org.strbio.mol.lib.pred2ary
Class TPSet

java.lang.Object
  java.util.AbstractCollection<E>
      java.util.AbstractList<E>
          java.util.Vector
              org.strbio.mol.PolymerSet
                  org.strbio.mol.ProteinSet
                      org.strbio.mol.ProfileSet
                          org.strbio.mol.lib.pred2ary.PCPSet
                              org.strbio.mol.lib.pred2ary.TPSet

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, java.lang.Iterable, java.util.Collection, java.util.List, java.util.RandomAccess

Direct Known Subclasses:

PredictionSet, TrainingSet

public class TPSet
extends PCPSet

Class to represent a set of profiles being used in 2ary structure prediction; this might be either a training or prediction set.

 Version 1.2, 6/7/99 - moved to org.strbio.mol.lib.pred2ary
 Version 1.12, 8/20/98 - moved translateEA() to protein.
 Version 1.11, 8/17/98 - changed some fatal() to exceptions.
 Version 1.1, 5/20/98 - added more ways to getNetworkStats
 Version 1.0, 5/12/98 - adapted from 2ary_set.cpp
 

Version:

1.2, 6/7/99

Author:

JMC

See Also:

PredClassProfile, Serialized Form

Nested Class Summary

static class

TPSet.NetVars Class with variables used to configure the neural nets.

Nested classes/interfaces inherited from class org.strbio.mol.PolymerSet

PolymerSet.PolymerEnumeration

Field Summary

static boolean	CLASS_2ARY turn on the 3 inputs with predicted secondary structure percentages and predicted number of alternations between helix and sheet.
static boolean	CLASS_AA set these to determine exactly what sort of class network to use note that the "sequence length" input is always on.
static boolean	CLASS_STRONG turn on the 2 inputs with "strong" secondary structure predictions.
SimMatrix	CSM
CorrectMatrix[]	CSM1
double	cutoff_2ary
double[]	cutoff_class
double	cutoff_lvl2
int	nstats
SimMatrix	r_SM
SimMatrix	SM
static boolean	USE_LENGTH_INFO Set this to allow the class prediction algorith to know the length rules without learning them from the training set.
TPSet.NetVars	vars

Fields inherited from class java.util.Vector

capacityIncrement, elementCount, elementData

Fields inherited from class java.util.AbstractList

modCount

Constructor Summary

TPSet()

TPSet(TPSet q)

Method Summary

void	addSM(TPSet x)
void	clear() Clear out all information in the set.
void	clearSM()
void	compute2arySM(int which, int sample) compute SM for a set.
void	computeClassSM() compute CSM for a set
void	copyLvl1HE() copy lvl1 predictions for all proteins in set
void	copyUnreducedStats(int sample) copy unreduced predictions for all proteins in set
void	deleteSM()
void	estAccy(IMatrix ea_h, IMatrix ea_e, IMatrix ea_c) estimate accuracy, replacing h, e, predStructure, and the SM matrices
void	estAccyCount(IMatrix ea_h, IMatrix ea_e, IMatrix ea_c) make count matrices for est_accy
double	fcut2ary(NeuralNet net, int sample, Printf outfile) find best 2ary cutoff
void	fcutClass(NeuralNet cnet4, NeuralNet[] cnet1, Printf outfile) find cutoff, and compute class stats
double	fcutLvl2(NeuralNet net, int sample, Printf outfile) find cutoff, and compute stats on a training set.
void	filterPred() filter out short predictions
void	findInputs() find class inputs for all proteins in set
java.lang.String	name2ary(int i) return name of various types of 2ary str
void	newSM(int x) make/delete stats for all proteins in a set
void	predict2ary(NeuralNet net, int sample, Printf outfile) do a 2ary prediction
void	predict2aryHE(NeuralNet net) Predict level 1 H and E using network.
void	predictClass(NeuralNet cnet4, NeuralNet[] cnet1, Printf outfile) do class prediction
void	predictClassDirectly(Printf outfile) do class prediction directly from 2ary prediction
void	predictClassRaw(NeuralNet cnet4, NeuralNet[] cnet1) predict class using both 4-output and 1-output networks.
void	predictLvl2(NeuralNet net, int sample, Printf outfile) do a prediction
void	predictLvl2HE(NeuralNet net) do level 2 prediction
void	print2ary(Printf outfile, int sample)
void	print2aryByClass(Printf outfile)
void	print2aryStats(Printf outfile)
void	print2aryUnreduced(Printf outfile)
void	printClass(Printf outfile)
void	printClassStats(Printf outfile)
void	printDirectClassPrediction(Printf outfile)
void	printDirectClassStats(Printf outfile)
void	smooth(int num) smooth h and e
void	translate2ary() translate he to predictions
void	translateClass() saved net outputs -> class predictions
void	translateEA()

