Simultaneously align multiple protein sequences. More...

#include <algo/cobalt/cobalt.hpp>

Inheritance diagram for CMultiAligner:

Collaboration diagram for CMultiAligner:

Classes
class	compare_sseg_db_idx

struct	SColumn
	Column in an alignment used for combining result from multiple alignment and pair-wise in-cluster alignments. More...

struct	SGraphNode

struct	SProgress
	Structure for reporting alignment progress. More...

struct	SSegmentLoc

Public Types
enum	EStatus { eSuccess = 0 , eOptionsError , eQueriesError , eDatabaseError , eInternalError , eInterrupt , eOutOfMemory , eLastStatus = eOutOfMemory }
	Return status. More...

enum	EAlignmentStage { eBegin , eQueryClustering , eDomainHitsSearch , eLocalHitsSearch , ePatternHitsSearch , eTreeComputation , eProgressiveAlignment , eIterativeAlignment }

typedef int	TStatus

typedef bool(*	FInterruptFn) (SProgress *progress)
	Prototype for function pointer to dertermine whether alignment should proceed of be interrupted. More...

typedef CSparseKmerCounts	TKmerCounts

typedef TKmerMethods< TKmerCounts >	TKMethods

typedef pair< TRange, TRange >	TRangePair

Public Types inherited from CObject
enum	EAllocFillMode { eAllocFillNone = 1 , eAllocFillZero , eAllocFillPattern }
	Control filling of newly allocated memory. More...

typedef CObjectCounterLocker	TLockerType
	Default locker type for CRef. More...

typedef atomic< Uint8 >	TCounter
	Counter type is CAtomiCounter. More...

typedef Uint8	TCount
	Alias for value type of counter. More...

Public Member Functions
	CMultiAligner (void)
	Create mutli aligner with default options. More...

	CMultiAligner (const string &rps_db)
	Create multi aligner with selected RPS data base and default options. More...

	CMultiAligner (const CConstRef< CMultiAlignerOptions > &options)
	Create mutli aligner with given options. More...

void	SetQueries (const vector< CRef< objects::CSeq_loc > > &queries, CRef< objects::CScope > scope)
	Set query sequences. More...

void	SetQueries (const vector< CRef< objects::CBioseq > > &queries)
	Set query sequences. More...

void	SetQueries (const blast::TSeqLocVector &queries)
	Set query sequences. More...

void	SetInputMSAs (const objects::CSeq_align &msa1, const objects::CSeq_align &msa2, const set< int > &representatives1, const set< int > &representatives2, CRef< objects::CScope > scope)
	Set input alignments. More...

const vector< CRef< objects::CSeq_loc > > &	GetQueries (void) const
	Get query sequences. More...

CRef< objects::CScope >	GetScope (void)
	Get scope. More...

CConstRef< CMultiAlignerOptions >	GetOptions (void) const
	Get mutli aligner parameters. More...

TStatus	Run (void)
	Align the current set of input sequences (reset any existing alignment information). More...

void	Reset (void)
	Clear out the state left by the previous alignment operation. More...

CRef< objects::CSeq_align >	GetResults (void) const
	Retrieve the current aligned results in Seq-align format. More...

CRef< objects::CSeq_align >	GetResults (vector< int > &indices) const
	Retrieve a selection of the current aligned results, in Seq-align format. More...

const vector< CSequence > &	GetSeqResults (void) const
	Retrieve the current aligned results in CSequence format. More...

const TPhyTreeNode *	GetTree (void) const
	Get ree used guide in progressive alignment. More...

CRef< objects::CBioTreeContainer >	GetTreeContainer (void) const
	Get serializable tree used as guide in progressive alignment. More...

const CClusterer::TClusters &	GetQueryClusters (void) const
	Get clusters of query sequences. More...

int	GetScore (void) const
	Get alignment score. More...

FInterruptFn	SetInterruptCallback (FInterruptFn fnptr, void *user_data=NULL)
	Set a function callback to be invoked by multi aligner to allow interrupting alignment in progress. More...

const vector< string > &	GetMessages (void) const
	Get Error/Warning messages. More...

bool	IsMessage (void) const
	Check whether there are any error/warning messages. More...

Public Member Functions inherited from CObject
	CObject (void)
	Constructor. More...

	CObject (const CObject &src)
	Copy constructor. More...

virtual	~CObject (void)
	Destructor. More...

CObject &	operator= (const CObject &src) THROWS_NONE
	Assignment operator. More...

bool	CanBeDeleted (void) const THROWS_NONE
	Check if object can be deleted. More...

bool	IsAllocatedInPool (void) const THROWS_NONE
	Check if object is allocated in memory pool (not system heap) More...

bool	Referenced (void) const THROWS_NONE
	Check if object is referenced. More...

bool	ReferencedOnlyOnce (void) const THROWS_NONE
	Check if object is referenced only once. More...

void	AddReference (void) const
	Add reference to object. More...

void	RemoveReference (void) const
	Remove reference to object. More...

void	ReleaseReference (void) const
	Remove reference without deleting object. More...

virtual void	DoNotDeleteThisObject (void)
	Mark this object as not allocated in heap – do not delete this object. More...

virtual void	DoDeleteThisObject (void)
	Mark this object as allocated in heap – object can be deleted. More...

void *	operator new (size_t size)
	Define new operator for memory allocation. More...

void *	operator new[] (size_t size)
	Define new[] operator for 'array' memory allocation. More...

void	operator delete (void *ptr)
	Define delete operator for memory deallocation. More...

void	operator delete[] (void *ptr)
	Define delete[] operator for memory deallocation. More...

void *	operator new (size_t size, void *place)
	Define new operator. More...

void	operator delete (void ptr, void place)
	Define delete operator. More...

void *	operator new (size_t size, CObjectMemoryPool *place)
	Define new operator using memory pool. More...

void	operator delete (void ptr, CObjectMemoryPool place)
	Define delete operator. More...

virtual void	DebugDump (CDebugDumpContext ddc, unsigned int depth) const
	Define method for dumping debug information. More...

Public Member Functions inherited from CDebugDumpable
	CDebugDumpable (void)

virtual	~CDebugDumpable (void)

void	DebugDumpText (ostream &out, const string &bundle, unsigned int depth) const

void	DebugDumpFormat (CDebugDumpFormatter &ddf, const string &bundle, unsigned int depth) const

void	DumpToConsole (void) const

Static Public Attributes
static const int	kMajorVersion = 3
	Version information. More...

static const int	kMinorVersion = 0

static const int	kPatchVersion = 0

static const CNWAligner::TScore	kDefaultGapOpen = -11
	Default gap open penalty. More...

static const CNWAligner::TScore	kDefaultGapExtend = -1
	Default gap extension penalty. More...

Static Public Attributes inherited from CObject
static const TCount	eCounterBitsCanBeDeleted = 1 << 0
	Define possible object states. More...

static const TCount	eCounterBitsInPlainHeap = 1 << 1
	Heap signature was found. More...

static const TCount	eCounterBitsPlaceMask
	Mask for 'in heap' state flags. More...

static const int	eCounterStep = 1 << 2
	Skip over the "in heap" bits. More...

static const TCount	eCounterValid = TCount(1) << (sizeof(TCount) * 8 - 2)
	Minimal value for valid objects (reference counter is zero) Must be a single bit value. More...

static const TCount	eCounterStateMask
	Valid object, and object in heap. More...

Protected Types
typedef struct CMultiAligner::SSegmentLoc	SSegmentLoc

Protected Member Functions
bool	x_ValidateQueries (void) const
	Validate query sequences. More...

bool	x_ValidateInputMSAs (void) const
	Validate input alignments. More...

bool	x_ValidateUserHits (void)
	Validate user constraints with queries. More...

void	x_CreateBlastQueries (blast::TSeqLocVector &queries, vector< int > &indices)
	Create query set for RPS Blast and Blastp searches along with indices in multiple alignment queries array. More...

void	x_CreatePatternQueries (vector< const CSequence * > &queries, vector< int > &indices)
	Create query set for PROSITE pattern search along with indices in multiple alignment queries array. More...

void	x_FindDomainHits (blast::TSeqLocVector &queries, const vector< int > &indices)
	Run RPS blast on seletced input sequences and postprocess the results. More...

void	x_FindLocalHits (const blast::TSeqLocVector &queries, const vector< int > &indices)
	Run blast on selected input sequences and postprocess the results. More...

void	x_FindLocalInClusterHits (const vector< TPhyTreeNode * > &cluster_trees)
	Run blast on sequences from each cluster subtree. More...

void	x_FindPatternHits (const vector< const CSequence * > &queries, const vector< int > &indices)
	Find PROSITE pattern hits on selected input sequences. More...

void	x_FindConsistentHitSubset (void)
	Find consistent subset of pair-wise hits that can be used as alignment constraints. More...

void	x_ComputeTree ()
	Given the current list of domain and local hits, generate a phylogenetic tree that clusters the current input sequences. More...

void	x_BuildAlignment ()
	Given the current domain, local, pattern and user hits, along with the current tree, compute a multiple alignment of the input sequences. More...

bool	x_FindQueryClusters ()
	Find clusters of similar queries, select cluster representative sequences, and prepare input to multiple alignement composed of only representatives. More...

void	x_AlignInClusters ()
	Pair-wise align each cluster sequence to cluster representative. More...

void	x_MakeClusterResidueFrequencies ()
	Compute profile residue frequencies for clusters. More...

void	x_MultiAlignClusters ()
	Combine pair-wise in-cluster alignements with multiple alignments of cluster prototypes. More...

void	x_ComputeClusterTrees (vector< TPhyTreeNode * > &trees)
	Compute independent phylogenetic trees each cluster. More...

void	x_AttachClusterTrees (const vector< TPhyTreeNode * > &cluster_trees, const vector< TPhyTreeNode * > &cluster_leaves)
	Replace leaves in the alignment guide tree of clusters with cluster trees. More...

void	x_BuildFullTree (const vector< TPhyTreeNode * > &cluster_trees)
	Combine alignment guide tree computed for clusters with guide trees computed for each cluster. More...

virtual void	x_Run (void)
	Align the current set of input sequences (reset any existing alignment information). More...

void	x_AlignMSAs (void)
	Align multiple sequence alignments. More...

void	x_SetDomainHits (const blast::TSeqLocVector &queruies, const vector< int > &indices, const objects::CBlast4_archive &archive)
	Set pre-computed domain hits using BLAST archive format. More...

Protected Member Functions inherited from CObject
virtual void	DeleteThis (void)
	Virtual method "deleting" this object. More...

Protected Attributes
CConstRef< CMultiAlignerOptions >	m_Options

vector< CRef< objects::CSeq_loc > >	m_tQueries

CRef< objects::CScope >	m_Scope

vector< CSequence >	m_QueryData

vector< CSequence >	m_InMSA1
	Input alignment. More...

vector< CSequence >	m_InMSA2
	Input alignment. More...

vector< int >	m_Msa1Repr
	Indices of sequence representatives in input alignment 1. More...

vector< int >	m_Msa2Repr
	Indices of sequence representatives in input alignment 2. More...

vector< CSequence >	m_Results

int	m_Score
	Alignment score. More...

CPSSMAligner	m_Aligner

CTree	m_Tree

CClusterer	m_Clusterer

CHitList	m_DomainHits

CHitList	m_LocalHits

CHitList	m_CombinedHits

CHitList	m_PatternHits

CHitList	m_LocalInClusterHits

CHitList	m_UserHits

vector< bool >	m_IsDomainSearched
	Marks sequences with pre-computed domain hits. More...

vector< vector< Uint4 > >	m_ClusterGapPositions

vector< CRef< objects::CSeq_loc > >	m_AllQueries

vector< CSequence >	m_AllQueryData

vector< vector< TRange > >	m_RPSLocs

FInterruptFn	m_Interrupt

SProgress	m_ProgressMonitor

vector< string >	m_Messages

CMultiAlignerOptions::EInClustAlnMethod	m_ClustAlnMethod

Static Protected Attributes
static const int	kClusterNodeId = 16000

Private Types
enum	EEndGapCostStrategy { fReduceLeft = 1 , fReduceRight = 2 , fReduceBoth = fReduceLeft \| fReduceRight }
	Strategy for reducing end gap penalties for profile-profile alignment. More...

typedef struct CMultiAligner::SGraphNode	SGraphNode

typedef struct CMultiAligner::SColumn	SColumn
	Column in an alignment used for combining result from multiple alignment and pair-wise in-cluster alignments. More...

