public void test_contaminant_check()
{
InputFile f = new InputFile("fake.txt", Purpose.ProteinDatabase);
f.ContaminantDB = true;
InputFile g = new InputFile("fake.txt", Purpose.ProteinDatabase);
InputFile h = new InputFile("fake.txt", Purpose.ProteinDatabase);
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>(), new List <GoTerm>());
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>(), new List <GoTerm>());
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>(), new List <GoTerm>());
Dictionary <InputFile, Protein[]> dict = new Dictionary <InputFile, Protein[]> {
{ f, new Protein[] { p1 } },
{ g, new Protein[] { p2 } },
{ h, new Protein[] { p3 } },
};
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);
TheoreticalProteoform w = new TheoreticalProteoformGroup(new List <TheoreticalProteoform> {
v, u, t
}.OrderByDescending(theo => theo.contaminant ? 1 : 0));
Assert.True(w.contaminant);
Assert.True(w.accession.Contains(p1.Accession));
//Not contaminant
TheoreticalProteoform x = new TheoreticalProteoformGroup(new List <TheoreticalProteoform> {
v, u
});
Assert.False(x.contaminant);
//PTM mass test
Assert.AreEqual(0, t.ptm_mass);
}
public void nodes_table_gives_meaningful_modified_theoreticals2()
{
ModificationMotif motif;
ModificationMotif.TryGetMotif("K", out motif);
Modification m = new Modification("id", _modificationType: "modtype", _target: motif, _locationRestriction: "Anywhere.", _monoisotopicMass: 1);
Proteoform p = ConstructorsForTesting.make_a_theoretical();
ProteoformFamily f = new ProteoformFamily(p);
f.construct_family();
string node_table = CytoscapeScript.get_cytoscape_nodes_tsv(new List <ProteoformFamily> {
f
},
null,
CytoscapeScript.color_scheme_names[0], Lollipop.edge_labels[0], Lollipop.node_labels[0], Lollipop.node_positioning[0], 2,
f.theoretical_proteoforms, false, Lollipop.gene_name_labels[1]);
Assert.True(node_table.Contains(CytoscapeScript.modified_theoretical_label));
Assert.AreNotEqual(f.theoretical_proteoforms[0].accession, CytoscapeScript.get_proteoform_shared_name(p, Lollipop.node_labels[0], 2));
}
public void aggregate_out_of_middle_monoisotopic_tolerance()
{
missed_monoisotopics -= 1;
Sweet.lollipop.set_missed_monoisotopic_range();
double max_monoisotopic_mass = starter_mass + missed_monoisotopics * Lollipop.MONOISOTOPIC_UNIT_MASS;
double min_monoisotopic_mass = starter_mass - missed_monoisotopics * Lollipop.MONOISOTOPIC_UNIT_MASS;
Sweet.lollipop.neucode_labeled = true;
List <IAggregatable> components = generate_neucode_components(starter_mass);
// below lowest monoisotopic tolerance
components[1].charge_states.Clear(); // must clear charge states because you can't set the weighted monoisotopic mass if there are charge states.
components[1].weighted_monoisotopic_mass = (min_monoisotopic_mass - min_monoisotopic_mass / 1000000 * Convert.ToDouble(Sweet.lollipop.mass_tolerance) - Double.MinValue);
ExperimentalProteoform e = ConstructorsForTesting.ExperimentalProteoform("E1", components[0], components, empty_quant_components_list, true);
Assert.AreEqual(1, e.aggregated.Count);
// above highest monoisotopic tolerance
components[1].weighted_monoisotopic_mass = (max_monoisotopic_mass + max_monoisotopic_mass / 1000000 * Convert.ToDouble(Sweet.lollipop.mass_tolerance) + Double.MinValue);
e = ConstructorsForTesting.ExperimentalProteoform("E1", components[0], components, empty_quant_components_list, true);
Assert.AreEqual(1, e.aggregated.Count);
}
public void wrong_relation_shifting()
{
ProteoformCommunity test_community = new ProteoformCommunity();
SaveState.lollipop.target_proteoform_community = test_community;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E1");
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E2");
ProteoformComparison wrong_comparison = ProteoformComparison.ExperimentalExperimental;
ProteoformRelation pr2 = new ProteoformRelation(pf3, pf4, wrong_comparison, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr3 = new ProteoformRelation(pf3, pf4, wrong_comparison, 0, TestContext.CurrentContext.TestDirectory);
List <ProteoformRelation> prs = new List <ProteoformRelation> {
pr2, pr3
};
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.False(SaveState.lollipop.et_peaks[0].shift_experimental_masses(1, true));
}
public void nodes_table_gives_meaningful_modified_theoreticals2()
{
ModificationMotif motif;
ModificationMotif.TryGetMotif("K", out motif);
ModificationWithMass m = new ModificationWithMass("oxidation", new Tuple <string, string>("", ""), motif, ModificationSites.K, 1, new Dictionary <string, IList <string> >(), new List <double>(), new List <double>(), "");
Proteoform p = ConstructorsForTesting.make_a_theoretical();
ProteoformFamily f = new ProteoformFamily(p);
f.construct_family();
string node_table = CytoscapeScript.get_cytoscape_nodes_tsv(new List <ProteoformFamily> {
f
},
false,
CytoscapeScript.color_scheme_names[0], Lollipop.edge_labels[0], Lollipop.node_labels[0], Lollipop.node_positioning[0], 2,
f.theoretical_proteoforms, false, Lollipop.gene_name_labels[1]);
Assert.True(node_table.Contains(CytoscapeScript.modified_theoretical_label));
Assert.AreNotEqual(f.theoretical_proteoforms[0].accession, CytoscapeScript.get_proteoform_shared_name(p, Lollipop.node_labels[0], 2));
}
public void test_same_ptmset()
{
PtmSet set1 = new PtmSet(new List <Ptm>()
{
new Ptm(1, ConstructorsForTesting.get_modWithMass("id1", 5)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id2", 6)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id4", 6))
});
PtmSet set2 = new PtmSet(new List <Ptm>()
{
new Ptm(1, ConstructorsForTesting.get_modWithMass("id1", 5)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id2", 6)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id4", 6))
});
Assert.IsTrue(set1.same_ptmset(set2, true));
Assert.IsTrue(set1.same_ptmset(set2, false));
set2 = new PtmSet(new List <Ptm>()
{
new Ptm(1, ConstructorsForTesting.get_modWithMass("id1", 5)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id2", 6)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id3", 6))
});
Assert.IsFalse(set1.same_ptmset(set2, true));
Assert.IsFalse(set1.same_ptmset(set2, false));
Assert.IsFalse(set2.same_ptmset(set1, true));
Assert.IsFalse(set2.same_ptmset(set1, false));
set2 = new PtmSet(new List <Ptm>()
{
new Ptm(1, ConstructorsForTesting.get_modWithMass("id1", 5)),
new Ptm(2, ConstructorsForTesting.get_modWithMass("id2", 6)),
});
Assert.IsFalse(set1.same_ptmset(set2, true));
Assert.IsFalse(set1.same_ptmset(set2, false));
Assert.IsFalse(set2.same_ptmset(set1, true));
Assert.IsFalse(set2.same_ptmset(set1, false));
}
