});
Sweet.lollipop.target_proteoform_community.theoretical_proteoforms = new List <TheoreticalProteoform>()
{
t1
}.ToArray();
Sweet.lollipop.topdown_proteoforms = new List <TopDownProteoform> {
td1, td2, td3
};
Sweet.lollipop.add_td_proteoforms = true;
Sweet.lollipop.aggregate_proteoforms(Sweet.lollipop.validate_proteoforms, Sweet.lollipop.raw_neucode_pairs, Sweet.lollipop.raw_experimental_components, Sweet.lollipop.raw_quantification_components, 0);
List <ProteoformRelation> relations = Sweet.lollipop.target_proteoform_community.relate(Sweet.lollipop.target_proteoform_community.experimental_proteoforms, Sweet.lollipop.target_proteoform_community.theoretical_proteoforms, ProteoformComparison.ExperimentalTheoretical, true, Environment.CurrentDirectory, true);
List <DeltaMassPeak> peaks = Sweet.lollipop.target_proteoform_community.accept_deltaMass_peaks(relations, new Dictionary <string, List <ProteoformRelation> >());
//should have 4 experimental proteoforms -- 3 topdown, 1 not topdown experimental
Assert.AreEqual(3, Sweet.lollipop.target_proteoform_community.experimental_proteoforms.Count(e => e.topdown_id));
Assert.AreEqual(4, Sweet.lollipop.target_proteoform_community.experimental_proteoforms.Count());
Assert.AreEqual(1, relations.Count);
Assert.AreEqual(1, relations.Count(r => r.RelationType == ProteoformComparison.ExperimentalTheoretical && (r.connected_proteoforms[0] as ExperimentalProteoform).topdown_id));
Assert.AreEqual(1, td1.relationships.Count(r => r.RelationType == ProteoformComparison.ExperimentalTheoretical && (r.connected_proteoforms[0] as ExperimentalProteoform).topdown_id));
Assert.AreEqual(0, td2.relationships.Count());
Assert.AreEqual(0, td3.relationships.Count());
//accession 3 has higher score... gets td
Sweet.lollipop.clear_td();
td3 = ConstructorsForTesting.TopDownProteoform("ACCESSION_3", 1000.0, 45);
TopDownHit h3 = new TopDownHit();
h3.score = 100;
td3.topdown_hits = new List <TopDownHit>()
{
h3
};
TopDownHit h4 = new TopDownHit();
h4.score = 1;
TopDownProteoform td4 = ConstructorsForTesting.TopDownProteoform("ACCESSION_4", 1001.0, 45);
td4.topdown_hits = new List <TopDownHit>()
{
h4
};
Sweet.lollipop.topdown_proteoforms = new List <TopDownProteoform> {
td4, td3
};
Sweet.lollipop.topdown_proteoforms.OrderBy(p => p.modified_mass);
Sweet.lollipop.aggregate_proteoforms(Sweet.lollipop.validate_proteoforms, Sweet.lollipop.raw_neucode_pairs, Sweet.lollipop.raw_experimental_components, Sweet.lollipop.raw_quantification_components, 0);
Assert.AreEqual(0, Math.Round(td3.modified_mass - td3.matching_experimental.modified_mass, 0));
Assert.IsNull(td4.matching_experimental);
}