Private Member Functions
void	x_InitParams (void)
	Initiate parameters using m_Options. More...

void	x_InitAligner (void)
	Initiate PSSM aligner parameters. More...

void	x_SetScoreMatrix (const char *matrix_name)
	Set the score matrix the aligner will use. More...

void	x_Init (void)
	Initiate class attributes that are not alignment parameters. More...

void	x_LoadBlockBoundaries (string blockfile, vector< SSegmentLoc > &blocklist)
	Given an RPS blast database, load a list of block offsets for each database sequence. More...

void	x_FindRPSHits (blast::TSeqLocVector &queries, const vector< int > &indices, CHitList &rps_hits)

void	x_RealignBlocks (CHitList &rps_hits, vector< SSegmentLoc > &blocklist, CProfileData &profile_data)

void	x_AssignRPSResFreqs (CHitList &rps_hits, CProfileData &profile_data)

void	x_AssignDefaultResFreqs ()

void	x_AddNewSegment (vector< CRef< objects::CSeq_loc > > &loc_list, const CRef< objects::CSeq_loc > &query, TOffset from, TOffset to, vector< SSegmentLoc > &seg_list, int query_index)
	Create a new query sequence that is a subset of a previous query sequence. More...

void	x_MakeFillerBlocks (const vector< int > &indices, vector< CRef< objects::CSeq_loc > > &filler_locs, vector< SSegmentLoc > &filler_segs)
	Turn all fragments of selected query sequence not already covered by a domain hit into a separate query sequence, used as input to a blast search. More...

void	x_AlignFillerBlocks (const blast::TSeqLocVector &queries, const vector< int > &indices, vector< CRef< objects::CSeq_loc > > &filler_locs, vector< SSegmentLoc > &filler_segs)
	Run blastp, aligning the collection of filler fragments against the entire input dataset. More...

void	x_FindAlignmentSubsets ()

SGraphNode *	x_FindBestPath (vector< SGraphNode > &nodes)
	Find a maximum weight path in a directed acyclic graph. More...

void	x_BuildAlignmentIterative (vector< CTree::STreeEdge > &edges, double cluster_cutoff)
	Main driver for the progressive alignment process. More...

void	x_FindConservedColumns (vector< CSequence > &new_alignment, CHitList &conserved)
	Create a list of constraints that reflect conserved columns in a multiple alignment. More...

void	x_AlignProgressive (const TPhyTreeNode *tree, vector< CSequence > &query_data, CNcbiMatrix< CHitList > &pair_info, int iteration, bool is_cluster)
	Main driver for progressive alignment. More...

double	x_RealignSequences (const TPhyTreeNode *input_cluster, vector< CSequence > &alignment, CNcbiMatrix< CHitList > &pair_info, double score, int iteration)
	Perform a single bipartition on a multiple alignment. More...

void	x_AlignProfileProfile (vector< CTree::STreeLeaf > &node_list1, vector< CTree::STreeLeaf > &node_list2, vector< CSequence > &alignment, CNcbiMatrix< CHitList > &pair_info, int iteration)
	Align two collections of sequences. More...

void	x_AlignProfileProfileUsingHit (vector< CTree::STreeLeaf > &node_list1, vector< CTree::STreeLeaf > &node_list2, vector< CSequence > &alignment, CNcbiMatrix< CHitList > &pair_info, int iteration)
	Align two profiles with all sequences that belong to the same cluster. More...

void	x_FindConstraints (vector< size_t > &constraint, vector< CSequence > &alignment, vector< CTree::STreeLeaf > &node_list1, vector< CTree::STreeLeaf > &node_list2, CNcbiMatrix< CHitList > &pair_info, int iteration)
	Find the set of constraints to use for a profile-profile alignment. More...

void	x_FindInClusterConstraints (vector< CSequence > &alignment, vector< CTree::STreeLeaf > &node_list1, vector< CTree::STreeLeaf > &node_list2, CNcbiMatrix< CHitList > &pair_info, vector< TRangePair > &match_ranges) const
	Find constraint to use for profile to profile alignment in clusters. More...

double	x_GetScoreOneCol (vector< CSequence > &align, int col)
	Calculate the entropy score of one column of a multiple alignment (see the COBALT papaer for details) More...

double	x_GetScore (vector< CSequence > &align)
	Compute the entropy score of a multiple alignment. More...

CRef< objects::CSeq_align >	x_GetSeqalign (const vector< CSequence > &align, vector< int > &indices) const
	Produce a seqalign representing the specified alignment, using a subset of the sequences. More...

void	x_AddRpsFreqsToCluster (const CClusterer::CSingleCluster &cluster, vector< CSequence > &query_data, const vector< TRange > &gaps)

unique_ptr< vector< int > >	x_AlignClusterQueries (const TPhyTreeNode *node)

void	x_ComputeProfileRangeAlignment (vector< CTree::STreeLeaf > &node_list1, vector< CTree::STreeLeaf > &node_list2, vector< CSequence > &alignment, vector< size_t > &constraints, const TRange &range1, const TRange &range2, int full_prof_len1, int full_prof_len2, EEndGapCostStrategy strat, CNWAligner::TTranscript &t)
	Compute profile profile alignmnet for a ranges of given profiles. More...

Static Private Member Functions
static void	x_InitColumn (vector< SColumn >::iterator &it, size_t len)

static void	x_InitInsertColumn (vector< SColumn >::iterator &it, size_t len, int num, int cluster)

Static Private Attributes
static const int	kRpsScaleFactor = 100

Friends
class	compare_sseg_db_idx

class	::CMultiAlignerTest

Additional Inherited Members
Static Public Member Functions inherited from CObject
static NCBI_XNCBI_EXPORT void	ThrowNullPointerException (void)
	Define method to throw null pointer exception. More...

static NCBI_XNCBI_EXPORT void	ThrowNullPointerException (const type_info &type)

static EAllocFillMode	GetAllocFillMode (void)

static void	SetAllocFillMode (EAllocFillMode mode)

static void	SetAllocFillMode (const string &value)
	Set mode from configuration parameter value. More...

Static Public Member Functions inherited from CDebugDumpable
static void	EnableDebugDump (bool on)

Detailed Description

Simultaneously align multiple protein sequences.

Definition at line 68 of file cobalt.hpp.

Member Typedef Documentation

◆ FInterruptFn

typedef bool(* CMultiAligner::FInterruptFn) (SProgress *progress)

Prototype for function pointer to dertermine whether alignment should proceed of be interrupted.

If this function returns true, all processing stops

Definition at line 112 of file cobalt.hpp.

◆ SColumn

typedef struct CMultiAligner::SColumn CMultiAligner::SColumn

private

Column in an alignment used for combining result from multiple alignment and pair-wise in-cluster alignments.

◆ SGraphNode

typedef struct CMultiAligner::SGraphNode CMultiAligner::SGraphNode

private

◆ SSegmentLoc

typedef struct CMultiAligner::SSegmentLoc CMultiAligner::SSegmentLoc

protected

◆ TKmerCounts

typedef CSparseKmerCounts CMultiAligner::TKmerCounts

Definition at line 114 of file cobalt.hpp.

◆ TKMethods

typedef TKmerMethods<TKmerCounts> CMultiAligner::TKMethods

Definition at line 115 of file cobalt.hpp.

◆ TRangePair

typedef pair<TRange, TRange> CMultiAligner::TRangePair

Definition at line 116 of file cobalt.hpp.

◆ TStatus

typedef int CMultiAligner::TStatus

Definition at line 88 of file cobalt.hpp.

Member Enumeration Documentation

◆ EAlignmentStage

enum CMultiAligner::EAlignmentStage

Enumerator
eBegin
eQueryClustering
eDomainHitsSearch
eLocalHitsSearch
ePatternHitsSearch
eTreeComputation
eProgressiveAlignment
eIterativeAlignment

Definition at line 90 of file cobalt.hpp.

◆ EEndGapCostStrategy

enum CMultiAligner::EEndGapCostStrategy

private

Strategy for reducing end gap penalties for profile-profile alignment.

Enumerator
fReduceLeft	Reduce penalty only for left end gaps.
fReduceRight	Reduce penalty only for right end gaps.
fReduceBoth	Reduce penalty for both end gaps.

Definition at line 553 of file cobalt.hpp.

◆ EStatus

enum CMultiAligner::EStatus

Return status.

Enumerator
eSuccess	Alignment successfully completed.
eOptionsError	Error related to options occured.
eQueriesError	Error related to query sequences occured.
eDatabaseError	Error related to RPS database occured.
eInternalError	Unexpected error occured.
eInterrupt	Alignment interruped through callback function.
eOutOfMemory	Out of memory error.
eLastStatus

Definition at line 77 of file cobalt.hpp.

Constructor & Destructor Documentation

◆ CMultiAligner() [1/3]

CMultiAligner::CMultiAligner ( void )

Create mutli aligner with default options.

Definition at line 51 of file cobalt.cpp.

◆ CMultiAligner() [2/3]

CMultiAligner::CMultiAligner ( const string & rps_db )

Create multi aligner with selected RPS data base and default options.

Parameters

rps_db RPS data base path [in]

Definition at line 61 of file cobalt.cpp.

References x_Init(), x_InitAligner(), and x_InitParams().

◆ CMultiAligner() [3/3]

CMultiAligner::CMultiAligner ( const CConstRef< CMultiAlignerOptions > & options )

Create mutli aligner with given options.

Parameters

options Parameters [in]

Definition at line 70 of file cobalt.cpp.

References x_Init(), x_InitAligner(), and x_InitParams().

Member Function Documentation

◆ GetMessages()

const vector<string>& CMultiAligner::GetMessages ( void ) const

inline

Get Error/Warning messages.

Errors are reported by exceptions, hence the messages will be mostly warnings.

Returns: Messages

Definition at line 284 of file cobalt.hpp.

Referenced by CMultiApplication::Run().

◆ GetOptions()

CConstRef<CMultiAlignerOptions> CMultiAligner::GetOptions ( void ) const

inline

Get mutli aligner parameters.

Returns: Options

Definition at line 196 of file cobalt.hpp.

Referenced by s_TestResults().

◆ GetQueries()

const vector< CRef<objects::CSeq_loc> >& CMultiAligner::GetQueries ( void ) const

inline

Get query sequences.

Returns: List of seq-ids and locations [in]

Definition at line 182 of file cobalt.hpp.

Referenced by CMultiAlignerTest::CompareDomainHits(), s_TestQueriesAsBioseqs(), s_TestQueriesAsSeq_locs(), and s_TestResults().

◆ GetQueryClusters()

const CClusterer::TClusters& CMultiAligner::GetQueryClusters ( void ) const

inline

Get clusters of query sequences.

Returns: Query clusters

Definition at line 255 of file cobalt.hpp.

Referenced by s_TestResults().

◆ GetResults() [1/2]

CRef< CSeq_align > CMultiAligner::GetResults ( vector< int > & indices ) const

Retrieve a selection of the current aligned results, in Seq-align format.

The Seq-align is of global type, with a single denseg. Columns that have gaps in all the selected sequences are removed

Parameters

indices List of ordinal IDs of sequences that the Seq-align will contain. Indices may appear in any order and may be repeated

Returns: The results

Definition at line 177 of file seqalign.cpp.

References CSequence::CompressSequences(), i, m_Results, NCBI_THROW, and x_GetSeqalign().

◆ GetResults() [2/2]

CRef< CSeq_align > CMultiAligner::GetResults ( void ) const

Retrieve the current aligned results in Seq-align format.