public void unlabeled_agg()
{
SaveState.lollipop.min_num_bioreps = 0;
double max_monoisotopic_mass = TestExperimentalProteoform.starter_mass + TestExperimentalProteoform.missed_monoisotopics * Lollipop.MONOISOTOPIC_UNIT_MASS;
double min_monoisotopic_mass = TestExperimentalProteoform.starter_mass - TestExperimentalProteoform.missed_monoisotopics * Lollipop.MONOISOTOPIC_UNIT_MASS;
List <Component> components = TestExperimentalProteoform.generate_unlabeled_components(TestExperimentalProteoform.starter_mass);
SaveState.lollipop.neucode_labeled = false;
SaveState.lollipop.remaining_components = new List <Component>(components);
SaveState.lollipop.remaining_verification_components = new List <Component>(components);
ExperimentalProteoform e = ConstructorsForTesting.ExperimentalProteoform("E");
e.root = components[0];
e.aggregate();
e.verify();
Assert.AreEqual(2, e.aggregated_components.Count);
Assert.AreEqual(2, e.lt_verification_components.Count);
Assert.AreEqual(0, e.hv_verification_components.Count); // everything goes into light with unlabeled
Assert.AreEqual(0, e.lt_quant_components.Count); // no quantitation for unlabeled, yet
Assert.AreEqual(0, e.hv_quant_components.Count);
}
public void cytoscape_improper_build_folder()
{
TheoreticalProteoform t = ConstructorsForTesting.make_a_theoretical();
ExperimentalProteoform e = ConstructorsForTesting.ExperimentalProteoform("E1");
ProteoformRelation et = new ProteoformRelation(e, t, ProteoformComparison.ExperimentalTheoretical, 0, TestContext.CurrentContext.TestDirectory);
e.agg_intensity = 999.99;
e.quant.TusherValues1.numeratorIntensitySum = 444.44m;
e.quant.TusherValues1.denominatorIntensitySum = 333.33m;
e.quant.intensitySum = 777.77m;
List <ProteoformFamily> f = new List <ProteoformFamily> {
new ProteoformFamily(e)
};
f.First().construct_family();
string message = CytoscapeScript.write_cytoscape_script(f, f,
"", "", "test",
null, false, false,
CytoscapeScript.color_scheme_names[0], Lollipop.edge_labels[0], Lollipop.node_labels[0], CytoscapeScript.node_label_positions[0], Lollipop.node_positioning[0], 2,
false, Lollipop.gene_name_labels[1]);
Assert.False(message.StartsWith("Finished"));
}
public void unlabeled_agg()
{
double max_monoisotopic_mass = TestExperimentalProteoform.starter_mass + TestExperimentalProteoform.missed_monoisotopics * Lollipop.MONOISOTOPIC_UNIT_MASS;
double min_monoisotopic_mass = TestExperimentalProteoform.starter_mass - TestExperimentalProteoform.missed_monoisotopics * Lollipop.MONOISOTOPIC_UNIT_MASS;
List <IAggregatable> components = TestExperimentalProteoform.generate_unlabeled_components(TestExperimentalProteoform.starter_mass);
Sweet.lollipop.neucode_labeled = false;
Sweet.lollipop.remaining_to_aggregate = new List <IAggregatable>(components);
Sweet.lollipop.remaining_verification_components = new HashSet <Component>(components.OfType <Component>());
Sweet.lollipop.missed_monoisotopics_range = Enumerable.Range(-3, 3 * 2 + 1).ToList();
ExperimentalProteoform e = ConstructorsForTesting.ExperimentalProteoform("E");
e.root = components[0];
e.aggregate();
e.verify();
Assert.AreEqual(2, e.aggregated.Count);
Assert.AreEqual(2, e.lt_verification_components.Count);
Assert.AreEqual(0, e.hv_verification_components.Count); // everything goes into light with unlabeled
Assert.AreEqual(0, e.lt_quant_components.Count); // no quantitation for unlabeled, yet
Assert.AreEqual(0, e.hv_quant_components.Count);
}
public void assign_quant_components_large_tolerance_split_range()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.neucode_labeled = true;
Sweet.lollipop.mass_tolerance = 10; //ppm
Sweet.lollipop.missed_monoisotopics_range = Enumerable.Range(-3, 3 * 2 + 1).ToList();
ExperimentalProteoform e = ConstructorsForTesting.ExperimentalProteoform("", 20000, 2, true); // tolerance is 0.2 Da
double hv_mass = e.agg_mass + e.lysine_count * Lollipop.NEUCODE_LYSINE_MASS_SHIFT; // 20000.0703
Component bb = new Component();
Component cc = new Component();
Component dd = new Component();
Component ee = new Component();
Component ff = new Component();
Component gg = new Component();
Component hh = new Component();
Component ii = new Component();
bb.weighted_monoisotopic_mass = 19999.79;
cc.weighted_monoisotopic_mass = 19999.99;
dd.weighted_monoisotopic_mass = 20000;
ee.weighted_monoisotopic_mass = 20000.03;
//boundary is 20000.036
ff.weighted_monoisotopic_mass = 20000.04;
gg.weighted_monoisotopic_mass = 20000.07;
hh.weighted_monoisotopic_mass = 20000.08;
ii.weighted_monoisotopic_mass = 20000.28;
Sweet.lollipop.remaining_quantification_components = new HashSet <Component> {
bb, cc, dd, ee, ff, gg, hh, ii
};
e.assign_quantitative_components();
Assert.AreEqual(3, e.lt_quant_components.Count);
Assert.AreEqual(3, e.hv_quant_components.Count);
Assert.False(e.lt_quant_components.Any(c => e.hv_quant_components.Contains(c)));
}
public void get_interesting_goterm_families()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.theoretical_database.aaIsotopeMassList = new AminoAcidMasses(Sweet.lollipop.carbamidomethylation, Sweet.lollipop.neucode_labeled).AA_Masses;
Sweet.lollipop.significance_by_permutation = true;
Sweet.lollipop.significance_by_log2FC = false;
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("ASDF", "T1", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new List <ProteolysisProduct> {
new ProteolysisProduct(0, 0, "")
}, "T2", "T3", true, false, new List <DatabaseReference> {
d1
}, new List <GoTerm> {
g1
});
ProteinWithGoTerms p2 = new ProteinWithGoTerms("ASDF", "T2", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new List <ProteolysisProduct> {
new ProteolysisProduct(0, 0, "")
}, "T2", "T3", true, false, new List <DatabaseReference> {
d2
}, new List <GoTerm> {
g2
});
ProteinWithGoTerms p3 = new ProteinWithGoTerms("ASDF", "T3", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new List <ProteolysisProduct> {
new ProteolysisProduct(0, 0, "")
}, "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.TusherValues1.significant = true;
e1.quant.tusherlogFoldChange = 1;
e2.quant.intensitySum = 1;
e2.quant.TusherValues1.significant = true;
e2.quant.tusherlogFoldChange = 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
};
t.begin = 1;
t.end = 1;
u.begin = 1;
u.end = 1;
v.begin = 1;
v.end = 1;
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 = Sweet.lollipop.getInducedOrRepressedProteins(fake_significant, Sweet.lollipop.TusherAnalysis1.GoAnalysis);
List <GoTermNumber> gtn = Sweet.lollipop.TusherAnalysis1.GoAnalysis.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 = Sweet.lollipop.getInterestingFamilies(gtn, families);