Methods inherited from class org.strbio.mol.lib.pred2ary.PCPSet

addClassPred, addPred, addPred, addPredToArray, clearClassPred, clearPred, combine, divideClassPred, dividePred, loadClassPred, loadPred, newPolymer, newPolymer, pcp, predDiffs, saveClassPred, savePred

Methods inherited from class org.strbio.mol.ProfileSet

blast, blast, write, writeClustal, writeClustal, writeMSF, writeMSF, writeProf, writeProf, writeSAF, writeSAF, writeTDP

Methods inherited from class org.strbio.mol.ProteinSet

findDSSP, findPDB, fixDistanceGaps, predictSS, predictSS, protein, residues, thread, thread, writeCASP, writeCASP, writeConv, writeConv, writeEA, writeEA, writePDB, writePDB, writePDB, writePDB, writeVar2, writeVar2

Methods inherited from class org.strbio.mol.PolymerSet

add, add, add, addReversedCopies, clearPolymers, clearProperties, clearProperty, ensureNames, findClosest, getNames, getPropertyAll, getPropertyOne, isEqual, keepOnlyChainID, keepOnlyNames, keepOnlyNamesFuzzy, load, n, nMonomers, noSpaceNames, nPolymersInFile, p, polymer, polymers, polymersInFile, polymersInFile, polymersInFile, polymersInFile, printNames, read, read, read, readList, remove, remove, removeRedundantSequences, save, searchByName, searchByNameFuzzy, searchByNameFuzzy, searchByNameFuzzyIndex, searchByNameFuzzyIndex, searchByNameIndex, setPolymerAt, setProperty, stripNoAtoms, writeFasta, writeFasta, writeList, writeList, writePTS, writePTS, writeYAPF, writeYAPF

Methods inherited from class java.util.Vector

add, add, addAll, addAll, addElement, capacity, clone, contains, containsAll, copyInto, elementAt, elements, ensureCapacity, equals, firstElement, get, hashCode, indexOf, indexOf, insertElementAt, isEmpty, lastElement, lastIndexOf, lastIndexOf, remove, removeAll, removeAllElements, removeElement, removeElementAt, removeRange, retainAll, set, setElementAt, setSize, size, subList, toArray, toArray, toString, trimToSize

Methods inherited from class java.util.AbstractList

iterator, listIterator, listIterator

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Methods inherited from interface java.util.List

iterator, listIterator, listIterator

Field Detail

USE_LENGTH_INFO

public static final boolean USE_LENGTH_INFO

Set this to allow the class prediction algorith to know the length rules without learning them from the training set.

See Also:

Constant Field Values

CLASS_AA

public static final boolean CLASS_AA

set these to determine exactly what sort of class network to use note that the "sequence length" input is always on. Turn on the 20 inputs with AA percentages.

See Also:

Constant Field Values

CLASS_2ARY

public static final boolean CLASS_2ARY

turn on the 3 inputs with predicted secondary structure percentages and predicted number of alternations between helix and sheet.

See Also:

Constant Field Values

CLASS_STRONG

public static final boolean CLASS_STRONG

turn on the 2 inputs with "strong" secondary structure predictions.