The Seq-align is of global type, with a single denseg

Returns: The results

Definition at line 157 of file seqalign.cpp.

References i, m_Results, NCBI_THROW, and x_GetSeqalign().

Referenced by CMultiApplication::Run(), and s_TestResults().

◆ GetScope()

CRef<objects::CScope> CMultiAligner::GetScope ( void )

inline

Get scope.

Returns: Scope

Definition at line 188 of file cobalt.hpp.

References m_Scope.

Referenced by CMultiApplication::Run(), and s_TestResults().

◆ GetScore()

int CMultiAligner::GetScore ( void ) const

inline

Get alignment score.

Returns: Alignment score

Definition at line 261 of file cobalt.hpp.

◆ GetSeqResults()

const vector<CSequence>& CMultiAligner::GetSeqResults ( void ) const

inline

Retrieve the current aligned results in CSequence format.

Included for backward compatibility with previous API.

Returns: The results, on CSequence for each input sequence

Definition at line 240 of file cobalt.hpp.

Referenced by CMultiApplication::Run(), and s_TestResults().

◆ GetTree()

const TPhyTreeNode* CMultiAligner::GetTree ( void ) const

inline

Get ree used guide in progressive alignment.

Returns: Tree

Definition at line 245 of file cobalt.hpp.

Referenced by s_TestResults(), x_BuildAlignment(), x_BuildAlignmentIterative(), and x_RealignSequences().

◆ GetTreeContainer()

CRef< objects::CBioTreeContainer > CMultiAligner::GetTreeContainer ( void ) const

Get serializable tree used as guide in progressive alignment.

Returns: Tree

Definition at line 357 of file cobalt.cpp.

References CTree::GetTree(), m_Tree, MakeBioTreeContainer(), and NCBI_THROW.

Referenced by s_TestResults().

◆ IsMessage()

bool CMultiAligner::IsMessage ( void ) const

inline

Check whether there are any error/warning messages.

Returns: True if there are messages, false otherwise

Definition at line 289 of file cobalt.hpp.

◆ Reset()

void CMultiAligner::Reset ( void )

Clear out the state left by the previous alignment operation.

Definition at line 403 of file cobalt.cpp.

References m_CombinedHits, m_DomainHits, m_LocalHits, m_PatternHits, m_Results, and CHitList::PurgeAllHits().

Referenced by SetInputMSAs(), and SetQueries().

◆ Run()

CMultiAligner::TStatus CMultiAligner::Run ( void )

Align the current set of input sequences (reset any existing alignment information).

This function handles the generation of all internal state in the correct order. It is sufficient for 'black box' applications that only want a final answer without tweaking internal state. x_Run() is called for computing alignment.

Returns: Computation status: success (0), warnings (1), error (>1)

Definition at line 683 of file cobalt.cpp.

References eDatabaseError, eInternalError, eInterrupt, CMultiAlignerException::eInterrupt, CMultiAlignerException::eInvalidInput, CMultiAlignerException::eInvalidOptions, CMultiAlignerException::eInvalidScoreMatrix, eOptionsError, eOutOfMemory, CMultiAlignerException::eOutOfMemory, eQueriesError, eSuccess, CException::GetErrCode(), CException::GetMsg(), m_Messages, CException::what(), and x_Run().

Referenced by BOOST_AUTO_TEST_CASE(), and CMultiApplication::Run().

◆ SetInputMSAs()

void CMultiAligner::SetInputMSAs	(	const objects::CSeq_align &	msa1,
		const objects::CSeq_align &	msa2,
		const set< int > &	representatives1,
		const set< int > &	representatives2,
		CRef< objects::CScope >	scope
	)

Set input alignments.

Parameters

msa1	The first input alignment [in]
msa2	The second input alignment [in]
representatives1	List of sequence indices in msa1 to be used for computing constraints [in]
representatives2	List of sequence indices in msa2 to be used for computing constraints [in]
scope	Scope [in]

Definition at line 297 of file cobalt.cpp.

References set< Key, Compare >::begin(), copy(), CSequence::CreateMsa(), set< Key, Compare >::empty(), set< Key, Compare >::end(), i, int, ITERATE, m_InMSA1, m_InMSA2, m_Msa1Repr, m_Msa2Repr, m_Scope, m_tQueries, Reset(), set< Key, Compare >::size(), x_ValidateInputMSAs(), and x_ValidateUserHits().

Referenced by CMultiApplication::Run().

◆ SetInterruptCallback()

CMultiAligner::FInterruptFn CMultiAligner::SetInterruptCallback	(	CMultiAligner::FInterruptFn	fnptr,
		void *	user_data = `NULL`
	)

Set a function callback to be invoked by multi aligner to allow interrupting alignment in progress.

Parameters

fnptr	Pointer to callback function [in]
user_data	user data to be attached to progress structure [in]

Returns: Previously set callback function

Definition at line 370 of file cobalt.cpp.

References m_Interrupt, m_ProgressMonitor, and CMultiAligner::SProgress::user_data.

◆ SetQueries() [1/3]

void CMultiAligner::SetQueries ( const blast::TSeqLocVector & queries )

Set query sequences.

This automatically clears out the intermediate state of the last alignment.

Parameters

queries List of query sequences [in]

Definition at line 263 of file cobalt.cpp.

References i, ITERATE, m_QueryData, m_Scope, m_tQueries, NCBI_THROW, Reset(), x_ValidateQueries(), and x_ValidateUserHits().

◆ SetQueries() [2/3]

void CMultiAligner::SetQueries ( const vector< CRef< objects::CBioseq > > & queries )

Set query sequences.

This automatically clears out the intermediate state of the last alignment.

Parameters

queries List of query sequences [in]

Definition at line 219 of file cobalt.cpp.

References CException::GetMsg(), ITERATE, m_QueryData, m_Scope, m_tQueries, NCBI_THROW, Reset(), CRef< C, Locker >::Reset(), x_ValidateQueries(), and x_ValidateUserHits().

◆ SetQueries() [3/3]

void CMultiAligner::SetQueries	(	const vector< CRef< objects::CSeq_loc > > &	queries,
		CRef< objects::CScope >	scope
	)

Set query sequences.

This automatically clears out the intermediate state of the last alignment.

Parameters

queries	List of query sequences or seq-ids [in]
scope	Scope object [in]

Definition at line 194 of file cobalt.cpp.

References copy(), ITERATE, m_QueryData, m_Scope, m_tQueries, NCBI_THROW, Reset(), x_ValidateQueries(), and x_ValidateUserHits().

Referenced by BOOST_AUTO_TEST_CASE(), CMultiApplication::Run(), s_TestQueriesAsBioseqs(), and s_TestQueriesAsSeq_locs().

◆ x_AddNewSegment()

void CMultiAligner::x_AddNewSegment	(	vector< CRef< objects::CSeq_loc > > &	loc_list,
		const CRef< objects::CSeq_loc > &	query,
		TOffset	from,
		TOffset	to,
		vector< SSegmentLoc > &	seg_list,
		int	query_index
	)

private

Create a new query sequence that is a subset of a previous query sequence.

Parameters

loc_list	List of previously generated sequence fragments [in/out]
query	Sequence that contains the current fragment [in]
from	Start offset of fragment [in]
to	End offset of fragment [in]
seg_list	List of simplified representations of previous fragments [in/out]
query_index	Ordinal ID of 'query'

Definition at line 68 of file blast.cpp.

References eNa_strand_unknown, GetId(), m_Scope, query, and CSeq_loc::SetInt().

Referenced by x_MakeFillerBlocks().

◆ x_AddRpsFreqsToCluster()

void CMultiAligner::x_AddRpsFreqsToCluster	(	const CClusterer::CSingleCluster &	cluster,
		vector< CSequence > &	query_data,
		const vector< TRange > &	gaps
	)

private

Definition at line 1600 of file prog.cpp.

References _ASSERT, CNcbiMatrix< T >::GetCols(), CMultiAlignerOptions::GetDomainResFreqBoost(), CSequence::GetFreqs(), CSequence::GetLetter(), CClusterer::CSingleCluster::GetPrototype(), CNcbiMatrix< T >::GetRows(), i, int, ITERATE, CSequence::kGapChar, m_Options, m_QueryData, m_RPSLocs, offset, and prot.

Referenced by x_AlignProgressive().

◆ x_AlignClusterQueries()

unique_ptr< vector< int > > CMultiAligner::x_AlignClusterQueries ( const TPhyTreeNode * node )

private

Definition at line 357 of file blast.cpp.

References _ASSERT, CHitList::AddToHitList(), CSeq_align_Base::C_Segs::e_Dendiag, CSeq_align_Base::C_Segs::e_Denseg, CMultiAlignerOptions::GetBlastpEvalue(), CSeq_align_Base::C_Segs::GetDendiag(), CSeq_align_Base::C_Segs::GetDenseg(), CClusterer::GetDistMatrix(), CScore_Base::GetId(), CScore_Base::C_Value::GetInt(), CScore_Base::C_Value::GetReal(), CSearchResultSet::GetResults(), CSeq_align_Base::GetScore(), CDense_diag_Base::GetScores(), CSeq_align_Base::GetSegs(), CSearchResults::GetSeqAlign(), CObject_id_Base::GetStr(), CTreeNode< TValue, TKeyGetterP >::GetValue(), CScore_Base::GetValue(), i, CTreeNode< TValue, TKeyGetterP >::IsLeaf(), ITERATE, m_Clusterer, m_LocalInClusterHits, m_Options, m_Scope, m_tQueries, max(), result, CBl2Seq::RunEx(), CTreeNode< TValue, TKeyGetterP >::SubNodeBegin(), CTreeNode< TValue, TKeyGetterP >::SubNodeEnd(), and CSeq_align_Base::C_Segs::Which().

Referenced by x_FindLocalInClusterHits().

◆ x_AlignFillerBlocks()

void CMultiAligner::x_AlignFillerBlocks	(	const blast::TSeqLocVector &	queries,
		const vector< int > &	indices,
		vector< CRef< objects::CSeq_loc > > &	filler_locs,
		vector< SSegmentLoc > &	filler_segs
	)

private

Run blastp, aligning the collection of filler fragments against the entire input dataset.

Parameters

queries	List of queries selected for blastp alignment [in]
indices	List of indices of each selected query in the queries array [in]
filler_locs	List of generated sequences [in]
filler_segs	Simplified representation of filler_locs [in]

Definition at line 169 of file blast.cpp.

References CHitList::AddToHitList(), CSeq_align_Base::C_Segs::e_Dendiag, CSeq_align_Base::C_Segs::e_Denseg, eInterrupt, Get(), CMultiAlignerOptions::GetBlastpEvalue(), CSeq_align_Base::C_Segs::GetDendiag(), CSeq_align_Base::C_Segs::GetDenseg(), CSeq_interval_Base::GetFrom(), CScore_Base::GetId(), CScore_Base::C_Value::GetInt(), CSeq_loc_Base::GetInt(), CScore_Base::C_Value::GetReal(), CSeq_align_Base::GetScore(), CDense_diag_Base::GetScores(), CSeq_align_Base::GetSegs(), CObject_id_Base::GetStr(), CSeq_interval_Base::GetTo(), CScore_Base::GetValue(), i, int, ITERATE, m_Interrupt, m_LocalHits, m_Options, m_ProgressMonitor, m_Scope, max(), NCBI_THROW, CBl2Seq::Run(), and CSeq_align_Base::C_Segs::Which().

Referenced by x_FindLocalHits().

◆ x_AlignInClusters()

void CMultiAligner::x_AlignInClusters ( void )

protected

Pair-wise align each cluster sequence to cluster representative.

Definition at line 1009 of file cobalt.cpp.