Assert.AreEqual(1, fams.Count);
Assert.AreEqual(1, fams[0].theoretical_proteoforms.Count);
}
public void test_goterm_analysis_with_custom_list()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.theoretical_database.aaIsotopeMassList = new AminoAcidMasses(Sweet.lollipop.carbamidomethylation, Sweet.lollipop.neucode_labeled).AA_Masses;
Sweet.lollipop.significance_by_permutation = true;
Sweet.lollipop.significance_by_log2FC = false;
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("ASDF", "T1", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new List <ProteolysisProduct> {
new ProteolysisProduct(0, 0, "")
}, "T2", "T3", true, false, new List <DatabaseReference> {
d1
}, new List <GoTerm> {
g1
});
ProteinWithGoTerms p2 = new ProteinWithGoTerms("ASDF", "T2", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new List <ProteolysisProduct> {
new ProteolysisProduct(0, 0, "")
}, "T2", "T3", true, false, new List <DatabaseReference> {
d2
}, new List <GoTerm> {
g2
});
ProteinWithGoTerms p3 = new ProteinWithGoTerms("ASDF", "T3", new List <Tuple <string, string> > {
new Tuple <string, string>("", "")
}, new Dictionary <int, List <Modification> >(), new List <ProteolysisProduct> {
new ProteolysisProduct(0, 0, "")
}, "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.TusherValues1.significant = true;
e1.quant.tusherlogFoldChange = 1;
e2.quant.intensitySum = 1;
e2.quant.TusherValues1.significant = true;
e2.quant.tusherlogFoldChange = 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;
Sweet.lollipop.TusherAnalysis1.inducedOrRepressedProteins = Sweet.lollipop.getInducedOrRepressedProteins(new List <ExperimentalProteoform> {
e1
}, Sweet.lollipop.TusherAnalysis1.GoAnalysis);
Sweet.lollipop.TusherAnalysis1.GoAnalysis.allTheoreticalProteins = true;
Sweet.lollipop.theoretical_database.expanded_proteins = new ProteinWithGoTerms[] { p1, p2, p3 };
Sweet.lollipop.TusherAnalysis1.GoAnalysis.backgroundProteinsList = Path.Combine(TestContext.CurrentContext.TestDirectory, "test_protein_list.txt");
Sweet.lollipop.TusherAnalysis1.GoAnalysis.GO_analysis(Sweet.lollipop.TusherAnalysis1.inducedOrRepressedProteins);
Assert.AreEqual(1, Sweet.lollipop.TusherAnalysis1.inducedOrRepressedProteins.Count); // only taking one ET connection by definition in forming ET relations; only one is used in identify theoreticals
Assert.AreEqual(1, Sweet.lollipop.TusherAnalysis1.GoAnalysis.goTermNumbers.Count);
Assert.AreEqual("1", Sweet.lollipop.TusherAnalysis1.GoAnalysis.goTermNumbers.First().Id);
Assert.AreEqual(0 - (decimal)Math.Log(2d / 3d, 2), Sweet.lollipop.TusherAnalysis1.GoAnalysis.goTermNumbers.First().log_odds_ratio);
List <ProteoformFamily> fams = Sweet.lollipop.getInterestingFamilies(Sweet.lollipop.TusherAnalysis1.GoAnalysis.goTermNumbers, families);
Assert.AreEqual(1, fams.Count);
Assert.AreEqual(2, fams[0].theoretical_proteoforms.Count);
}
public void TestUnabeledProteoformCommunityRelate_ET()
{
SaveState.lollipop.neucode_labeled = false;
// 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, true, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, 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, true, 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, true, 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, true, 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, true, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(0, prList.Count);
}
public void TestNeuCodeLabeledProteoformCommunityRelate_EE()
{
SaveState.lollipop.neucode_labeled = true;
// Two proteoforms; lysine count equal; mass difference < 250 -- return 1
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true);
ExperimentalProteoform pf2 = ConstructorsForTesting.ExperimentalProteoform("A2", 1010.0, 1, true);
ExperimentalProteoform[] pa1 = new ExperimentalProteoform[2];
pa1[0] = pf1;
pa1[1] = pf2;
List <ProteoformRelation> prList = new List <ProteoformRelation>();
prList = community.relate(pa1, pa1, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(1, prList.Count);
// Two proteoforms; lysine count equal; mass difference > 250 -- return 0
pf1.modified_mass = 1000;
pf1.lysine_count = 1;
pf2.modified_mass = 2000;
pf2.lysine_count = 1;
pa1[0] = pf1;
pa1[1] = pf2;
prList = community.relate(pa1, pa1, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(0, prList.Count);
// Two proteoforms; lysine count NOT equal; mass difference < 250 -- return 0
pf1.modified_mass = 1000;
pf1.lysine_count = 1;
pf2.modified_mass = 1100;
pf2.lysine_count = 2;
pa1[0] = pf1;
pa1[1] = pf2;
prList = community.relate(pa1, pa1, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(0, prList.Count);
//Three proteoforms; lysine count equal; mass difference < 250 Da -- return 3
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true);
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("A2", 1010.0, 1, true);
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("A3", 1020.0, 1, true);
ExperimentalProteoform[] pa2 = new ExperimentalProteoform[3];
pa2[0] = pf3;
pa2[1] = pf4;
pa2[2] = pf5;
prList = community.relate(pa2, pa2, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(3, prList.Count);
//Three proteoforms; lysine count equal; one mass difference < 250 Da; one mass difference > 500 -- return 1
pf3.modified_mass = 1000;
pf3.lysine_count = 1;
pf4.modified_mass = 1010;
pf4.lysine_count = 1;
pf5.modified_mass = 2020;
pf5.lysine_count = 1;
pa2[0] = pf3;
pa2[1] = pf4;
pa2[2] = pf5;
prList = community.relate(pa2, pa2, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(1, prList.Count);
//Three proteoforms; lysine count NOT equal; mass difference < 250 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;
pa2[0] = pf3;
pa2[1] = pf4;
pa2[2] = pf5;
prList = community.relate(pa2, pa2, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(0, prList.Count);
//Three proteoforms; lysine count equal; mass difference > 250 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;
pa2[0] = pf3;
pa2[1] = pf4;
pa2[2] = pf5;
prList = community.relate(pa2, pa2, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
Assert.AreEqual(0, prList.Count);
}
public void shift_et_peak_neucode()
{
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(100);
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, true);
Assert.IsTrue(pf3.mass_shifted);
Assert.IsTrue(pf4.mass_shifted);
foreach (Component c in
n3.
Concat(n4).
Concat(n3.Select(n => ((NeuCodePair)n).neuCodeLight)).
Concat(n4.Select(n => ((NeuCodePair)n).neuCodeLight)))
{
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);
}
foreach (Component c in
n3.Select(n => ((NeuCodePair)n).neuCodeHeavy).
