//!!!very similar to other code
public static Dictionary <Pair <NEC, Hla>, bool> ReadTable(HlaFactory hlaFactory, string fileName, bool dedup)
{
Dictionary <Pair <NEC, Hla>, bool> table = new Dictionary <Pair <NEC, Hla>, bool>();
foreach (Dictionary <string, string> row in SpecialFunctions.TabFileTable(fileName, "N\tepitope\tC\thla\tlabel", false))
{
string n = row["N"];
string epitope = row["epitope"];
SpecialFunctions.CheckCondition(Biology.GetInstance().LegalPeptide(epitope), string.Format("Peptide, '{0}', contains illegal char.", epitope));
string c = row["C"];
NEC nec = NEC.GetInstance(n, epitope, c);
Hla hla = hlaFactory.GetGroundInstance(row["hla"]);
string labelString = row["label"];
SpecialFunctions.CheckCondition(labelString == "0" || labelString == "1", "Expect label to be '0' or '1'");
Pair <NEC, Hla> pair = new Pair <NEC, Hla>(nec, hla);
bool labelAsBool = (labelString == "1");
if (dedup && table.ContainsKey(pair))
{
SpecialFunctions.CheckCondition(table[pair] == labelAsBool, "The example " + pair.ToString() + " appears with contradictory labels.");
continue;
}
table.Add(pair, labelAsBool);
}
return(table);
}
//!!!very similar to other code
public static Dictionary <string, int> ReadTable(string fileName)
{
Dictionary <string, int> table = new Dictionary <string, int>();
foreach (Dictionary <string, string> row in SpecialFunctions.TabFileTable(fileName, "epitope\tdifferenceCount\tid\tstartAABase1", false))
{
string epitope = row["epitope"];
SpecialFunctions.CheckCondition(Biology.GetInstance().LegalPeptide(epitope), string.Format("Peptide, '{0}', contains illegal char.", epitope));
int differenceCount = int.Parse(row["differenceCount"]);
MaxMaxMismatches = Math.Max(MaxMaxMismatches, differenceCount);
table.Add(epitope, differenceCount);
}
return(table);
}
//static private Dictionary<Pair<NEC, Hla>, bool> CloseHuman = null;
//GeneratorType.Hla | GeneratorType.Position | GeneratorType.Property | GeneratorType.AndHla | GeneratorType.Zero6Supertype | GeneratorType.AndZero6Supertype
private static Set <IHashableFeature> GenerateFeatureSet(
object entity, string supertypeTableSource,
int?flankSizeOrNull,
bool includeFlankNECFeatures,
bool includeChemicalProperties, bool includeAAFeatures,
bool addEiFeatures
)
{
bool includeAndHlaAndSTWithEpitopeAdjFeatures = false;
bool subtractSupertypeFeatures = false;
bool subtractHlaFeatures = false;
bool substractChemAACrissCrossFeatures = false;
SpecialFunctions.CheckCondition(!includeAndHlaAndSTWithEpitopeAdjFeatures || includeFlankNECFeatures);
Pair <NEC, Hla> necAndHlaX = (Pair <NEC, Hla>)entity;
NEC nec = (null == flankSizeOrNull) ? necAndHlaX.First : NEC.GetInstance(necAndHlaX.First, (int)flankSizeOrNull);
Hla hla = necAndHlaX.Second;
Debug.Assert(nec.N.Length == nec.C.Length); // real assert
Pair <NEC, Hla> necAndHla = new Pair <NEC, Hla>(nec, hla);
Set <IHashableFeature> hlaishFeatureSet = new Set <IHashableFeature>();
CreateAndAddHlaFeature(subtractHlaFeatures, hla, necAndHla, ref hlaishFeatureSet);
CreateAndAddFeatureSupertype(supertypeTableSource, subtractSupertypeFeatures, hla, necAndHla, ref hlaishFeatureSet, Assembly.GetExecutingAssembly(), Predictor.ResourceString);
Set <IHashableFeature> featureSet = Set <IHashableFeature> .GetInstance(hlaishFeatureSet);
if (addEiFeatures)
{
AddEiFeatures(includeChemicalProperties, includeAAFeatures, substractChemAACrissCrossFeatures, nec, necAndHla, hlaishFeatureSet, featureSet);
}
if (includeFlankNECFeatures)
{
List <IHashableFeature> aaInNFlankFeatureList = new List <IHashableFeature>(In.GetAASeqInRegionInstance(1, necAndHla, NFlank.GetInstance()));
DebugCheckThatEvaluatesToTrue(necAndHla, aaInNFlankFeatureList);
if (includeAAFeatures)
{
featureSet.AddNewRange(aaInNFlankFeatureList); //AA in N flank