References ASSERT, eInterrupt, ctll::empty(), CNWAligner::eTS_Delete, CNWAligner::eTS_Insert, CClusterer::GetClusters(), CMultiAlignerOptions::GetEndGapExtendPenalty(), CMultiAlignerOptions::GetEndGapOpenPenalty(), CMultiAlignerOptions::GetGapExtendPenalty(), CMultiAlignerOptions::GetGapOpenPenalty(), CSequence::GetLength(), CSequence::GetLetter(), CSequence::GetPrintableLetter(), CClusterer::CSingleCluster::GetPrototype(), CSequence::GetSequence(), CNWAligner::GetTranscript(), CMultiAlignerOptions::GetVerbose(), i, ITERATE, CSequence::kGapChar, m_Aligner, m_AllQueryData, m_Clusterer, m_ClusterGapPositions, m_Interrupt, m_Options, m_ProgressMonitor, m_Results, NCBI_THROW, CSequence::PropagateGaps(), CPSSMAligner::Run(), CNWAligner::SetEndSpaceFree(), CPSSMAligner::SetEndWg(), CPSSMAligner::SetEndWs(), CPSSMAligner::SetSequences(), CPSSMAligner::SetStartWg(), CPSSMAligner::SetStartWs(), CPSSMAligner::SetWg(), CPSSMAligner::SetWs(), ncbi::grid::netcache::search::fields::size, and t.

Referenced by x_Run().

◆ x_AlignMSAs()

void CMultiAligner::x_AlignMSAs ( void )

protected

Align multiple sequence alignments.

Definition at line 513 of file cobalt.cpp.

References _ASSERT, CSequence::CompressSequences(), CMultiAlignerOptions::eNone, CNcbiMatrix< T >::GetCols(), CHitList::GetHit(), CNcbiMatrix< T >::GetRows(), i, int, ITERATE, m_ClustAlnMethod, m_CombinedHits, m_InMSA1, m_InMSA2, m_QueryData, m_Results, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHitList::Size(), x_AlignProfileProfile(), x_CreateBlastQueries(), x_CreatePatternQueries(), x_FindConsistentHitSubset(), x_FindDomainHits(), x_FindLocalHits(), and x_FindPatternHits().

Referenced by x_Run().

◆ x_AlignProfileProfile()

void CMultiAligner::x_AlignProfileProfile	(	vector< CTree::STreeLeaf > &	node_list1,
		vector< CTree::STreeLeaf > &	node_list2,
		vector< CSequence > &	alignment,
		CNcbiMatrix< CHitList > &	pair_info,
		int	iteration
	)

private

Align two collections of sequences.

All sequences within a single collection begin with the same size

Parameters

node_list1	List of sequence number in first collection [in]
node_list2	List of sequence number in second collection [in]
alignment	Complete list of aligned sequences (may contain sequences that will not be aligned). On output, new gaps will be propagated to all affected sequences [in][out]
pair_info	Constraints that may be used in the alignment [in]
iteration	The iteration number [in]

Definition at line 793 of file prog.cpp.

References CNWAligner::eTS_Delete, CNWAligner::eTS_Insert, CMultiAlignerOptions::GetEndGapExtendPenalty(), CMultiAlignerOptions::GetEndGapOpenPenalty(), CMultiAlignerOptions::GetGapExtendPenalty(), CMultiAlignerOptions::GetGapOpenPenalty(), CNWAligner::GetTranscript(), CMultiAlignerOptions::GetVerbose(), i, int, kAlphabetSize, kRpsScaleFactor, kScale, m_Aligner, m_Options, query, CPSSMAligner::Run(), CNWAligner::SetEndSpaceFree(), CPSSMAligner::SetEndWg(), CPSSMAligner::SetEndWs(), CNWAligner::SetPattern(), CPSSMAligner::SetSequences(), CPSSMAligner::SetStartWg(), CPSSMAligner::SetStartWs(), CPSSMAligner::SetWg(), CPSSMAligner::SetWs(), t, x_FillResidueFrequencies(), x_FindConstraints(), and x_NormalizeResidueFrequencies().

Referenced by x_AlignMSAs(), x_AlignProfileProfileUsingHit(), x_AlignProgressive(), and x_RealignSequences().

◆ x_AlignProfileProfileUsingHit()

void CMultiAligner::x_AlignProfileProfileUsingHit	(	vector< CTree::STreeLeaf > &	node_list1,
		vector< CTree::STreeLeaf > &	node_list2,
		vector< CSequence > &	alignment,
		CNcbiMatrix< CHitList > &	pair_info,
		int	iteration
	)

private

Align two profiles with all sequences that belong to the same cluster.

A pair-wise constraint alignment between the most similar sequences is used for aligning the profiles. Sequence positions that match in the pair-wise alignment will also match in the profile alignment. Ranges between the matching positions are aligned with CPSSMAligner.

Definition at line 1056 of file prog.cpp.

References _ASSERT, COpenRange< Position >::Empty(), CNWAligner::eTS_Delete, CNWAligner::eTS_Insert, CNWAligner::eTS_Match, fReduceBoth, fReduceLeft, fReduceRight, CMultiAlignerOptions::GetFastAlign(), GetLength(), COpenRange< Position >::GetLength(), CMultiAlignerOptions::GetVerbose(), i, int, NStr::IntToString(), ITERATE, m_Messages, m_Options, COpenRange< Position >::NotEmpty(), query, t, x_AlignProfileProfile(), x_ComputeProfileRangeAlignment(), and x_FindInClusterConstraints().

Referenced by x_AlignProgressive().

◆ x_AlignProgressive()

void CMultiAligner::x_AlignProgressive	(	const TPhyTreeNode *	tree,
		vector< CSequence > &	query_data,
		CNcbiMatrix< CHitList > &	pair_info,
		int	iteration,
		bool	is_cluster
	)

private

Main driver for progressive alignment.

Parameters

tree	Alignment guide tree [in]
query_data	The sequences to align. The first call assumes this list contains the unaligned sequences, and intermediate alignment stages will update the list as alignment progresses. On output query_data contains the aligned version of all sequences [in][out]
pair_info	List of alignment constraints [in]
iteration	The iteration number [in]
is_cluster	Is the curretly traversed node inside a cluster subtree [in]

Definition at line 1534 of file prog.cpp.

References eInterrupt, CHitList::Empty(), CClusterer::GetClusters(), CTreeNode< TValue, TKeyGetterP >::GetValue(), CTreeNode< TValue, TKeyGetterP >::IsLeaf(), kClusterNodeId, CTree::ListTreeLeaves(), m_Clusterer, m_DomainHits, m_Interrupt, m_ProgressMonitor, NCBI_THROW, x_AddRpsFreqsToCluster(), x_AlignProfileProfile(), x_AlignProfileProfileUsingHit(), and x_GetClusterGapLocations().

Referenced by x_BuildAlignmentIterative().

◆ x_AssignDefaultResFreqs()

void CMultiAligner::x_AssignDefaultResFreqs ( )

private

Definition at line 603 of file rps.cpp.

References Blast_ResFreqFree(), Blast_ResFreqNew(), Blast_ResFreqStdComp(), BLASTAA_SEQ_CODE, BlastScoreBlkFree(), BlastScoreBlkNew(), eInterrupt, CMultiAlignerOptions::eToPrototype, CMultiAlignerOptions::GetLocalResFreqBoost(), i, kAlphabetSize, m_AllQueryData, m_ClustAlnMethod, m_Interrupt, m_Options, m_ProgressMonitor, m_QueryData, NCBI_THROW, Blast_ResFreq::prob, query, and x_MakeClusterResidueFrequencies().

Referenced by x_FindDomainHits(), and x_FindLocalHits().

◆ x_AssignRPSResFreqs()

void CMultiAligner::x_AssignRPSResFreqs	(	CHitList &	rps_hits,
		CProfileData &	profile_data
	)

private

◆ x_AttachClusterTrees()

void CMultiAligner::x_AttachClusterTrees	(	const vector< TPhyTreeNode * > &	cluster_trees,
		const vector< TPhyTreeNode * > &	cluster_leaves
	)

protected

Replace leaves in the alignment guide tree of clusters with cluster trees.

Parameters

cluster_trees	List of phylogenetic trees computed for each cluster of input sequences [in]
cluster_leaves	List of pointers to leaves of the alignment guide tree computed for clusters [in]

Definition at line 1790 of file cobalt.cpp.

References _ASSERT, CTreeNode< TValue, TKeyGetterP >::AddNode(), CTreeNode< TValue, TKeyGetterP >::DetachNode(), CClusterer::GetClusters(), CTreeNode< TValue, TKeyGetterP >::GetValue(), NStr::IntToString(), CTreeNode< TValue, TKeyGetterP >::IsLeaf(), ITERATE, kClusterNodeId, m_Clusterer, NULL, CTreeNode< TValue, TKeyGetterP >::SubNodeBegin(), and CTreeNode< TValue, TKeyGetterP >::SubNodeEnd().

Referenced by x_BuildFullTree().

◆ x_BuildAlignment()

void CMultiAligner::x_BuildAlignment ( )

protected

Given the current domain, local, pattern and user hits, along with the current tree, compute a multiple alignment of the input sequences.

Intended for applications that want fine-grained control of the alignment process

Definition at line 2010 of file prog.cpp.

References _ASSERT, eProgressiveAlignment, GetTree(), CMultiAlignerOptions::GetVerbose(), i, INT4_MAX, kClusterNodeId, CTree::ListTreeEdges(), m_Options, m_ProgressMonitor, m_QueryData, min(), ct::sort(), CMultiAligner::SProgress::stage, and x_BuildAlignmentIterative().

Referenced by x_Run().

◆ x_BuildAlignmentIterative()

void CMultiAligner::x_BuildAlignmentIterative	(	vector< CTree::STreeEdge > &	edges,
		double	cluster_cutoff
	)

private

Main driver for the progressive alignment process.

Parameters

edges	List of tree edges sorted by decreasing edge length [in]
cluster_cutoff	The minimum distance a tree edge must have to be the subject of a bipartition [in]

Definition at line 1812 of file prog.cpp.

References eIterativeAlignment, CMultiAlignerOptions::eMulti, eOutOfMemory, fabs, CHitList::GetHit(), CMultiAlignerOptions::GetIterate(), GetLength(), COpenRange< Position >::GetLength(), CMultiAlignerOptions::GetRealign(), GetTree(), CMultiAlignerOptions::GetVerbose(), i, int, INT4_MIN, m_ClustAlnMethod, m_CombinedHits, m_LocalInClusterHits, m_Options, m_PatternHits, m_ProgressMonitor, m_QueryData, m_Results, m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, m_UserHits, max(), NCBI_THROW, CHitList::PurgeAllHits(), s_CleanUpConstraints(), CHitList::Size(), CMultiAligner::SProgress::stage, x_AlignProgressive(), x_FindConservedColumns(), x_GetScore(), and x_RealignSequences().

Referenced by x_BuildAlignment().

◆ x_BuildFullTree()

void CMultiAligner::x_BuildFullTree ( const vector< TPhyTreeNode * > & cluster_trees )

protected

Combine alignment guide tree computed for clusters with guide trees computed for each cluster.

Leaves of the guide tree computed for cluster prototypes are replaced with clusters trees. Cluster trees are rescaled so that distance from root to cluster prototype in cluster tree is the same as length of the leaf edge in the alignment guide tree (for cluster prototypes).

Parameters

cluster_trees List of cluster trees [in]

Definition at line 1844 of file cobalt.cpp.

References _ASSERT, CClusterer::GetClusters(), CClusterer::CSingleCluster::GetPrototype(), CTree::GetTree(), CMultiAlignerOptions::GetVerbose(), i, m_Clusterer, m_Options, m_QueryData, m_Tree, NULL, CTree::PrintTree(), s_FindLeafDistances(), s_RescaleTree(), and x_AttachClusterTrees().

Referenced by x_Run().

◆ x_ComputeClusterTrees()

void CMultiAligner::x_ComputeClusterTrees ( vector< TPhyTreeNode * > & trees )

protected

Compute independent phylogenetic trees each cluster.

Parameters

Array of tree roots. Each array element corresponds to the cluster. One element clusters yield NULL tree. [out]

Definition at line 1593 of file cobalt.cpp.