Concat(n4.Select(n => ((NeuCodePair)n).neuCodeHeavy)))
{
Assert.AreEqual(-1.0 * Lollipop.MONOISOTOPIC_UNIT_MASS, c.manual_mass_shift);
Assert.AreEqual(200 + TestExperimentalProteoform.starter_lysine_count * Lollipop.NEUCODE_LYSINE_MASS_SHIFT - 1.0 * Lollipop.MONOISOTOPIC_UNIT_MASS, c.weighted_monoisotopic_mass);
}
}
public void test_accept_from_presets()
{
Sweet.lollipop = new Lollipop();
ProteoformCommunity test_community = new ProteoformCommunity();
Sweet.lollipop.target_proteoform_community = test_community;
Sweet.lollipop.theoretical_database.uniprotModifications = new Dictionary <string, List <Modification> >
{
{ "unmodified", new List <Modification>()
{
ConstructorsForTesting.get_modWithMass("unmodified", 0)
} }
};
//Testing the acceptance of peaks. The FDR is tested above, so I'm not going to work with that here.
//Four proteoforms, three relations (linear), middle one isn't accepted; should give 2 families
Sweet.lollipop.min_peak_count_ee = 2;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E1");
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E2");
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("E3");
ExperimentalProteoform pf6 = ConstructorsForTesting.ExperimentalProteoform("E4");
ProteoformComparison comparison34 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison45 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison56 = ProteoformComparison.ExperimentalExperimental;
ProteoformRelation pr2 = new ProteoformRelation(pf3, pf4, comparison34, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr3 = new ProteoformRelation(pf4, pf5, comparison45, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr4 = new ProteoformRelation(pf5, pf6, comparison56, 0, TestContext.CurrentContext.TestDirectory);
//Test display strings
Assert.AreEqual("E1", pr2.connected_proteoforms[0].accession);
Assert.AreEqual("E2", pr2.connected_proteoforms[1].accession);
List <ProteoformRelation> prs2 = new List <ProteoformRelation> {
pr2, pr3, pr4
};
foreach (ProteoformRelation pr in prs2)
{
pr.set_nearby_group(prs2, prs2.Select(r => r.InstanceId).ToList());
}
Assert.AreEqual(3, pr2.nearby_relations_count);
Assert.AreEqual(3, pr3.nearby_relations_count);
Assert.AreEqual(3, pr4.nearby_relations_count);
Sweet.lollipop.theoretical_database.all_possible_ptmsets = new List <PtmSet> {
new PtmSet(new List <Ptm> {
new Ptm(-1, ConstructorsForTesting.get_modWithMass("unmodified", 0))
})
};
Sweet.lollipop.theoretical_database.possible_ptmset_dictionary = Sweet.lollipop.theoretical_database.make_ptmset_dictionary();
Sweet.unaccept_peak_action(pr2);
using (StreamWriter file = new StreamWriter(Path.Combine(TestContext.CurrentContext.TestDirectory, "method.xml")))
file.WriteLine(Sweet.save_method());
Sweet.open_method(Path.Combine(TestContext.CurrentContext.TestDirectory, "method.xml"), string.Join(Environment.NewLine, File.ReadAllLines(Path.Combine(TestContext.CurrentContext.TestDirectory, "method.xml"))), true, out string warning);
Sweet.lollipop.ee_peaks = test_community.accept_deltaMass_peaks(prs2, new List <ProteoformRelation>());
Assert.AreEqual(1, Sweet.lollipop.ee_peaks.Count);
DeltaMassPeak peak = Sweet.lollipop.ee_peaks[0];
Assert.IsFalse(peak.Accepted); // <-- even though it's above the threshold
Assert.AreEqual(3, peak.grouped_relations.Count);
Assert.AreEqual(3, pr2.peak.peak_relation_group_count);
Assert.AreEqual(0, pr2.peak.DeltaMass);
Assert.AreEqual("[unmodified]", peak.possiblePeakAssignments_string);
//Test that the relations in the peak are added to each of the proteoforms referenced in the peak
Assert.True(pf3.relationships.Contains(pr2));
Assert.True(pf4.relationships.Contains(pr2) && pf4.relationships.Contains(pr3));
Assert.True(pf5.relationships.Contains(pr3) && pf5.relationships.Contains(pr4));
}
public static void accept_peaks_doesnt_crash_with_weird_relation()
{
ProteoformCommunity c = new ProteoformCommunity();
ProteoformRelation r = new ProteoformRelation(ConstructorsForTesting.ExperimentalProteoform("E1"), ConstructorsForTesting.ExperimentalProteoform("E1"), ProteoformComparison.ExperimentalFalse, 0, TestContext.CurrentContext.TestDirectory);
r.outside_no_mans_land = true;
r.nearby_relations = new List <ProteoformRelation>();
Assert.Throws <ArgumentException>(() => c.accept_deltaMass_peaks(new List <ProteoformRelation> {
r
}, new List <ProteoformRelation>()));
}
public void shift_et_peak_neucode_from_actions()
{
Sweet.lollipop = new Lollipop();
ProteoformCommunity test_community = new ProteoformCommunity();
Sweet.lollipop.target_proteoform_community = test_community;
//Make a few experimental proteoforms
List <IAggregatable> n1 = TestExperimentalProteoform.generate_neucode_components(100);
List <IAggregatable> n2 = TestExperimentalProteoform.generate_neucode_components(100);
List <IAggregatable> n3 = TestExperimentalProteoform.generate_neucode_components(200);
List <IAggregatable> n4 = TestExperimentalProteoform.generate_neucode_components(200);
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("E1");
pf1.aggregated = n1;
ExperimentalProteoform pf2 = ConstructorsForTesting.ExperimentalProteoform("E2");
pf2.aggregated = n2;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E3");
pf3.aggregated = n3;
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E4");
pf4.aggregated = n4;
Sweet.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, Sweet.lollipop.et_peaks.Count);
//Shift the peaks, which shifts all of the proteoforms
DeltaMassPeak d2 = Sweet.lollipop.et_peaks[1];
d2.mass_shifter = "-1";
Sweet.shift_peak_action(d2);
d2.mass_shifter = null;
using (StreamWriter file = new StreamWriter(Path.Combine(TestContext.CurrentContext.TestDirectory, "method.xml")))
file.WriteLine(Sweet.save_method());
Sweet.open_method(Path.Combine(TestContext.CurrentContext.TestDirectory, "method.xml"), string.Join(Environment.NewLine, File.ReadAllLines(Path.Combine(TestContext.CurrentContext.TestDirectory, "method.xml"))), true, out string warning);
Sweet.mass_shifts_from_presets();
d2.shift_experimental_masses(Convert.ToInt32(d2.mass_shifter), true);
foreach (Component c in
n3.OfType <NeuCodePair>().Select(n => n.neuCodeLight).
Concat(n4.OfType <NeuCodePair>().Select(n => n.neuCodeLight)))
{
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);
}
foreach (Component c in
n3.OfType <NeuCodePair>().Select(n => n.neuCodeHeavy).