featureSet.AddNewRange(In.GetAASeqInRegionInstance(2, necAndHla, NFlank.GetInstance())); //AA1-AA2 in Nflank
featureSet.AddNewRange(SubSeq.GetInSubSeqEnumeration(NFlank.GetInstance(), false, 1, necAndHla)); //AA@x in N flank (numbering is 5 4 3 2 1)
featureSet.AddNewRange(SubSeq.GetInSubSeqEnumeration(NFlank.GetInstance(), false, 2, necAndHla)); //AA1-AA2@x in Nflank (x is position of AA2, i.e., the smaller number)
featureSet.AddNewRange(In.GetAASeqInRegionInstance(1, necAndHla, CFlank.GetInstance())); //AA in Cflank
featureSet.AddNewRange(In.GetAASeqInRegionInstance(2, necAndHla, CFlank.GetInstance())); //AA1-AA2 in Cflank
featureSet.AddNewRange(SubSeq.GetInSubSeqEnumeration(CFlank.GetInstance(), true, 1, necAndHla)); //AA@x in C flank (numbering is 1 2 3 4 5)
featureSet.AddNewRange(SubSeq.GetInSubSeqEnumeration(CFlank.GetInstance(), true, 2, necAndHla)); //AA1-AA2@x in Cflank (x is position of AA1, i.e., the smaller number)
}
if (includeChemicalProperties)
{
featureSet.AddNewOrOldRange(InProperty.GetPropertySeqInRegionInstance(1, necAndHla, NFlank.GetInstance()));
featureSet.AddNewOrOldRange(SubSeq.GetInPropertySubSeqEnumeration(NFlank.GetInstance(), false, 1, necAndHla));
featureSet.AddNewOrOldRange(InProperty.GetPropertySeqInRegionInstance(1, necAndHla, CFlank.GetInstance()));
featureSet.AddNewOrOldRange(SubSeq.GetInPropertySubSeqEnumeration(CFlank.GetInstance(), true, 1, necAndHla));
featureSet.AddNewOrOldRange(InProperty.GetPropertySeqInRegionInstance(2, necAndHla, NFlank.GetInstance()));
featureSet.AddNewOrOldRange(SubSeq.GetInPropertySubSeqEnumeration(NFlank.GetInstance(), false, 2, necAndHla));
featureSet.AddNewOrOldRange(InProperty.GetPropertySeqInRegionInstance(2, necAndHla, CFlank.GetInstance()));
featureSet.AddNewOrOldRange(SubSeq.GetInPropertySubSeqEnumeration(CFlank.GetInstance(), true, 2, necAndHla));
}
}
if (includeFlankNECFeatures)
{
if (includeAAFeatures)
{
//EV in Epitope
AddFeatureWithOptionalAndHlaAndST(In.GetAASeqInRegionInstance(2, necAndHla, Epitope.GetInstance()), includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, false, ref featureSet);//AA1-AA2 in Epitope
//RR in Epitope[@1-2]
AddFeatureWithOptionalAndHlaAndST(SubSeq.GetInSubSeqEnumeration(Epitope.GetInstance(), true, 2, necAndHla), includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, false, ref featureSet);//AA1-AA2@x in Epitope (x is position of AA1, i.e., the smaller number)
}
if (includeChemicalProperties)
{
//polar,cyclic in Epitope
AddFeatureWithOptionalAndHlaAndST(InProperty.GetPropertySeqInRegionInstance(2, necAndHla, Epitope.GetInstance()), includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, false, ref featureSet);
//polar,large in Epitope[@8-9]
AddFeatureWithOptionalAndHlaAndST(SubSeq.GetInPropertySubSeqEnumeration(Epitope.GetInstance(), true, 2, necAndHla), includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, false, ref featureSet);
}
//AA1-AA2 in Nflank,Epitope, etc
if (null != flankSizeOrNull && (int)flankSizeOrNull > 0)
{
string epitope = (string)Epitope.GetInstance().Evaluate(entity);
SubSeq lastNAAFeature = SubSeq.GetInstance(1, 1, false, NFlank.GetInstance());
string lastNAA = (string)lastNAAFeature.Evaluate(entity);
In inLastNAA = In.GetInstance(lastNAA, lastNAAFeature);
SubSeq firstEAAFeature = SubSeq.GetInstance(1, 1, true, Epitope.GetInstance());
string firstEAA = (string)firstEAAFeature.Evaluate(entity);
Debug.Assert(firstEAA == epitope.Substring(0, 1));// real assert
In inFirstEAA = In.GetInstance(firstEAA, firstEAAFeature);
SubSeq lastEAAFeature = SubSeq.GetInstance(epitope.Length, epitope.Length, true, Epitope.GetInstance());
string lastEAA = (string)lastEAAFeature.Evaluate(entity);
In inLastEAA = In.GetInstance(lastEAA, lastEAAFeature);
SubSeq firstCAAFeature = SubSeq.GetInstance(1, 1, true, CFlank.GetInstance());