References _ASSERT, CMultiAlignerOptions::eClusters, CMultiAlignerOptions::eFastME, CClusterer::GetClusterDistMatrix(), CClusterer::GetClusters(), CClusterer::GetDistMatrix(), CMultiAlignerOptions::GetTreeMethod(), CMultiAlignerOptions::GetVerbose(), i, int, m_Clusterer, m_Options, NULL, CTree::PrintTree(), CTree::ReleaseTree(), CClusterer::ReleaseTrees(), s_MakeTwoLeafTree(), s_SetLeafIds(), CClusterer::CSingleCluster::size(), and ncbi::grid::netcache::search::fields::size.

Referenced by x_Run().

◆ x_ComputeProfileRangeAlignment()

void CMultiAligner::x_ComputeProfileRangeAlignment	(	vector< CTree::STreeLeaf > &	node_list1,
		vector< CTree::STreeLeaf > &	node_list2,
		vector< CSequence > &	alignment,
		vector< size_t > &	constraints,
		const TRange &	range1,
		const TRange &	range2,
		int	full_prof_len1,
		int	full_prof_len2,
		CMultiAligner::EEndGapCostStrategy	strat,
		CNWAligner::TTranscript &	t
	)

private

Compute profile profile alignmnet for a ranges of given profiles.

Resambles x_AlignProfileProfile, but works on sequence ranges, does not allow end space free for large sequence lengths difference, returns transcript, and does not update the vector of input sequences.

Parameters

node_list1	List of sequence numbers in first collecton [in]
node_list2	List of sequence numbers in second collection [in]
alignment	List of sequences [in]
constraints	Constraints for alignment [in]
range1	Range for alignment of the first profile [in]
range2	Range for alignment of the second profile [in]
t	Alignmet transcript [out]

Definition at line 964 of file prog.cpp.

References fReduceLeft, fReduceRight, CMultiAlignerOptions::GetEndGapExtendPenalty(), CMultiAlignerOptions::GetEndGapOpenPenalty(), CRange_Base::GetFrom(), CMultiAlignerOptions::GetGapExtendPenalty(), CMultiAlignerOptions::GetGapOpenPenalty(), CRange_Base::GetTo(), CNWAligner::GetTranscript(), i, kAlphabetSize, kRpsScaleFactor, kScale, m_Aligner, m_Options, CPSSMAligner::Run(), CNWAligner::SetEndSpaceFree(), CPSSMAligner::SetEndWg(), CPSSMAligner::SetEndWs(), CNWAligner::SetPattern(), CPSSMAligner::SetSequences(), CPSSMAligner::SetStartWg(), CPSSMAligner::SetStartWs(), CPSSMAligner::SetWg(), CPSSMAligner::SetWs(), t, x_FillResidueFrequencies(), and x_NormalizeResidueFrequencies().

Referenced by x_AlignProfileProfileUsingHit().

◆ x_ComputeTree()

void CMultiAligner::x_ComputeTree ( void )

protected

Given the current list of domain and local hits, generate a phylogenetic tree that clusters the current input sequences.

Intended for applications that want fine-grained control of the alignment process

Definition at line 414 of file cobalt.cpp.

References _ASSERT, Blast_KarlinBlkGappedLoadFromTables(), CClusterer::ComputeClusters(), CTree::ComputeTree(), CMultiAlignerOptions::eClusters, CClusterer::eCompleteLinkage, CMultiAlignerOptions::eFastME, eInterrupt, CMultiAlignerOptions::eMulti, eTreeComputation, CClusterer::GetClusters(), CNcbiMatrix< T >::GetCols(), CPSSMAligner::GetMatrix(), CDistances::GetMatrix(), CNcbiMatrix< T >::GetRows(), CMultiAlignerOptions::GetScoreMatrixName(), CTree::GetTree(), CMultiAlignerOptions::GetTreeMethod(), CMultiAlignerOptions::GetVerbose(), i, int, m_Aligner, m_ClustAlnMethod, m_Clusterer, m_CombinedHits, m_Interrupt, m_Options, m_ProgressMonitor, m_QueryData, m_Tree, NCBI_THROW, CTree::PrintTree(), CClusterer::ReleaseTree(), CNcbiMatrix< T >::Resize(), CTree::SetTree(), and CMultiAligner::SProgress::stage.

Referenced by x_Run().

◆ x_CreateBlastQueries()

void CMultiAligner::x_CreateBlastQueries	(	blast::TSeqLocVector &	queries,
		vector< int > &	indices
	)

protected

Create query set for RPS Blast and Blastp searches along with indices in multiple alignment queries array.

Searches for conserved regions can be performed for a subset of multiple alignment queries (typically to reduce computation time).

Parameters

queries	Queries for RPS Blast and Blastp searches [out]
indices	Indexes of each query in m_tQueries [out]

Definition at line 1433 of file cobalt.cpp.

References _ASSERT, CMultiAlignerOptions::eMulti, CMultiAlignerOptions::eNone, CMultiAlignerOptions::eToPrototype, CClusterer::GetClusters(), i, int, ITERATE, m_ClustAlnMethod, m_Clusterer, m_InMSA1, m_InMSA2, m_Msa1Repr, m_Msa2Repr, m_Scope, m_tQueries, and NCBI_THROW.

Referenced by CMultiAlignerTest::SetDomainHits(), x_AlignMSAs(), and x_Run().

◆ x_CreatePatternQueries()

void CMultiAligner::x_CreatePatternQueries	(	vector< const CSequence * > &	queries,
		vector< int > &	indices
	)

protected

Create query set for PROSITE pattern search along with indices in multiple alignment queries array.

Searches for conserved regions can be performed for a subset of multiple alignment queries (typically to reduce computation time).

Parameters

queries	Queries for PROSITE patterns searches [out]
indices	Indexes of each query in m_tQueries [out]

Definition at line 1493 of file cobalt.cpp.

References CMultiAlignerOptions::eMulti, CMultiAlignerOptions::eNone, CMultiAlignerOptions::eToPrototype, CClusterer::GetClusters(), i, int, m_ClustAlnMethod, m_Clusterer, m_QueryData, and NCBI_THROW.

Referenced by x_AlignMSAs(), and x_Run().

◆ x_FindAlignmentSubsets()

void CMultiAligner::x_FindAlignmentSubsets ( )

private

Definition at line 120 of file seg.cpp.

References CHitList::GetHit(), i, CMultiAligner::SGraphNode::list_pos, m_CombinedHits, m_QueryData, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, NULL, CMultiAligner::SGraphNode::path_next, CHitList::PurgeUnwantedHits(), CHitList::SetKeepHit(), CHitList::Size(), and x_FindBestPath().

Referenced by x_FindConsistentHitSubset().

◆ x_FindBestPath()

CMultiAligner::SGraphNode * CMultiAligner::x_FindBestPath ( vector< SGraphNode > & nodes )

private

Find a maximum weight path in a directed acyclic graph.

Parameters

nodes The graph [in/modified]

Returns: Pointer to the first node in the optimal path

Definition at line 52 of file seg.cpp.

References i, INT4_MIN, CHit::m_SeqRange1, CHit::m_SeqRange2, NULL, and CLocalRange< Position >::StrictlyBelow().

Referenced by x_FindAlignmentSubsets(), and x_FindConstraints().

◆ x_FindConservedColumns()

void CMultiAligner::x_FindConservedColumns	(	vector< CSequence > &	new_alignment,
		CHitList &	conserved
	)

private

Create a list of constraints that reflect conserved columns in a multiple alignment.

Parameters

new_alignment	The multiple alignment to analyze [in]
conserved	The list of pairwise constraints [out]

Definition at line 1677 of file prog.cpp.

References CHitList::AddToHitList(), Empty(), CMultiAlignerOptions::GetConservedCutoffScore(), CRange_Base::GetFrom(), GetLength(), COpenRange< Position >::GetLength(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), i, int, CSequence::kGapChar, m_Options, and x_GetScoreOneCol().

Referenced by x_BuildAlignmentIterative().

◆ x_FindConsistentHitSubset()

void CMultiAligner::x_FindConsistentHitSubset ( void )

protected

Find consistent subset of pair-wise hits that can be used as alignment constraints.

Definition at line 175 of file seg.cpp.

References CRange_Base::GetFrom(), CHitList::GetHit(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), i, m_CombinedHits, m_Options, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, CHitList::MakeCanonical(), CHitList::Size(), and x_FindAlignmentSubsets().

Referenced by x_AlignMSAs(), and x_Run().

◆ x_FindConstraints()

void CMultiAligner::x_FindConstraints	(	vector< size_t > &	constraint,
		vector< CSequence > &	alignment,
		vector< CTree::STreeLeaf > &	node_list1,
		vector< CTree::STreeLeaf > &	node_list2,
		CNcbiMatrix< CHitList > &	pair_info,
		int	iteration
	)

private

Find the set of constraints to use for a profile-profile alignment.

This routine considers only constraints between one collection of equal-size sequences and another collection of equal-size sequences

Parameters

constraint	List of compute constraints, in the format expected by CPSSMAligner [out]
alignment	Current multiple alignment of sequences [in]
node_list1	List of sequences in first collection [in]
node_list2	List of sequences in second collection [in]
pair_info	List of pairwise constraints (between sequences) [in]
iteration	The iteration number [in]

Definition at line 382 of file prog.cpp.

References _ASSERT, abs, CHitList::AddToHitList(), eInterrupt, CHitList::Empty(), CRange_Base::GetFrom(), CHitList::GetHit(), COpenRange< Position >::GetLength(), CHit::GetSubHit(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), CHit::HasSubHits(), CMultiAligner::SGraphNode::hit, i, CHit::InsertSubHit(), int, m_Interrupt, m_Options, m_ProgressMonitor, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, NCBI_THROW, NON_CONST_ITERATE, CMultiAligner::SGraphNode::path_next, CHitList::Size(), CHitList::SortByStartOffset(), CLocalRange< Position >::StrictlyBelow(), x_ExpandRange(), x_FindBestPath(), and x_HitToConstraints().

Referenced by x_AlignProfileProfile().

◆ x_FindDomainHits()

void CMultiAligner::x_FindDomainHits	(	blast::TSeqLocVector &	queries,
		const vector< int > &	indices
	)

protected

Run RPS blast on seletced input sequences and postprocess the results.

Intended for applications that want fine-grained control of the alignment process.

Parameters

queries	Queries for RPS Blast search [in]
indices	Index of each RPS Blast query in the array of input sequences [in]

Definition at line 866 of file rps.cpp.

References _ASSERT, CMultiAlignerOptions::CanGetDomainHits(), CProfileData::Clear(), eDomainHitsSearch, CProfileData::eGetPssm, CProfileData::eGetResFreqs, eInterrupt, CHitList::Empty(), CMultiAlignerOptions::GetDomainHits(), CRange_Base::GetFrom(), CHitList::GetHit(), CMultiAlignerOptions::GetRpsDb(), CHit::GetSubHit(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), i, kRpsScaleFactor, CProfileData::Load(), m_CombinedHits, m_DomainHits, m_Interrupt, m_IsDomainSearched, m_Options, m_ProgressMonitor, m_RPSLocs, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, m_tQueries, CHitList::MatchOverlappingSubHits(), NCBI_THROW, NON_CONST_ITERATE, CHitList::PurgeAllHits(), CHitList::Size(), CMultiAligner::SProgress::stage, x_AssignDefaultResFreqs(), x_AssignRPSResFreqs(), x_FindRPSHits(), x_LoadBlockBoundaries(), x_RealignBlocks(), and x_SetDomainHits().

Referenced by x_AlignMSAs(), and x_Run().

◆ x_FindInClusterConstraints()

void CMultiAligner::x_FindInClusterConstraints	(	vector< CSequence > &	alignment,
		vector< CTree::STreeLeaf > &	node_list1,
		vector< CTree::STreeLeaf > &	node_list2,
		CNcbiMatrix< CHitList > &	pair_info,
		vector< TRangePair > &	match_ranges
	)		const

private

Find constraint to use for profile to profile alignment in clusters.