Concat(n4.OfType <NeuCodePair>().Select(n => n.neuCodeHeavy)))
{
Assert.AreEqual(-1.0 * Lollipop.MONOISOTOPIC_UNIT_MASS, c.manual_mass_shift);
Assert.AreEqual(200 + TestExperimentalProteoform.starter_lysine_count * Lollipop.NEUCODE_LYSINE_MASS_SHIFT - 1.0 * Lollipop.MONOISOTOPIC_UNIT_MASS, c.weighted_monoisotopic_mass);
}
}
public void TestAcceptDeltaMassPeaks()
{
Sweet.lollipop = new Lollipop();
ProteoformCommunity test_community = new ProteoformCommunity();
Sweet.lollipop.target_proteoform_community = test_community;
Sweet.lollipop.theoretical_database.uniprotModifications = new Dictionary <string, List <Modification> >
{
{ "unmodified", new List <Modification>()
{
ConstructorsForTesting.get_modWithMass("unmodified", 0)
} }
};
//Testing the acceptance of peaks. The FDR is tested above, so I'm not going to work with that here.
//Four proteoforms, three relations (linear), middle one isn't accepted; should give 2 families
Sweet.lollipop.min_peak_count_ee = 2;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E1");
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E2");
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("E3");
ExperimentalProteoform pf6 = ConstructorsForTesting.ExperimentalProteoform("E4");
ProteoformComparison comparison34 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison45 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison56 = ProteoformComparison.ExperimentalExperimental;
ProteoformRelation pr2 = new ProteoformRelation(pf3, pf4, comparison34, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr3 = new ProteoformRelation(pf4, pf5, comparison45, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr4 = new ProteoformRelation(pf5, pf6, comparison56, 0, TestContext.CurrentContext.TestDirectory);
//Test display strings
Assert.AreEqual("E1", pr2.connected_proteoforms[0].accession);
Assert.AreEqual("E2", pr2.connected_proteoforms[1].accession);
pr2.RelationType = ProteoformComparison.ExperimentalExperimental;
pr2.RelationType = ProteoformComparison.ExperimentalTheoretical;
pr2.RelationType = ProteoformComparison.ExperimentalDecoy;
pr2.RelationType = ProteoformComparison.ExperimentalFalse;
pr2.RelationType = comparison34;
List <ProteoformRelation> prs2 = new List <ProteoformRelation> {
pr2, pr3, pr4
};
foreach (ProteoformRelation pr in prs2)
{
pr.set_nearby_group(prs2, prs2.Select(r => r.InstanceId).ToList());
}
Assert.AreEqual(3, pr2.nearby_relations_count);
Assert.AreEqual(3, pr3.nearby_relations_count);
Assert.AreEqual(3, pr4.nearby_relations_count);
Sweet.lollipop.theoretical_database.all_possible_ptmsets = new List <PtmSet> {
new PtmSet(new List <Ptm> {
new Ptm(-1, ConstructorsForTesting.get_modWithMass("unmodified", 0))
})
};
Sweet.lollipop.theoretical_database.possible_ptmset_dictionary = Sweet.lollipop.theoretical_database.make_ptmset_dictionary();
//auto accept set to false
Sweet.lollipop.ee_accept_peaks_based_on_rank = false;
Sweet.lollipop.ee_peaks = test_community.accept_deltaMass_peaks(prs2, new List <ProteoformRelation>());
Assert.AreEqual(1, Sweet.lollipop.ee_peaks.Count);
DeltaMassPeak peak = Sweet.lollipop.ee_peaks[0];
Assert.IsTrue(peak.Accepted);
Assert.AreEqual(3, peak.grouped_relations.Count);
Assert.AreEqual(3, pr2.peak.peak_relation_group_count);
Assert.AreEqual(0, pr2.peak.DeltaMass);
Assert.AreEqual("[unmodified]", peak.possiblePeakAssignments_string);
//Test that the relations in the peak are added to each of the proteoforms referenced in the peak
Assert.True(pf3.relationships.Contains(pr2));
Assert.True(pf4.relationships.Contains(pr2) && pf4.relationships.Contains(pr3));
Assert.True(pf5.relationships.Contains(pr3) && pf5.relationships.Contains(pr4));
//autoaccept set to true, must be less than first quartile rank...
Sweet.lollipop.clear_ee();
Sweet.lollipop.mod_rank_first_quartile = 0;
Sweet.lollipop.ee_accept_peaks_based_on_rank = true;
Sweet.lollipop.ee_peaks = test_community.accept_deltaMass_peaks(prs2, new List <ProteoformRelation>());
peak = Sweet.lollipop.ee_peaks[0];
Assert.IsFalse(peak.Accepted);
Assert.AreEqual(0, peak.possiblePeakAssignments.Min(p => p.ptm_rank_sum));
Sweet.lollipop.clear_ee();
Sweet.lollipop.mod_rank_first_quartile = 1;
Sweet.lollipop.ee_accept_peaks_based_on_rank = true;
Sweet.lollipop.ee_peaks = test_community.accept_deltaMass_peaks(prs2, new List <ProteoformRelation>());
peak = Sweet.lollipop.ee_peaks[0];
Assert.IsTrue(peak.Accepted);
Assert.AreEqual(0, peak.possiblePeakAssignments.Min(p => p.ptm_rank_sum));
}
public void test_construct_two_families()
{
//Five experimental proteoforms, four relations (linear), second on not accepted into a peak, one peak; should give 2 families
SaveState.lollipop = new Lollipop();
ProteoformCommunity test_community = new ProteoformCommunity();
SaveState.lollipop.target_proteoform_community = test_community;
SaveState.lollipop.theoretical_database.uniprotModifications = new Dictionary <string, List <Modification> > {
{ "unmodified", new List <Modification> {
new Modification("unmodified", "unknown")
} }
};
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....