See Also:

Constant Field Values

nstats

public int nstats

r_SM

public SimMatrix r_SM

SM

public SimMatrix SM

CSM

public SimMatrix CSM

CSM1

public CorrectMatrix[] CSM1

cutoff_2ary

public double cutoff_2ary

cutoff_lvl2

public double cutoff_lvl2

cutoff_class

public double[] cutoff_class

vars

public TPSet.NetVars vars

Constructor Detail

TPSet

public TPSet()

TPSet

public TPSet(TPSet q)

Method Detail

clear

public void clear()

Clear out all information in the set.

Specified by:

clear in interface java.util.Collection

Specified by:

clear in interface java.util.List

Overrides:

clear in class PolymerSet

predict2aryHE

public final void predict2aryHE(NeuralNet net)

Predict level 1 H and E using network.

predictLvl2HE

public final void predictLvl2HE(NeuralNet net)

do level 2 prediction

predictClassRaw

public final void predictClassRaw(NeuralNet cnet4,
                                  NeuralNet[] cnet1)

predict class using both 4-output and 1-output networks.

compute2arySM

public final void compute2arySM(int which,
                                int sample)

compute SM for a set.

See Also:

PredClassProfile.compute2arySM(int, int)

computeClassSM

public final void computeClassSM()

compute CSM for a set

predict2ary

public final void predict2ary(NeuralNet net,
                              int sample,
                              Printf outfile)

do a 2ary prediction

predictLvl2

public final void predictLvl2(NeuralNet net,
                              int sample,
                              Printf outfile)

do a prediction

predictClassDirectly

public final void predictClassDirectly(Printf outfile)

do class prediction directly from 2ary prediction

predictClass

public final void predictClass(NeuralNet cnet4,
                               NeuralNet[] cnet1,
                               Printf outfile)

do class prediction

fcut2ary

public final double fcut2ary(NeuralNet net,
                             int sample,
                             Printf outfile)

find best 2ary cutoff

fcutLvl2

public final double fcutLvl2(NeuralNet net,
                             int sample,
                             Printf outfile)

find cutoff, and compute stats on a training set.

fcutClass

public final void fcutClass(NeuralNet cnet4,
                            NeuralNet[] cnet1,
                            Printf outfile)

find cutoff, and compute class stats

smooth

public final void smooth(int num)

smooth h and e

translate2ary

public final void translate2ary()

translate he to predictions

translateEA

public final void translateEA()

translateClass

public final void translateClass()

saved net outputs -> class predictions

filterPred

public final void filterPred()

filter out short predictions

newSM

public void newSM(int x)

make/delete stats for all proteins in a set

deleteSM

public final void deleteSM()

clearSM

public final void clearSM()

addSM

public final void addSM(TPSet x)

copyLvl1HE

public final void copyLvl1HE()

copy lvl1 predictions for all proteins in set

findInputs

public final void findInputs()

find class inputs for all proteins in set

copyUnreducedStats

public final void copyUnreducedStats(int sample)

copy unreduced predictions for all proteins in set

name2ary

public final java.lang.String name2ary(int i)

return name of various types of 2ary str

print2aryStats

public final void print2aryStats(Printf outfile)

print2aryByClass

public final void print2aryByClass(Printf outfile)

print2aryUnreduced

public final void print2aryUnreduced(Printf outfile)

printClassStats

public final void printClassStats(Printf outfile)

printDirectClassStats

public final void printDirectClassStats(Printf outfile)

print2ary

public final void print2ary(Printf outfile,
                            int sample)

printClass

public final void printClass(Printf outfile)

printDirectClassPrediction

public final void printDirectClassPrediction(Printf outfile)

estAccyCount

public final void estAccyCount(IMatrix ea_h,
                               IMatrix ea_e,
                               IMatrix ea_c)

make count matrices for est_accy

estAccy

public final void estAccy(IMatrix ea_h,
                          IMatrix ea_e,
                          IMatrix ea_c)

estimate accuracy, replacing h, e, predStructure, and the SM matrices