Finds in-cluster constraint (blastp hit) between two most similar sequences from each profile.

Parameters

alignment	Current multiple alignment of sequences [in]
node_list1	List of sequences in first collection [in]
node_list2	List of sequences in second collection [in]
pair_info	List of pairwise constraints (between sequences) [in]
match_ranges	A list of pairs of ungapped blastp alignment ranges on the profiles [out]

Definition at line 700 of file prog.cpp.

References _ASSERT, first(), CClusterer::GetDistMatrix(), CHit::GetEditScript(), CRange_Base::GetFrom(), CMultiAlignerOptions::GetVerbose(), i, CEditScript::ListMatchRegions(), m_Clusterer, m_Options, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, and x_GetProfileMatchRanges().

Referenced by x_AlignProfileProfileUsingHit().

◆ x_FindLocalHits()

void CMultiAligner::x_FindLocalHits	(	const blast::TSeqLocVector &	queries,
		const vector< int > &	indices
	)

protected

Run blast on selected input sequences and postprocess the results.

Intended for applications that want fine-grained control of the alignment process

Parameters

queries	Queries for Blast alignment [in]
indices	Index of each Blast query in the array of input sequences [in]

Definition at line 306 of file blast.cpp.

References CHitList::Append(), eLocalHitsSearch, CHitList::Empty(), CRange_Base::GetFrom(), CHitList::GetHit(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), i, m_CombinedHits, m_DomainHits, m_LocalHits, m_Options, m_ProgressMonitor, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, CHitList::PurgeAllHits(), CHitList::Size(), CMultiAligner::SProgress::stage, x_AlignFillerBlocks(), x_AssignDefaultResFreqs(), and x_MakeFillerBlocks().

Referenced by x_AlignMSAs(), and x_Run().

◆ x_FindLocalInClusterHits()

void CMultiAligner::x_FindLocalInClusterHits ( const vector< TPhyTreeNode * > & cluster_trees )

protected

Run blast on sequences from each cluster subtree.

Each tree is traversed. Blast is run for pairs of most similar sequences such that each belongs to different subtree. Indented for use with query clustering for faster in-cluster pair-wise alignments.

Parameters

cluster_trees List of query cluster trees [in]

Definition at line 481 of file blast.cpp.

References CRange_Base::GetFrom(), CHitList::GetHit(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), i, ITERATE, m_LocalInClusterHits, m_Options, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, CHitList::PurgeAllHits(), CHitList::Size(), and x_AlignClusterQueries().

Referenced by x_Run().

◆ x_FindPatternHits()

void CMultiAligner::x_FindPatternHits	(	const vector< const CSequence * > &	queries,
		const vector< int > &	indices
	)

protected

Find PROSITE pattern hits on selected input sequences.

Intended for applications that want fine-grained control of the alignment process

Parameters

queries	Queries for PROSITE patter search [in]
indices	Index of each PROSITE pattern search query in the array of input sequences [in]

Definition at line 62 of file phi.cpp.

References _ASSERT, CHitList::AddToHitList(), BLASTAA_SEQ_CODE, BlastScoreBlkFree(), BlastScoreBlkNew(), eInterrupt, ePatternHitsSearch, FALSE, FindPatternHits(), CMultiAlignerOptions::GetCddPatterns(), CRange_Base::GetFrom(), CHitList::GetHit(), GetLength(), GetSequence(), CRange_Base::GetTo(), CMultiAlignerOptions::GetVerbose(), i, int, m_Interrupt, m_Options, m_PatternHits, m_ProgressMonitor, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, NCBI_THROW, NULL, patterns, PHI_MAX_HIT, CHitList::PurgeAllHits(), CHitList::Size(), SPHIPatternSearchBlkFree(), SPHIPatternSearchBlkNew(), and CMultiAligner::SProgress::stage.

Referenced by x_AlignMSAs(), and x_Run().

◆ x_FindQueryClusters()

bool CMultiAligner::x_FindQueryClusters ( )

protected

Find clusters of similar queries, select cluster representative sequences, and prepare input to multiple alignement composed of only representatives.

Returns: True if at least one cluster was found, false otherwise

Definition at line 744 of file cobalt.cpp.

Referenced by x_Run().

◆ x_FindRPSHits()

void CMultiAligner::x_FindRPSHits	(	blast::TSeqLocVector &	queries,
		const vector< int > &	indices,
		CHitList &	rps_hits
	)

private

Definition at line 411 of file rps.cpp.

References _ASSERT, CHitList::AddToHitList(), CBlastOptionsFactory::Create(), CSearchDatabase::eBlastDbIsProtein, eInterrupt, CSeqDB::eProtein, eRPSBlast, Get(), CSeq_align_Base::C_Segs::GetDenseg(), CMultiAlignerOptions::GetDomainHitlistSize(), CRange_Base::GetFrom(), CHitList::GetHit(), CScore_Base::GetId(), CDense_seg_Base::GetIds(), CScore_Base::C_Value::GetInt(), CRef< C, Locker >::GetNonNullPointer(), CScore_Base::C_Value::GetReal(), CMultiAlignerOptions::GetRpsDb(), CMultiAlignerOptions::GetRpsEvalue(), CSeq_align_Base::GetScore(), CSeq_align_Base::GetSegs(), CObject_id_Base::GetStr(), CRange_Base::GetTo(), CScore_Base::GetValue(), CMultiAlignerOptions::GetVerbose(), i, ITERATE, m_Interrupt, m_Options, m_ProgressMonitor, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, max(), NCBI_THROW, results, CLocalBlast::Run(), CSeqDB::SeqidToOid(), CBlastOptionsHandle::SetEvalueThreshold(), CBlastOptionsHandle::SetFilterString(), CBlastOptionsHandle::SetHitlistSize(), and CHitList::Size().

Referenced by x_FindDomainHits().

◆ x_GetScore()

double CMultiAligner::x_GetScore ( vector< CSequence > & align )

private

Compute the entropy score of a multiple alignment.

Parameters

align complete multiple alignment [in]

Returns: the alignment score

Definition at line 1392 of file prog.cpp.

References H, i, and x_GetScoreOneCol().

Referenced by x_BuildAlignmentIterative(), and x_RealignSequences().

◆ x_GetScoreOneCol()

double CMultiAligner::x_GetScoreOneCol	(	vector< CSequence > &	align,
		int	col
	)

private

Calculate the entropy score of one column of a multiple alignment (see the COBALT papaer for details)

Parameters

align	The alignment [in]
col	the column number to score

Returns: the computed score

Definition at line 1351 of file prog.cpp.

References count, CMultiAlignerOptions::GetPseudocount(), kDerivedFreqs, kNumClasses, kRes2Class, log, m_Options, and r().

Referenced by x_FindConservedColumns(), and x_GetScore().

◆ x_GetSeqalign()

CRef< CSeq_align > CMultiAligner::x_GetSeqalign	(	const vector< CSequence > &	align,
		vector< int > &	indices
	)		const

private

Produce a seqalign representing the specified alignment, using a subset of the sequences.

Parameters

align	The alignment to convert
indices	0-based list of sequence numbers that will participate in the resulting Seq-align. indices may appear out of order and may be repeated

Returns: The generated seqalign

Definition at line 58 of file seqalign.cpp.

References CSeq_align_Base::eType_global, CSeq_interval_Base::GetFrom(), CSeq_interval_Base::GetId(), CSeq_loc_Base::GetInt(), CSeq_loc_Base::GetWhole(), i, CSeq_loc_Base::IsInt(), CSeq_loc_Base::IsWhole(), CSequence::kGapChar, CDense_seg_Base::SetDim(), CSeq_align_Base::SetDim(), CDense_seg_Base::SetIds(), CDense_seg_Base::SetLens(), CDense_seg_Base::SetNumseg(), CSeq_align_Base::SetSegs(), CDense_seg_Base::SetStarts(), and CSeq_align_Base::SetType().

Referenced by GetResults().

◆ x_Init()

void CMultiAligner::x_Init ( void )

inlineprivate

Initiate class attributes that are not alignment parameters.

Definition at line 579 of file cobalt.hpp.

References NULL.

Referenced by CMultiAligner().

◆ x_InitAligner()

void CMultiAligner::x_InitAligner ( void )

private

Initiate PSSM aligner parameters.

Definition at line 102 of file cobalt.cpp.

References CMultiAlignerOptions::GetEndGapExtendPenalty(), CMultiAlignerOptions::GetEndGapOpenPenalty(), CMultiAlignerOptions::GetGapExtendPenalty(), CMultiAlignerOptions::GetGapOpenPenalty(), CMultiAlignerOptions::GetScoreMatrixName(), m_Aligner, m_Options, CPSSMAligner::SetEndWg(), CPSSMAligner::SetEndWs(), CPSSMAligner::SetStartWg(), CPSSMAligner::SetStartWs(), CPSSMAligner::SetWg(), CPSSMAligner::SetWs(), and x_SetScoreMatrix().

Referenced by CMultiAligner().

◆ x_InitColumn()

void CMultiAligner::x_InitColumn	(	vector< SColumn >::iterator &	it,
		size_t	len
	)

staticprivate

Definition at line 1160 of file cobalt.cpp.

References i, and len.

Referenced by x_MultiAlignClusters().

◆ x_InitInsertColumn()

void CMultiAligner::x_InitInsertColumn	(	vector< SColumn >::iterator &	it,
		size_t	len,
		int	num,
		int	cluster
	)

staticprivate

Definition at line 1175 of file cobalt.cpp.

References i, and len.

Referenced by x_MultiAlignClusters().

◆ x_InitParams()

void CMultiAligner::x_InitParams ( void )

private

Initiate parameters using m_Options.

Definition at line 79 of file cobalt.cpp.

References _ASSERT, CHitList::AddToHitList(), CConstRef< C, Locker >::Empty(), CMultiAlignerOptions::eNone, CMultiAlignerOptions::GetInClustAlnMethod(), CMultiAlignerOptions::GetUseQueryClusters(), CMultiAlignerOptions::GetUserConstraints(), CMultiAlignerOptions::GetUserConstraintsScore(), ITERATE, m_ClustAlnMethod, m_Options, m_UserHits, and CHitList::PurgeAllHits().

Referenced by CMultiAligner().

◆ x_LoadBlockBoundaries()

void CMultiAligner::x_LoadBlockBoundaries	(	string	blockfile,
		vector< SSegmentLoc > &	blocklist
	)

private

Given an RPS blast database, load a list of block offsets for each database sequence.

The list is resident in a text file, where each line is as follows

[seq ID] [oid of block] [start block offset] [end block offset]

Note that block offsets are zero-based

Parameters

blockfile	Name of file containing list of offsets [in]
blocklist	the list of offsets read from file [out]

Definition at line 75 of file rps.cpp.

References buf, NCBI_THROW, and tmp.

Referenced by x_FindDomainHits().

◆ x_MakeClusterResidueFrequencies()

void CMultiAligner::x_MakeClusterResidueFrequencies ( void )

protected

Compute profile residue frequencies for clusters.

Frequencies are not normalized.

Definition at line 1377 of file cobalt.cpp.

References _ASSERT, eInterrupt, CClusterer::GetClusters(), CSequence::GetFreqs(), CSequence::GetLength(), CClusterer::CSingleCluster::GetPrototype(), CNcbiMatrix< T >::GetRows(), i, ITERATE, kAlphabetSize, m_AllQueryData, m_Clusterer, m_ClusterGapPositions, m_Interrupt, m_ProgressMonitor, m_QueryData, NCBI_THROW, offset, CClusterer::CSingleCluster::size(), and ncbi::grid::netcache::search::fields::size.

Referenced by x_AssignDefaultResFreqs().

◆ x_MakeFillerBlocks()

void CMultiAligner::x_MakeFillerBlocks	(	const vector< int > &	blastp_indices,
		vector< CRef< objects::CSeq_loc > > &	filler_locs,
		vector< SSegmentLoc > &	filler_segs
	)

private

Turn all fragments of selected query sequence not already covered by a domain hit into a separate query sequence, used as input to a blast search.