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E1");
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E2");
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("E3");
ExperimentalProteoform pf6 = ConstructorsForTesting.ExperimentalProteoform("E4");
ExperimentalProteoform pf7 = ConstructorsForTesting.ExperimentalProteoform("E5");
ProteoformComparison comparison34 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison45 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison56 = ProteoformComparison.ExperimentalExperimental;
ProteoformComparison comparison67 = ProteoformComparison.ExperimentalExperimental;
ProteoformRelation pr2 = new ProteoformRelation(pf3, pf4, comparison34, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr3 = new ProteoformRelation(pf4, pf5, comparison45, 19, TestContext.CurrentContext.TestDirectory); //not accepted
ProteoformRelation pr4 = new ProteoformRelation(pf5, pf6, comparison56, 0, TestContext.CurrentContext.TestDirectory);
ProteoformRelation pr5 = new ProteoformRelation(pf6, pf7, comparison67, 0, TestContext.CurrentContext.TestDirectory);
List <ProteoformRelation> prs2 = new List <ProteoformRelation> {
pr2, pr3, pr4, pr5
}.OrderBy(r => r.DeltaMass).ToList();
foreach (ProteoformRelation pr in prs2)
{
pr.set_nearby_group(prs2, prs2.Select(r => r.InstanceId).ToList());
}
Assert.AreEqual(3, pr2.nearby_relations_count);
Assert.AreEqual(1, pr3.nearby_relations_count);
Assert.AreEqual(3, pr4.nearby_relations_count);
Assert.AreEqual(3, pr5.nearby_relations_count);
test_community.accept_deltaMass_peaks(prs2, new List <ProteoformRelation>());
Assert.AreEqual(2, SaveState.lollipop.ee_peaks.Count);
Assert.AreEqual(1, SaveState.lollipop.ee_peaks.Where(peak => peak.Accepted).Count());
Assert.AreEqual(3, SaveState.lollipop.ee_peaks.Where(peak => peak.Accepted).First().grouped_relations.Count());
test_community.experimental_proteoforms = new ExperimentalProteoform[] { pf3, pf4, pf5, pf6, pf7 };
test_community.construct_families();
Assert.AreEqual(2, test_community.families.Count);
Assert.AreEqual("", test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf3)).accession_list);
Assert.AreEqual(2, test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf3)).proteoforms.Count);
Assert.AreEqual(2, test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf3)).experimental_proteoforms.Count);
Assert.AreEqual(0, test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf3)).theoretical_proteoforms.Count);
Assert.AreEqual("", test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf5)).accession_list);
Assert.AreEqual(3, test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf5)).proteoforms.Count);
Assert.AreEqual(3, test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf5)).experimental_proteoforms.Count);
Assert.AreEqual(0, test_community.families.FirstOrDefault(f => f.proteoforms.Contains(pf5)).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 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 void agg_proteoforms_meeting_criteria2()
{
Sweet.lollipop = new Lollipop();
string anysingle = "Biorep+Techreps From Any Single Condition";
string any = "Biorep+Techreps From Any Condition";
string fromeach = "Biorep+Techreps From Each Condition";
List <ExperimentalProteoform> exps = new List <ExperimentalProteoform> {
ConstructorsForTesting.ExperimentalProteoform("E")
};
List <string> conditions = new List <string> {
"s", "ns"
};
Component c1 = new Component();
c1.input_file = new InputFile("somePath", Purpose.Identification);
c1.input_file.lt_condition = conditions[0];
c1.input_file.biological_replicate = "1";
c1.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c1);
List <ExperimentalProteoform> exps_out = new List <ExperimentalProteoform>();
//PASSES WHEN THERE ARE ENOUGH IN SPECIFIED CONDITIONS
//One biorep+techrep obs passes any-single-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 1);
Assert.AreEqual(1, exps_out.Count);
//Two biorep+techrep obs in single condition passes any-single-conditon test
Component c2 = new Component();
c2.input_file = new InputFile("somePath", Purpose.Identification);
c2.input_file.lt_condition = conditions[0];
c2.input_file.biological_replicate = "1";
c2.input_file.technical_replicate = "2";
exps[0].aggregated.Add(c2);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 2);
Assert.AreEqual(1, exps_out.Count);
//Two biorep+techrep obs in single condition passes any-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, any, 2);
Assert.AreEqual(1, exps_out.Count);
//Two biorep+techrep obs in single condition doesn't pass for-each-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 1);
Assert.AreEqual(0, exps_out.Count);
// DOESN'T PASS WHEN LESS THAN THRESHOLD
//Two biorep+techrep obs in single condition doesn't pass 3 from any-single-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 3);
Assert.AreEqual(0, exps_out.Count);
//Two biorep+techrep obs in single condition doesn't pass 3 from any-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, any, 3);
Assert.AreEqual(0, exps_out.Count);
//Two biorep+techrep obs in single condition doesn't pass 2 from for-each-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 2);
Assert.AreEqual(0, exps_out.Count);
Component c3 = new Component();
c3.input_file = new InputFile("somePath", Purpose.Identification);
c3.input_file.lt_condition = conditions[1];
c3.input_file.biological_replicate = "100";
c3.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c3);
//At least one biorep+techrep in each condition passes for-each-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 1);
Assert.AreEqual(1, exps_out.Count);
// DOESN'T PASS WHEN LESS THAN THRESHOLD IN SPECIFIC CONDITIONS UNLESS ANY-CONDITION
//Two and one biorep+techrep obs in two different conditions doesn't pass 3 from any-single-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 3);
Assert.AreEqual(0, exps_out.Count);
//Two and one biorep+techrep obs in two different conditions passes 3 from any-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, any, 3);
Assert.AreEqual(1, exps_out.Count);
//Two and one biorep+techrep obs in two different conditions doesn't pass 3 from for-each-conditon test
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 3);
Assert.AreEqual(0, exps_out.Count);
//DOESN'T PASS WHEN NOT MATCHING LISTED CONDITIONS, EXCEPT FOR ANY-CONDITION
foreach (var x in exps[0].aggregated)
{
x.input_file.lt_condition = "not_a_condition";
}
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 3);
Assert.AreEqual(0, exps_out.Count);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, any, 3);
Assert.AreEqual(1, exps_out.Count);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 3);
Assert.AreEqual(0, exps_out.Count);
//NOT JUST COUNTING BIOREP INTENSITIES, BUT RATHER BIOREPS WITH OBSERVATIONS
exps[0].aggregated.Clear();
Component c4 = new Component();
c4.input_file = new InputFile("somePath", Purpose.Identification);
c4.input_file.lt_condition = conditions[0];
c4.input_file.biological_replicate = "1";
c4.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c4);
Component c5 = new Component();
c5.input_file = new InputFile("somePath", Purpose.Identification);
c5.input_file.lt_condition = conditions[0];
c5.input_file.biological_replicate = "1";
c5.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c5);
Component c6 = new Component();
c6.input_file = new InputFile("somePath", Purpose.Identification);
c6.input_file.lt_condition = conditions[0];
c6.input_file.biological_replicate = "1";
c6.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c6);
Component c7 = new Component();
c7.input_file = new InputFile("somePath", Purpose.Identification);
c7.input_file.lt_condition = conditions[0];
c7.input_file.biological_replicate = "1";
c7.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c7);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 2);
Assert.AreEqual(0, exps_out.Count);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, any, 2);
Assert.AreEqual(0, exps_out.Count);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 2);
Assert.AreEqual(0, exps_out.Count);