string firstCAA = (string)firstCAAFeature.Evaluate(entity);
In inFirstCAA = In.GetInstance(firstCAA, firstCAAFeature);
if (includeAAFeatures)
{
And andLastNNAAFirstEAA = And.GetInstance(inLastNAA, inFirstEAA);
AddFeatureWithOptionalAndHlaAndST(andLastNNAAFirstEAA, includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, /*checkThatNew*/ true, ref featureSet);
And andLastEAAFirstCAA = And.GetInstance(inLastEAA, inFirstCAA);
AddFeatureWithOptionalAndHlaAndST(andLastEAAFirstCAA, includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, /*checkThatNew*/ true, ref featureSet);
}
if (includeChemicalProperties)
{
foreach (string lastNProperty in KmerProperties.AaToPropList[Biology.GetInstance().OneLetterAminoAcidAbbrevTo3Letter[lastNAA[0]]])
{
InProperty inLastNProperty = InProperty.GetInstance(lastNProperty, lastNAAFeature);
foreach (string firstEProperty in KmerProperties.AaToPropList[Biology.GetInstance().OneLetterAminoAcidAbbrevTo3Letter[firstEAA[0]]])
{
InProperty inFirstEProperty = InProperty.GetInstance(firstEProperty, firstEAAFeature); //!!!get this out of the loop?
And andLastNPropertyFirstEProperty = And.GetInstance(inLastNProperty, inFirstEProperty);
AddFeatureWithOptionalAndHlaAndST(andLastNPropertyFirstEProperty, includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, /*checkThatNew*/ false, ref featureSet);
Debug.Assert((bool)andLastNPropertyFirstEProperty.Evaluate(necAndHla));
}
}
foreach (string lastEProperty in KmerProperties.AaToPropList[Biology.GetInstance().OneLetterAminoAcidAbbrevTo3Letter[lastEAA[0]]])
{
InProperty inlastEProperty = InProperty.GetInstance(lastEProperty, lastEAAFeature);
foreach (string firstCProperty in KmerProperties.AaToPropList[Biology.GetInstance().OneLetterAminoAcidAbbrevTo3Letter[firstCAA[0]]])
{
InProperty infirstCProperty = InProperty.GetInstance(firstCProperty, firstCAAFeature); //!!!get this out of the loop?
And andlastEPropertyfirstCProperty = And.GetInstance(inlastEProperty, infirstCProperty);
AddFeatureWithOptionalAndHlaAndST(andlastEPropertyfirstCProperty, includeAndHlaAndSTWithEpitopeAdjFeatures, hlaishFeatureSet, /*checkThatNew*/ false, ref featureSet);
Debug.Assert((bool)andlastEPropertyfirstCProperty.Evaluate(necAndHla));
}
}
}
}
}
return(featureSet);
}
private static List <Set <char> > CreateSequence(string aaSeqAsString)
{
List <Set <char> > sequence = new List <Set <char> >();
Set <char> set = null;
foreach (char ch in aaSeqAsString)
{
switch (ch)
{
case '{':
{
SpecialFunctions.CheckCondition(set == null, "Nested '{''s are not allowed in aaSeq strings");
set = new Set <char>();
sequence.Add(set);
break;
}
case '}':
{
SpecialFunctions.CheckCondition(set != null, "'}' must follow a '{' in aaSeq strings");
SpecialFunctions.CheckCondition(set.Count > 0, "Empty sets not allow in aaSeq strings");
set = null;
break;
}
case ' ':
{
SpecialFunctions.CheckCondition(false, "Sequences should not contain blanks. Use '?' for missing.");
break;
}
case '?':
case '-':
{
SpecialFunctions.CheckCondition(set == null, string.Format("'{0}' must not appear in sets", ch));
sequence.Add(Set <char> .GetInstance(ch));
break;
}
default:
{
//!!!most this to Biology?
SpecialFunctions.CheckCondition(Biology.GetInstance().OneLetterAminoAcidAbbrevTo3Letter.ContainsKey(ch),
string.Format("The character {0} is not an amino acid", ch));
string aminoAcid = Biology.GetInstance().OneLetterAminoAcidAbbrevTo3Letter[ch];
SpecialFunctions.CheckCondition(Biology.GetInstance().KnownAminoAcid(aminoAcid),
string.Format("The character {0} is not a standard amino acid", ch));
if (set == null)
{
sequence.Add(Set <char> .GetInstance(ch));
}
else
{
set.AddNew(ch);
}
break;
}
}
}
SpecialFunctions.CheckCondition(set == null, "Missing '}' in aaSeq string");
return(sequence);
}