Parameters

blastp_indices	Indices of query sequences selected for blastp search [in]
filler_locs	List of generated sequences [out]
filler_segs	Simplified representation of filler_locs [out]

Definition at line 95 of file blast.cpp.

References _ASSERT, CHitList::GetHit(), GetLength(), CHit::GetSubHit(), CMultiAlignerOptions::GetVerbose(), CHit::HasSubHits(), i, int, ITERATE, CHit::kMinHitSize, m_CombinedHits, m_Options, m_QueryData, m_Scope, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, m_tQueries, CHitList::Size(), and x_AddNewSegment().

Referenced by x_FindLocalHits().

◆ x_MultiAlignClusters()

void CMultiAligner::x_MultiAlignClusters ( void )

protected

Combine pair-wise in-cluster alignements with multiple alignments of cluster prototypes.

Definition at line 1188 of file cobalt.cpp.

References columns, eInterrupt, CClusterer::GetClusters(), CSequence::GetLength(), CSequence::GetPrintableLetter(), CMultiAlignerOptions::GetVerbose(), i, int, ITERATE, CSequence::kGapChar, letter(), m_AllQueries, m_AllQueryData, m_Clusterer, m_ClusterGapPositions, m_Interrupt, m_Options, m_ProgressMonitor, m_Results, m_tQueries, NCBI_THROW, NON_CONST_ITERATE, offset, results, ncbi::grid::netcache::search::fields::size, x_InitColumn(), and x_InitInsertColumn().

Referenced by x_Run().

◆ x_RealignBlocks()

void CMultiAligner::x_RealignBlocks	(	CHitList &	rps_hits,
		vector< SSegmentLoc > &	blocklist,
		CProfileData &	profile_data
	)

private

Definition at line 114 of file rps.cpp.

References _ASSERT, CHit::AddUpSubHits(), eInterrupt, COpenRange< Position >::Empty(), CNWAligner::eTS_Delete, CNWAligner::eTS_Insert, CNWAligner::eTS_Match, CHit::GetEditScript(), CMultiAlignerOptions::GetEndGapExtendPenalty(), CMultiAlignerOptions::GetEndGapOpenPenalty(), CPSSMAligner::GetEndWg(), CMultiAligner::SSegmentLoc::GetFrom(), CRange_Base::GetFrom(), CMultiAlignerOptions::GetGapExtendPenalty(), CMultiAlignerOptions::GetGapOpenPenalty(), CHitList::GetHit(), COpenRange< Position >::GetLength(), CProfileData::GetPssm(), CHit::GetRangeFromSeq2(), CEditScript::GetScore(), CProfileData::GetSeqOffsets(), CMultiAligner::SSegmentLoc::GetTo(), CRange_Base::GetTo(), CNWAligner::GetTranscript(), CMultiAlignerOptions::GetVerbose(), CNWAligner::GetWg(), CNWAligner::GetWs(), CHit::HasSubHits(), i, CHit::InsertSubHit(), CHit::kMinHitSize, kRpsScaleFactor, m_Aligner, m_Interrupt, m_Options, m_ProgressMonitor, m_QueryData, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, CEditScript::MakeEditScript(), max(), min(), NCBI_THROW, CHitList::PurgeUnwantedHits(), query, CMultiAligner::SSegmentLoc::range, CHit::ResolveSubHitConflicts(), CPSSMAligner::Run(), CMultiAligner::SSegmentLoc::seq_index, COpenRange< Position >::Set(), CNWAligner::SetEndSpaceFree(), CPSSMAligner::SetEndWg(), CPSSMAligner::SetEndWs(), CRange_Base::SetFrom(), CHitList::SetKeepHit(), CPSSMAligner::SetSequences(), CPSSMAligner::SetStartWg(), CPSSMAligner::SetStartWs(), CRange_Base::SetTo(), CPSSMAligner::SetWg(), CPSSMAligner::SetWs(), and CHitList::Size().

Referenced by x_FindDomainHits().

◆ x_RealignSequences()

double CMultiAligner::x_RealignSequences	(	const TPhyTreeNode *	input_cluster,
		vector< CSequence > &	alignment,
		CNcbiMatrix< CHitList > &	pair_info,
		double	score,
		int	iteration
	)

private

Perform a single bipartition on a multiple alignment.

Parameters

input_cluster	A tree describing sequences that form one half of the bipartition [in]
alignment	A complete multiple alignment, updated with the the bipartition results if they have a higher score [in][out]
pair_info	Pairwise constraints on the alignment [in]
score	Starting alignment score [in]
iteration	The iteration number [in]

Returns: The score of the new multiple alignment

Definition at line 1415 of file prog.cpp.

References CSequence::CompressSequences(), GetTree(), CMultiAlignerOptions::GetVerbose(), i, int, CTree::ListTreeLeaves(), m_Options, m_QueryData, x_AlignProfileProfile(), and x_GetScore().

Referenced by x_BuildAlignmentIterative().

◆ x_Run()

void CMultiAligner::x_Run ( void )

protectedvirtual

Align the current set of input sequences (reset any existing alignment information).

This function handles the generation of all internal state in the correct order. It is sufficient for 'black box' applications that only want a final answer without tweaking internal state.

Definition at line 590 of file cobalt.cpp.

References CMultiAlignerOptions::eMulti, CMultiAlignerOptions::eNone, CMultiAlignerOptions::eToPrototype, CClusterer::GetClusters(), NStr::IntToString(), kClusterNodeId, m_AllQueries, m_ClustAlnMethod, m_Clusterer, m_InMSA1, m_tQueries, m_Tree, NCBI_THROW, CTree::SetTree(), x_AlignInClusters(), x_AlignMSAs(), x_BuildAlignment(), x_BuildFullTree(), x_ComputeClusterTrees(), x_ComputeTree(), x_CreateBlastQueries(), x_CreatePatternQueries(), x_FindConsistentHitSubset(), x_FindDomainHits(), x_FindLocalHits(), x_FindLocalInClusterHits(), x_FindPatternHits(), x_FindQueryClusters(), and x_MultiAlignClusters().

Referenced by Run().

◆ x_SetDomainHits()

void CMultiAligner::x_SetDomainHits	(	const blast::TSeqLocVector &	queruies,
		const vector< int > &	indices,
		const objects::CBlast4_archive &	archive
	)

protected

Set pre-computed domain hits using BLAST archive format.

Parameters

queries	Sequences for which alignment hits will be searched for in the pre-computed set [in]
indices	Index of each query in the array of sequences to align [in]
archive	BLAST archive format [in]

Queries from the archive are matched to COBALT queries by Seq_ids. The two sets of queries do not need to be the same. Domain hits for the archive queries that do match to any of the COBALT queries are ignored. COBALT will do RPS-BLAST search for all of its sequences that were not matched to the archive queries. It is the responsibility of user to ensure that the pre-computed hits and COBALT refer to the same domain database.

Definition at line 664 of file rps.cpp.

References _ASSERT, CHitList::AddToHitList(), compare_seqids(), CSeq_id::CompareOrdered(), ConstBegin(), eDetectLoops, CSeqDB::eProtein, CSeq_align_set_Base::Get(), CBlast4_get_search_results_reply_Base::GetAlignments(), CBlast4_queries_Base::GetBioseq_set(), CBlast4_request_Base::GetBody(), CSeq_align_Base::C_Segs::GetDenseg(), CRange_Base::GetFrom(), CHitList::GetHit(), CScore_Base::GetId(), CDense_seg_Base::GetIds(), CScore_Base::C_Value::GetInt(), CBlast4_queue_search_request_Base::GetQueries(), CBlast4_request_body_Base::GetQueue_search(), CScore_Base::C_Value::GetReal(), CBlast4_archive_Base::GetRequest(), CBlast4_archive_Base::GetResults(), CMultiAlignerOptions::GetRpsDb(), CMultiAlignerOptions::GetRpsEvalue(), CSeq_align_Base::GetScore(), CSeq_align_Base::GetSegs(), CSeq_align::GetSeq_id(), CBlast4_queries_Base::GetSeq_loc_list(), CObject_id_Base::GetStr(), CRange_Base::GetTo(), CScore_Base::GetValue(), CMultiAlignerOptions::GetVerbose(), i, CBlast4_queries_Base::IsBioseq_set(), CBlast4_queries_Base::IsSeq_loc_list(), ITERATE, m_DomainHits, m_IsDomainSearched, m_Options, CHit::m_Score, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, m_tQueries, NCBI_THROW, NULL, CHitList::PurgeAllHits(), CSeqDB::SeqidToOid(), CHitList::Size(), and ct::sort().

Referenced by CMultiAlignerTest::SetDomainHits(), and x_FindDomainHits().

◆ x_SetScoreMatrix()

void CMultiAligner::x_SetScoreMatrix ( const char * matrix_name )

private

Set the score matrix the aligner will use.

NOTE that at present any hits between sequences will always be scored using BLOSUM62; the matrix chosen here is only used when forming the complete alignment

Parameters

matrix_name The score matrix to use; limited to the same list of matrices as blast [in]

Definition at line 381 of file cobalt.cpp.

References m_Aligner, NCBI_THROW, NCBISM_Blosum45, NCBISM_Blosum62, NCBISM_Blosum80, NCBISM_Pam250, NCBISM_Pam30, NCBISM_Pam70, CPSSMAligner::SetScoreMatrix(), and util::strcmp().

Referenced by x_InitAligner().

◆ x_ValidateInputMSAs()

bool CMultiAligner::x_ValidateInputMSAs ( void ) const

protected

Validate input alignments.

Throws if alignments do not pass validation.

Parameters

True	if validation passed

Definition at line 129 of file cobalt.cpp.

References NStr::IntToString(), ITERATE, m_InMSA1, m_InMSA2, m_Msa1Repr, m_Msa2Repr, and NCBI_THROW.

Referenced by SetInputMSAs().

◆ x_ValidateQueries()

bool CMultiAligner::x_ValidateQueries ( void ) const

protected

Validate query sequences.

Check for gaps in sequences. Throws if queries do not pass validation.

Returns: True if validation passed

Definition at line 114 of file cobalt.cpp.

References i, ITERATE, CSequence::kGapChar, m_QueryData, and NCBI_THROW.

Referenced by SetQueries().

◆ x_ValidateUserHits()

bool CMultiAligner::x_ValidateUserHits ( void )

protected

Validate user constraints with queries.

Throws if constraints do not pass validation.

Returns: True if validation passed

Definition at line 159 of file cobalt.cpp.

References CRange_Base::GetFrom(), CHitList::GetHit(), GetLength(), CRange_Base::GetTo(), i, m_QueryData, CHit::m_SeqIndex1, CHit::m_SeqIndex2, CHit::m_SeqRange1, CHit::m_SeqRange2, m_UserHits, NCBI_THROW, and CHitList::Size().

Referenced by SetInputMSAs(), and SetQueries().

Friends And Related Function Documentation

◆ ::CMultiAlignerTest

friend class ::CMultiAlignerTest

friend

Definition at line 749 of file cobalt.hpp.

◆ compare_sseg_db_idx

friend class compare_sseg_db_idx

friend

Definition at line 519 of file cobalt.hpp.

Member Data Documentation

◆ kClusterNodeId

const int CMultiAligner::kClusterNodeId = 16000

staticprotected

Definition at line 747 of file cobalt.hpp.

Referenced by x_AlignProgressive(), x_AttachClusterTrees(), x_BuildAlignment(), and x_Run().

◆ kDefaultGapExtend

const CNWAligner::TScore CMultiAligner::kDefaultGapExtend = -1

static

Default gap extension penalty.

Definition at line 300 of file cobalt.hpp.

◆ kDefaultGapOpen

const CNWAligner::TScore CMultiAligner::kDefaultGapOpen = -11

static

Default gap open penalty.

Definition at line 297 of file cobalt.hpp.

◆ kMajorVersion

const int CMultiAligner::kMajorVersion = 3

static

Version information.