Component c8 = new Component();
c8.input_file = new InputFile("somePath", Purpose.Identification);
c8.input_file.lt_condition = conditions[1];
c8.input_file.biological_replicate = "1";
c8.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c8);
Component c9 = new Component();
c9.input_file = new InputFile("somePath", Purpose.Identification);
c9.input_file.lt_condition = conditions[1];
c9.input_file.biological_replicate = "1";
c9.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c9);
Component c10 = new Component();
c10.input_file = new InputFile("somePath", Purpose.Identification);
c10.input_file.lt_condition = conditions[1];
c10.input_file.biological_replicate = "1";
c10.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c10);
Component c11 = new Component();
c11.input_file = new InputFile("somePath", Purpose.Identification);
c11.input_file.lt_condition = conditions[1];
c11.input_file.biological_replicate = "1";
c11.input_file.technical_replicate = "1";
exps[0].aggregated.Add(c11);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, anysingle, 2);
Assert.AreEqual(0, exps_out.Count);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, any, 2);
Assert.AreEqual(1, exps_out.Count);
exps_out = Sweet.lollipop.determineAggProteoformsMeetingCriteria(conditions, exps, fromeach, 2);
Assert.AreEqual(0, exps_out.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 TestDeltaMassPeakConstructor()
{
Sweet.lollipop = new Lollipop();
ConstructorsForTesting.read_mods();
Sweet.lollipop.et_high_mass_difference = 250;
Sweet.lollipop.et_low_mass_difference = -250;
Sweet.lollipop.peak_width_base_ee = 0.015;
Sweet.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, new HashSet <ProteoformRelation>(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 TestUnabeledProteoformCommunityRelate_EF()
{
ProteoformCommunity test_community;
List <ProteoformRelation> unequal_relations;
//Two equal, two unequal lysine count. Each should create two unequal relations, so eight relations total
//However, it shouldn't compare to itself, so that would make 4 total relations
test_community = new ProteoformCommunity();
Sweet.lollipop.neucode_labeled = true;
test_community.experimental_proteoforms = new ExperimentalProteoform[] {
ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A2", 1000.0, 5, true),
ConstructorsForTesting.ExperimentalProteoform("A3", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A4", 1000.0, 2, true)
};
Sweet.lollipop.ee_relations = test_community.relate(test_community.experimental_proteoforms, test_community.experimental_proteoforms, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
unequal_relations = test_community.relate_ef(test_community.experimental_proteoforms, test_community.experimental_proteoforms);
Assert.AreNotEqual(test_community.experimental_proteoforms[0], test_community.experimental_proteoforms[2]);
Assert.False(test_community.allowed_relation(test_community.experimental_proteoforms[0], test_community.experimental_proteoforms[0], ProteoformComparison.ExperimentalExperimental));
Assert.AreNotEqual(test_community.experimental_proteoforms[0].lysine_count, test_community.experimental_proteoforms[1].lysine_count);
//Assert.False(test_community.allowed_relation(test_community.experimental_proteoforms[0], test_community.experimental_proteoforms[1], ProteoformComparison.ExperimentalExperimental)); // this is taken care of in relate, now
Assert.True(test_community.allowed_relation(test_community.experimental_proteoforms[0], test_community.experimental_proteoforms[2], ProteoformComparison.ExperimentalExperimental));
//Assert.False(test_community.allowed_relation(test_community.experimental_proteoforms[0], test_community.experimental_proteoforms[3], ProteoformComparison.ExperimentalExperimental)); // this is taken care of in relate, now
Assert.AreEqual(2, unequal_relations.Count); //only 2 relations are > 3 lysines apart
//Two equal, two unequal lysine count. But one each has mass_difference > 250, so no relations
test_community = new ProteoformCommunity();
Sweet.lollipop.ee_relations.Clear();
test_community.experimental_proteoforms = new ExperimentalProteoform[] {
ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A2", 2000, 2, true),
ConstructorsForTesting.ExperimentalProteoform("A3", 3000, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A4", 4000, 2, true)
};
Sweet.lollipop.ee_relations = test_community.relate(test_community.experimental_proteoforms, test_community.experimental_proteoforms, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
unequal_relations = test_community.relate_ef(test_community.experimental_proteoforms, test_community.experimental_proteoforms);
Assert.AreEqual(0, unequal_relations.Count);
//None equal lysine count (apart from itself), four unequal lysine count. Each should create no unequal relations, so no relations total
test_community = new ProteoformCommunity();
test_community.experimental_proteoforms = new ExperimentalProteoform[] {
ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A2", 1000.0, 2, true),
ConstructorsForTesting.ExperimentalProteoform("A3", 1000.0, 3, true),
ConstructorsForTesting.ExperimentalProteoform("A4", 1000.0, 4, true)
};
Sweet.lollipop.ee_relations = test_community.relate(test_community.experimental_proteoforms, test_community.experimental_proteoforms, ProteoformComparison.ExperimentalExperimental, true, TestContext.CurrentContext.TestDirectory, false);
unequal_relations = test_community.relate_ef(test_community.experimental_proteoforms, test_community.experimental_proteoforms);
Assert.AreEqual(0, unequal_relations.Count);
//All equal, no unequal lysine count because there's an empty list of unequal lysine-count proteoforms. Each should create no unequal relations, so no relations total
test_community = new ProteoformCommunity();
test_community.experimental_proteoforms = new ExperimentalProteoform[] {
ConstructorsForTesting.ExperimentalProteoform("A1", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A2", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A3", 1000.0, 1, true),
ConstructorsForTesting.ExperimentalProteoform("A4", 1000.0, 1, true)
};
unequal_relations = test_community.relate_ef(test_community.experimental_proteoforms, test_community.experimental_proteoforms);
Assert.AreEqual(0, unequal_relations.Count);
}
public void shift_et_peak_unlabeled()
{
Sweet.lollipop = new Lollipop();
ProteoformCommunity test_community = new ProteoformCommunity();
Sweet.lollipop.target_proteoform_community = test_community;
Sweet.lollipop.neucode_labeled = false;
//Make a few experimental proteoforms
List <IAggregatable> n1 = TestExperimentalProteoform.generate_neucode_components(100);
List <IAggregatable> n2 = TestExperimentalProteoform.generate_neucode_components(200);
List <IAggregatable> n3 = TestExperimentalProteoform.generate_neucode_components(200);
List <IAggregatable> n4 = TestExperimentalProteoform.generate_neucode_components(200);
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("E1");
pf1.aggregated = n1.Select(n => (n as NeuCodePair).neuCodeLight).ToList <IAggregatable>();
ExperimentalProteoform pf2 = ConstructorsForTesting.ExperimentalProteoform("E2");
pf2.aggregated = n2.Select(n => (n as NeuCodePair).neuCodeLight).ToList <IAggregatable>();
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("E3");
pf3.aggregated = n3.Select(n => (n as NeuCodePair).neuCodeLight).ToList <IAggregatable>();
ExperimentalProteoform pf4 = ConstructorsForTesting.ExperimentalProteoform("E4");
pf4.aggregated = n4.Select(n => (n as NeuCodePair).neuCodeLight).ToList <IAggregatable>();
Sweet.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, Sweet.lollipop.et_peaks.Count);
//Shift the peaks, which shifts all of the proteoforms
DeltaMassPeak d2 = Sweet.lollipop.et_peaks[1];
d2.shift_experimental_masses(-1, false);
foreach (Component c in pf3.aggregated.Concat(pf4.aggregated).OfType <Component>())
{
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);
}
Sweet.lollipop.clear_et();
Assert.AreEqual(0, Sweet.lollipop.et_peaks.Count);
//don't double shift E if it's in two peaks...
