Inheritance diagram for MyAbstractGainAlgo< StateType >:

Public Member Functions
	MyAbstractGainAlgo (MyGainGraph &g, MyAnnotations &a, MyGainParams &p, Reporter &r, GeneOntology *go=NULL)
virtual void	computeCrossValidationResults (const BioFunction &function, const MyNodeIdList &cvNodes)=0
	Compute results of cross validation.
virtual void	computeCut (MyNodeIdList &nodesToAnnotate, MyNodeIdSet &cutNodes)
	After assinging states to all nodes, compute the induced cut.
virtual void	computePredictions (const BioFunction &function, const MyNodeIdList &nodesToAnnotate, MyNodeIdList &predictedNodes)=0
	Compute the confidence in each prediction.
virtual MyGainState< StateType >	computeState (MyNode &node)=0
virtual void	constructCrossValidationLists (vector< MyNodeIdList > &cvLists)=0
	Construct lists of nodes for cross validation.
virtual bool	hasConverged (const MyNodeIdList &nodesToAnnotate)=0
virtual void	crossValidate (const BioFunction &function)
virtual void	maskNodeStates (const MyNodeIdList &nodes)=0
	"Forget" known annotations/states of nodes prior to cross validation.
virtual void	evaluatePredictions (MyAnnotations &newAnnotations)
	"Remember" known annotations/states of nodes after cross validation.
virtual string	getName ()=0
virtual void	initialiseNodeStates (const BioFunction &function, MyNodeIdList &nodesToAnnotate)=0
virtual void	pipeline (const BioFunction &function)
	The complete pipeline of computation for functional annotation.
virtual void	predict (const BioFunction &function)
virtual void	run (const MyNodeIdList &nodesToAnnotate)=0
virtual void	setCrossValidationList (MyNodeIdList &nodesToAnnotate, const MyNodeIdList &cvList)
virtual void	visualiseCrossValidation (const BioFunction &currentFunction)
virtual void	visualiseCut (const BioFunction &currentFunction, MyNodeIdList &predictedNodes)
	Layout a graph that highlights the cut seprating nodes annotated with or predicted to be annotated with currentFunction from all the other nodes.
virtual void	visualisePredictions (const BioFunction &currentFunction, MyNodeIdList &nodesToAnnotate)
virtual void	initialiseEdgeWeights (const BioFunction &function)
virtual void	initialiseEdgeWeightsCutoff (const BioFunction &function)
virtual void	initialiseEdgeWeightsDepth (const BioFunction &function)
virtual void	removeUnpredictableNodes (const MyNodeIdSet &unpredictableNodes, MyNodeIdList &oldNodes)
Protected Attributes
MyAnnotations &	_annotations
BioFunction	_currentFunction
GeneOntology *	_go
MyGainGraph &	_graph
MyGainParams &	_params
Reporter &	_reporter
map< StateType, MyNodeIdList >	_nodesByInitialStateType
map< StateType, MyNodeIdList >	_nodesByRealStateType
map< MyNodeId, MyGainStateInfo < MyGainState< StateType > > >	_states
map< MyNodeId, StateType >	_realStateTypes
map< MyNodeId, StateType >	_initialStateTypes
MyNodeIdSet	_cutNodes

template<typename StateType>
class MyAbstractGainAlgo< StateType >

Member Function Documentation

template<typename StateType>

virtual void MyAbstractGainAlgo< StateType >::computeCrossValidationResults	(	const BioFunction &	function,
		const MyNodeIdList &	cvNodes
	)		`[pure virtual]`

Compute results of cross validation.

After invoking crossValidate(), compute the results. Each algorithm needs to define this method, since the algorithm stores the node states.

Implemented in MyOneVersusAllGeneManiaGainAlgo, MyOneVersusAllSinkSourceGainAlgo, and MyOneVersusAllGainAlgo.

template<typename StateType>

virtual void MyAbstractGainAlgo< StateType >::computeCut	(	MyNodeIdList &	nodesToAnnotate,
		MyNodeIdSet &	cutNodes
	)		`[inline, virtual]`

After assinging states to all nodes, compute the induced cut.

An edge is inconsistent if it is incident on nodes with different states. We define the cut as the set of inconsistent edges. Removing these edges decomposes the graph into connected components such that in component, all nodes are in the same state. The method computes the nodes predicted to be annotated with the current function that are incident on edges in the cut.

Parameters:

[in]	nodesToAnnotate,the	list of nodes to be annotated.
[out]	cutNodes,the	list of nodes in the cut.

Note:: By definition, only edges incident on nodes in nodesToAnnotate are eligible to belong to the cut since the evidence propagation algorithms do not propagate any information across other edges.

Reimplemented in MyOneVersusAllGainAlgo.

template<typename StateType>

virtual void MyAbstractGainAlgo< StateType >::computePredictions	(	const BioFunction &	function,
		const MyNodeIdList &	nodesToAnnotate,
		MyNodeIdList &	predictedNodes
	)		`[pure virtual]`

Compute the confidence in each prediction.