Definition at line 72 of file cobalt.hpp.

◆ kMinorVersion

const int CMultiAligner::kMinorVersion = 0

static

Definition at line 73 of file cobalt.hpp.

◆ kPatchVersion

const int CMultiAligner::kPatchVersion = 0

static

Definition at line 74 of file cobalt.hpp.

◆ kRpsScaleFactor

const int CMultiAligner::kRpsScaleFactor = 100

staticprivate

Definition at line 500 of file cobalt.hpp.

Referenced by x_AlignProfileProfile(), x_ComputeProfileRangeAlignment(), x_FindDomainHits(), and x_RealignBlocks().

◆ m_Aligner

CPSSMAligner CMultiAligner::m_Aligner

protected

Definition at line 710 of file cobalt.hpp.

Referenced by x_AlignInClusters(), x_AlignProfileProfile(), x_ComputeProfileRangeAlignment(), x_ComputeTree(), x_InitAligner(), x_RealignBlocks(), and x_SetScoreMatrix().

◆ m_AllQueries

vector< CRef<objects::CSeq_loc> > CMultiAligner::m_AllQueries

protected

Definition at line 727 of file cobalt.hpp.

Referenced by x_FindQueryClusters(), x_MultiAlignClusters(), and x_Run().

◆ m_AllQueryData

vector<CSequence> CMultiAligner::m_AllQueryData

protected

Definition at line 728 of file cobalt.hpp.

Referenced by x_AlignInClusters(), x_AssignDefaultResFreqs(), x_FindQueryClusters(), x_MakeClusterResidueFrequencies(), and x_MultiAlignClusters().

◆ m_ClustAlnMethod

CMultiAlignerOptions::EInClustAlnMethod CMultiAligner::m_ClustAlnMethod

protected

Definition at line 744 of file cobalt.hpp.

Referenced by x_AlignMSAs(), x_AssignDefaultResFreqs(), x_BuildAlignmentIterative(), x_ComputeTree(), x_CreateBlastQueries(), x_CreatePatternQueries(), x_FindQueryClusters(), x_InitParams(), and x_Run().

◆ m_Clusterer

CClusterer CMultiAligner::m_Clusterer

protected

Definition at line 712 of file cobalt.hpp.

Referenced by x_AlignClusterQueries(), x_AlignInClusters(), x_AlignProgressive(), x_AttachClusterTrees(), x_BuildFullTree(), x_ComputeClusterTrees(), x_ComputeTree(), x_CreateBlastQueries(), x_CreatePatternQueries(), x_FindInClusterConstraints(), x_FindQueryClusters(), x_MakeClusterResidueFrequencies(), x_MultiAlignClusters(), and x_Run().

◆ m_ClusterGapPositions

vector< vector<Uint4> > CMultiAligner::m_ClusterGapPositions

protected

Definition at line 726 of file cobalt.hpp.

Referenced by x_AlignInClusters(), x_MakeClusterResidueFrequencies(), and x_MultiAlignClusters().

◆ m_CombinedHits

CHitList CMultiAligner::m_CombinedHits

protected

Definition at line 718 of file cobalt.hpp.

Referenced by Reset(), x_AlignMSAs(), x_BuildAlignmentIterative(), x_ComputeTree(), x_FindAlignmentSubsets(), x_FindConsistentHitSubset(), x_FindDomainHits(), x_FindLocalHits(), and x_MakeFillerBlocks().

◆ m_DomainHits

CHitList CMultiAligner::m_DomainHits

protected

Definition at line 716 of file cobalt.hpp.

Referenced by CMultiAlignerTest::CompareDomainHits(), CMultiAlignerTest::GetDomainHits(), Reset(), x_AlignProgressive(), x_FindDomainHits(), x_FindLocalHits(), and x_SetDomainHits().

◆ m_InMSA1

vector<CSequence> CMultiAligner::m_InMSA1

protected

Input alignment.

Definition at line 694 of file cobalt.hpp.

Referenced by SetInputMSAs(), x_AlignMSAs(), x_CreateBlastQueries(), x_Run(), and x_ValidateInputMSAs().

◆ m_InMSA2

vector<CSequence> CMultiAligner::m_InMSA2

protected

Input alignment.

Definition at line 696 of file cobalt.hpp.

Referenced by SetInputMSAs(), x_AlignMSAs(), x_CreateBlastQueries(), and x_ValidateInputMSAs().

◆ m_Interrupt

FInterruptFn CMultiAligner::m_Interrupt

protected

Definition at line 736 of file cobalt.hpp.

Referenced by SetInterruptCallback(), x_AlignFillerBlocks(), x_AlignInClusters(), x_AlignProgressive(), x_AssignDefaultResFreqs(), x_AssignRPSResFreqs(), x_ComputeTree(), x_FindConstraints(), x_FindDomainHits(), x_FindPatternHits(), x_FindRPSHits(), x_MakeClusterResidueFrequencies(), x_MultiAlignClusters(), and x_RealignBlocks().

◆ m_IsDomainSearched

vector<bool> CMultiAligner::m_IsDomainSearched

protected

Marks sequences with pre-computed domain hits.

Definition at line 724 of file cobalt.hpp.

Referenced by CMultiAlignerTest::GetIsDomainSearched(), x_FindDomainHits(), and x_SetDomainHits().

◆ m_LocalHits

CHitList CMultiAligner::m_LocalHits

protected

Definition at line 717 of file cobalt.hpp.

Referenced by Reset(), x_AlignFillerBlocks(), and x_FindLocalHits().

◆ m_LocalInClusterHits

CHitList CMultiAligner::m_LocalInClusterHits

protected

Definition at line 720 of file cobalt.hpp.

Referenced by x_AlignClusterQueries(), x_BuildAlignmentIterative(), and x_FindLocalInClusterHits().

◆ m_Messages

vector<string> CMultiAligner::m_Messages

protected

Definition at line 739 of file cobalt.hpp.

Referenced by Run(), and x_AlignProfileProfileUsingHit().

◆ m_Msa1Repr

vector<int> CMultiAligner::m_Msa1Repr

protected

Indices of sequence representatives in input alignment 1.

Definition at line 699 of file cobalt.hpp.

Referenced by SetInputMSAs(), x_CreateBlastQueries(), and x_ValidateInputMSAs().

◆ m_Msa2Repr

vector<int> CMultiAligner::m_Msa2Repr

protected

Indices of sequence representatives in input alignment 2.

Definition at line 701 of file cobalt.hpp.

Referenced by SetInputMSAs(), x_CreateBlastQueries(), and x_ValidateInputMSAs().

◆ m_Options

CConstRef<CMultiAlignerOptions> CMultiAligner::m_Options

protected

◆ m_PatternHits

CHitList CMultiAligner::m_PatternHits

protected

Definition at line 719 of file cobalt.hpp.

Referenced by Reset(), x_BuildAlignmentIterative(), and x_FindPatternHits().

◆ m_ProgressMonitor

SProgress CMultiAligner::m_ProgressMonitor

protected

◆ m_QueryData

vector<CSequence> CMultiAligner::m_QueryData

protected

◆ m_Results

vector<CSequence> CMultiAligner::m_Results

protected

Definition at line 703 of file cobalt.hpp.

Referenced by GetResults(), Reset(), x_AlignInClusters(), x_AlignMSAs(), x_BuildAlignmentIterative(), and x_MultiAlignClusters().

◆ m_RPSLocs

vector< vector<TRange> > CMultiAligner::m_RPSLocs

protected

Definition at line 731 of file cobalt.hpp.

Referenced by x_AddRpsFreqsToCluster(), x_AssignRPSResFreqs(), and x_FindDomainHits().

◆ m_Scope

CRef<objects::CScope> CMultiAligner::m_Scope

protected

Definition at line 689 of file cobalt.hpp.

Referenced by SetInputMSAs(), SetQueries(), x_AlignClusterQueries(), x_AlignFillerBlocks(), x_CreateBlastQueries(), x_FindQueryClusters(), and x_MakeFillerBlocks().

◆ m_Score

int CMultiAligner::m_Score

protected

Alignment score.

Definition at line 706 of file cobalt.hpp.

Referenced by x_BuildAlignmentIterative().

◆ m_tQueries

vector< CRef<objects::CSeq_loc> > CMultiAligner::m_tQueries

protected

Definition at line 688 of file cobalt.hpp.

Referenced by CMultiAlignerTest::SetDomainHits(), SetInputMSAs(), SetQueries(), CMultiAlignerTest::SetQuerySeqlocs(), x_AlignClusterQueries(), x_CreateBlastQueries(), x_FindDomainHits(), x_FindQueryClusters(), x_MakeFillerBlocks(), x_MultiAlignClusters(), x_Run(), and x_SetDomainHits().

◆ m_Tree

CTree CMultiAligner::m_Tree

protected

Definition at line 711 of file cobalt.hpp.

Referenced by GetTreeContainer(), x_BuildFullTree(), x_ComputeTree(), and x_Run().

◆ m_UserHits

CHitList CMultiAligner::m_UserHits

protected

Definition at line 721 of file cobalt.hpp.

Referenced by x_BuildAlignmentIterative(), x_FindQueryClusters(), x_InitParams(), and x_ValidateUserHits().

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

include/algo/cobalt/cobalt.hpp
src/algo/cobalt/blast.cpp
src/algo/cobalt/cobalt.cpp
src/algo/cobalt/phi.cpp
src/algo/cobalt/prog.cpp
src/algo/cobalt/rps.cpp
src/algo/cobalt/seg.cpp
src/algo/cobalt/seqalign.cpp

Classes

Public Types

Public Member Functions

Static Public Attributes

Protected Types

Protected Member Functions

Protected Attributes

Static Protected Attributes

Private Types

Private Member Functions

Static Private Member Functions

Static Private Attributes

Friends

Additional Inherited Members

Detailed Description

Member Typedef Documentation

◆ FInterruptFn

◆ SColumn

◆ SGraphNode

◆ SSegmentLoc

◆ TKmerCounts

◆ TKMethods

◆ TRangePair

◆ TStatus

Member Enumeration Documentation

◆ EAlignmentStage

◆ EEndGapCostStrategy

◆ EStatus

Constructor & Destructor Documentation

◆ CMultiAligner() [1/3]

◆ CMultiAligner() [2/3]

◆ CMultiAligner() [3/3]

Member Function Documentation

◆ GetMessages()

◆ GetOptions()

◆ GetQueries()

◆ GetQueryClusters()

◆ GetResults() [1/2]

◆ GetResults() [2/2]

◆ GetScope()

◆ GetScore()

◆ GetSeqResults()

◆ GetTree()

◆ GetTreeContainer()

◆ IsMessage()

◆ Reset()

◆ Run()

◆ SetInputMSAs()

◆ SetInterruptCallback()

◆ SetQueries() [1/3]

◆ SetQueries() [2/3]

◆ SetQueries() [3/3]

◆ x_AddNewSegment()

◆ x_AddRpsFreqsToCluster()

◆ x_AlignClusterQueries()

◆ x_AlignFillerBlocks()

◆ x_AlignInClusters()

◆ x_AlignMSAs()

◆ x_AlignProfileProfile()

◆ x_AlignProfileProfileUsingHit()

◆ x_AlignProgressive()

◆ x_AssignDefaultResFreqs()

◆ x_AssignRPSResFreqs()

◆ x_AttachClusterTrees()

◆ x_BuildAlignment()

◆ x_BuildAlignmentIterative()

◆ x_BuildFullTree()

◆ x_ComputeClusterTrees()

◆ x_ComputeProfileRangeAlignment()

◆ x_ComputeTree()

◆ x_CreateBlastQueries()

◆ x_CreatePatternQueries()

◆ x_FindAlignmentSubsets()

◆ x_FindBestPath()

◆ x_FindConservedColumns()

◆ x_FindConsistentHitSubset()

◆ x_FindConstraints()

◆ x_FindDomainHits()

◆ x_FindInClusterConstraints()

◆ x_FindLocalHits()