pr1 = new ProteoformRelation(pf1, pf5, comparison14, 1, TestContext.CurrentContext.TestDirectory);
pr2 = new ProteoformRelation(pf1, pf6, comparison14, 1, TestContext.CurrentContext.TestDirectory);
prs = new List <ProteoformRelation> {
pr1, pr2
};
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(1, Sweet.lollipop.et_peaks.Count);
//Shift the peaks, which shifts all of the proteoforms
d2 = Sweet.lollipop.et_peaks[0];
d2.shift_experimental_masses(-1, false);
foreach (Component c in pf3.aggregated.Concat(pf4.aggregated).OfType <Component>())
{
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 TestUnabeledProteoformCommunityRelateWithNotches_ET()
{
Sweet.lollipop.neucode_labeled = false;
Sweet.lollipop.et_use_notch = true;
// 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, true, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, 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, true, 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, true, 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, true, 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, true, 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, true, TestContext.CurrentContext.TestDirectory, true);
Assert.AreEqual(1, prList.Count);
}
public void TestDeltaMassPeakConstructorWithNotches()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.et_use_notch = true;
Sweet.lollipop.enter_input_files(new string[] { Path.Combine(TestContext.CurrentContext.TestDirectory, "ptmlist.txt") }, Lollipop.acceptable_extensions[2], Lollipop.file_types[2], Sweet.lollipop.input_files, false);
ConstructorsForTesting.read_mods();
Sweet.lollipop.et_high_mass_difference = 250;
Sweet.lollipop.et_low_mass_difference = -250;
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("acession1");
TheoreticalProteoform pf2 = ConstructorsForTesting.make_a_theoretical();
pf1.modified_mass = 1000;
ProteoformComparison relation_type = ProteoformComparison.ExperimentalTheoretical;
double delta_mass = 1 - 1e-7;
ExperimentalProteoform pf3 = ConstructorsForTesting.ExperimentalProteoform("acession3");
TheoreticalProteoform pf4 = ConstructorsForTesting.make_a_theoretical();
pf3.modified_mass = 1000;
ProteoformComparison relation_type2 = ProteoformComparison.ExperimentalTheoretical;
double delta_mass2 = 1;
ExperimentalProteoform pf5 = ConstructorsForTesting.ExperimentalProteoform("acession5");
TheoreticalProteoform pf6 = ConstructorsForTesting.make_a_theoretical();
pf5.modified_mass = 1000;
ProteoformComparison relation_type3 = ProteoformComparison.ExperimentalTheoretical;
double delta_mass3 = 1 + 1e-7;
ExperimentalProteoform pf55 = ConstructorsForTesting.ExperimentalProteoform("acession5");
TheoreticalProteoform pf65 = ConstructorsForTesting.make_a_theoretical();
pf55.modified_mass = 1000;
ProteoformComparison relation_type35 = ProteoformComparison.ExperimentalTheoretical;
double delta_mass35 = 1 + 2e-7;
TestProteoformCommunityRelate.prepare_for_et(new List <double>()
{
1
});
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, new HashSet <ProteoformRelation>(theList));
Console.WriteLine("Created deltaMassPeak");
Assert.AreEqual(0, deltaMassPeak.peak_group_fdr);
Assert.AreEqual(4, deltaMassPeak.grouped_relations.Count);
Assert.AreEqual("[fake1]", deltaMassPeak.possiblePeakAssignments_string);
Assert.AreEqual(1.0, deltaMassPeak.DeltaMass);
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();
pf7.modified_mass = 1000;
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 TestProteoformCommunityRelateWithNotches_ED()
{
Sweet.lollipop = new Lollipop();
Sweet.lollipop.neucode_labeled = true;
Sweet.lollipop.et_use_notch = true;
Sweet.lollipop.decoy_databases = 1;
// In empty comminity, relate ed is empty
Assert.AreEqual(0, Sweet.lollipop.ed_relations.Count);
//create a decoy proteoform community
Sweet.lollipop.decoy_proteoform_communities.Add(Sweet.lollipop.decoy_community_name_prefix + "0", new ProteoformCommunity());
TheoreticalProteoform pf2 = ConstructorsForTesting.make_a_theoretical("decoyProteoform1", 0, -1);
Sweet.lollipop.decoy_proteoform_communities[Sweet.lollipop.decoy_community_name_prefix + "0"].theoretical_proteoforms = new TheoreticalProteoform[1] {
pf2
};
Sweet.lollipop.relate_ed();
// Have a single decoy community --> have single ed_relations
Assert.AreEqual(1, Sweet.lollipop.ed_relations.Count);
// But it's empty
Assert.IsEmpty(Sweet.lollipop.ed_relations[Sweet.lollipop.decoy_community_name_prefix + "0"]);
// In order to make it not empty, we must have relate_et method output a non-empty List
// it must take as arguments non-empty pfs1 and pfs2
// So testProteoformCommunity.experimental_proteoforms must be non-empty
// And decoy_proteoforms["fake_decoy_proteoform1"] must be non-empty
ExperimentalProteoform pf1 = ConstructorsForTesting.ExperimentalProteoform("experimentalProteoform1");
Sweet.lollipop.decoy_proteoform_communities[Sweet.lollipop.decoy_community_name_prefix + "0"].theoretical_proteoforms.First().ExpandedProteinList = new List <ProteinWithGoTerms> {
p1
};
Assert.IsEmpty(Sweet.lollipop.decoy_proteoform_communities[Sweet.lollipop.decoy_community_name_prefix + "0"].experimental_proteoforms);
Sweet.lollipop.decoy_proteoform_communities[Sweet.lollipop.decoy_community_name_prefix + "0"].experimental_proteoforms = new ExperimentalProteoform[] { pf1 };
pf1.modified_mass = 1000;
pf2.modified_mass = 1000;
Sweet.lollipop.clear_et();
prepare_for_et(new List <double> {
pf1.modified_mass - pf2.modified_mass
});
Sweet.lollipop.relate_ed();
// Make sure there is one relation total, because only a single decoy was provided
Assert.AreEqual(1, Sweet.lollipop.ed_relations.Count);
Assert.IsNotEmpty(Sweet.lollipop.ed_relations);
Assert.AreEqual(1, Sweet.lollipop.ed_relations[Sweet.lollipop.decoy_community_name_prefix + "0"].Count); // Make sure there is one relation for the provided fake_decoy_proteoform1
ProteoformRelation rel = Sweet.lollipop.ed_relations[Sweet.lollipop.decoy_community_name_prefix + "0"][0];
Assert.IsFalse(rel.Accepted);
Assert.AreEqual("decoyProteoform1", rel.connected_proteoforms[1].accession);
Assert.AreEqual(0, rel.DeltaMass);
Assert.IsEmpty(((TheoreticalProteoform)rel.connected_proteoforms[1]).fragment);
Assert.AreEqual(1, rel.nearby_relations_count); //shows that calculate_unadjusted_group_count works
//Assert.AreEqual(1, rel.mass_difference_group.Count); //I don't think we need this test anymore w/ way peaks are made -LVS
Assert.AreEqual(-1, rel.lysine_count);
Assert.AreEqual("T2", ((TheoreticalProteoform)rel.connected_proteoforms[1]).name);
Assert.AreEqual(0, ((ExperimentalProteoform)rel.connected_proteoforms[0]).aggregated.Count); //nothing aggregated with the basic constructor
Assert.IsTrue(rel.outside_no_mans_land);
Assert.IsNull(rel.peak);
Assert.True(string.Equals("unmodified", ((TheoreticalProteoform)rel.connected_proteoforms[1]).ptm_set.ptm_description, StringComparison.CurrentCultureIgnoreCase));
Assert.AreEqual(1, rel.nearby_relations_count);
}