For each prediction, compute the algorithm's confidence in the prediction as the distance of the node from the cut.

Implemented in MyOneVersusAllGeneManiaGainAlgo, MyOneVersusNoneFunctionalFlowGainAlgo, MyOneVersusAllSinkSourceGainAlgo, MyOneVersusAllSemiHierarchicalHopfieldGainAlgo, and MyOneVersusAllGainAlgo.

template<typename StateType>

virtual MyGainState< StateType > MyAbstractGainAlgo< StateType >::computeState ( MyNode & node ) [pure virtual]

This method should be a member of this class since the method needs access to the states of the nodes (stored in subclasses of the class).

Implemented in MyOneVersusAllGeneManiaGainAlgo, MyOneVersusNoneLocalGainAlgo, MyOneVersusNoneFunctionalFlowGainAlgo, MyOneVersusAllSinkSourceGainAlgo, MyOneVersusAllHopfieldGainAlgo, MyOneVersusAllLocalGainAlgo, and MyOneVersusAllGainAlgo.

template<typename StateType>

virtual void MyAbstractGainAlgo< StateType >::constructCrossValidationLists ( vector< MyNodeIdList > & cvLists ) [pure virtual]

Construct lists of nodes for cross validation.

Parameters:

[out] cvLists,a vector of MyNodeIdLists.

Each element of this vector contains a list of node ids (stored in a MyNodeIdList) whose annotation details should be deleted for cross validation. In the case of leave-one-out cross validation, each element of the vector will contain just one node.

Implemented in MyOneVersusAllGainAlgo.

template<typename StateType >

void MyAbstractGainAlgo< StateType >::evaluatePredictions ( MyAnnotations & newAnnotations ) [virtual]

"Remember" known annotations/states of nodes after cross validation.

Parameters:

[in] nodes,a MyNodeIdList containing a list of node IDs for which I should unmask state and annotation information.Evaluate the predictions made by the algorithm based on the (new) annotations.

The method compares the current annotations (on which the algorithm based the predictions) with newAnnotations. It first finds those genes for which the algorithm made predictions and that have new annotations in newAnnotations. For each such gene, the method computes the distance of each prediction to the closest new annotation for that gene. The method outputs results to a file.

Note:: Invoke this method only after making predictions for all functions.

template<typename StateType>

virtual string MyAbstractGainAlgo< StateType >::getName ( ) [pure virtual]

Return the name of the algorithm.

Implements MyReallyAbstractGainAlgo.

template<typename StateType>

virtual void MyAbstractGainAlgo< StateType >::maskNodeStates ( const MyNodeIdList & nodes ) [pure virtual]

"Forget" known annotations/states of nodes prior to cross validation.

Parameters:

[in] nodes,a MyNodeIdList containing a list of node IDs for which I should mask state and annotation information.

Implemented in MyOneVersusAllGeneManiaGainAlgo, MyOneVersusAllSinkSourceGainAlgo, and MyOneVersusAllGainAlgo.

template<typename StateType >

void MyAbstractGainAlgo< StateType >::setCrossValidationList	(	MyNodeIdList &	nodesToAnnotate,
		const MyNodeIdList &	cvList
	)		`[virtual]`

Given a list of unannotated nodes (nodesToAnnotate) and a list of nodes to cross validate (cvList),update nodesToAnnotate to include nodes in cvList, so that the annotation algorithm also annotates the nodes in cvList.

Warning:: Call this method only after invoking maskNodeStates().

Note:: This method is virtual since some algorithms can set nodesToAnnotate to be equal to cvList, for example, algorithms that use only one round of "guilt-by-association," where each node computes its state from the states of its neighbours.

Reimplemented in MyOneVersusAllGeneManiaGainAlgo, MyOneVersusNoneSinkSourceGainAlgo, MyOneVersusAllAbstractSVMGainAlgo, and MyOneVersusNoneGainAlgo.

template<typename StateType >

void MyAbstractGainAlgo< StateType >::visualiseCrossValidation ( const BioFunction & currentFunction ) [virtual]

Layout the functional linkage network to visualise the results.

template<typename StateType >

void MyAbstractGainAlgo< StateType >::visualiseCut	(	const BioFunction &	currentFunction,
		MyNodeIdList &	predictedNodes
	)		`[virtual]`

Layout a graph that highlights the cut seprating nodes annotated with or predicted to be annotated with currentFunction from all the other nodes.

The method lays out the graph induced by the edges in the cut and all the nodes on the "currentFunction" side of the cut.

Reimplemented in MyOneVersusAllAbstractSVMGainAlgo, and MyOneVersusAllGainAlgo.

The documentation for this class was generated from the following file:

/home/poirel/src/c++/biorithm/gain/gain-algorithm.h

Public Member Functions

Protected Attributes

template<typename StateType> class MyAbstractGainAlgo< StateType >

Member Function Documentation

template<typename StateType>
class MyAbstractGainAlgo< StateType >