BLAST filtering functions. More...

#include <algo/blast/core/ncbi_std.h>
#include <algo/blast/core/blast_def.h>
#include <algo/blast/core/blast_program.h>
#include <algo/blast/core/blast_query_info.h>
#include <algo/blast/core/blast_message.h>
#include <algo/blast/core/blast_options.h>

Include dependency graph for blast_filter.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Go to the SVN repository for this file.

Macros
#define	REPEATS_SEARCH_EVALUE 0.1
	Repeats filtering default options. More...

#define	REPEATS_SEARCH_MINSCORE 26
	Default score cutoff. More...

#define	REPEATS_SEARCH_PENALTY -1
	Default mismatch penalty. More...

#define	REPEATS_SEARCH_REWARD 1
	Default match reward. More...

#define	REPEATS_SEARCH_GAP_OPEN 2
	Default gap opening cost. More...

#define	REPEATS_SEARCH_GAP_EXTEND 1
	Default gap extension cost. More...

#define	REPEATS_SEARCH_WORD_SIZE 11
	Default word size. More...

#define	REPEATS_SEARCH_XDROP_UNGAPPED 40
	Default X-dropoff for ungapped extension. More...

#define	REPEATS_SEARCH_XDROP_FINAL 90
	Default X-dropoff for gapped extension with traceback. More...

#define	REPEATS_SEARCH_FILTER_STRING "F"
	Default filter string - no filtering. More...

#define	REPEAT_MASK_LINK_VALUE 5
	Largest gap allowed to be filled between repeat mask intervals. More...

Functions
BlastSeqLoc *	BlastSeqLocNew (BlastSeqLoc **head, Int4 from, Int4 to)
	Create and initialize a new sequence interval. More...

BlastSeqLoc *	BlastSeqLocAppend (BlastSeqLoc *head, BlastSeqLoc node)
	Appends the BlastSeqLoc to the list of BlastSeqLoc-s pointed to by head. More...

BlastSeqLoc *	BlastSeqLocNodeFree (BlastSeqLoc *node)
	Deallocate a single BlastSeqLoc structure and its contents, without following its next pointer. More...

BlastSeqLoc *	BlastSeqLocFree (BlastSeqLoc *loc)
	Deallocate all BlastSeqLoc objects in a chain. More...

BlastSeqLoc *	BlastSeqLocListDup (BlastSeqLoc *head)
	Make a deep copy of the linked list of BlastSeqLoc-s pointed to by its argument. More...

void	BlastSeqLocReverse (BlastSeqLoc *masks, Int4 query_length)
	Converts reverse strand coordinates to forward strand in place. More...

void	BlastSeqLocCombine (BlastSeqLoc **mask_loc, Int4 link_value)
	Go through all mask locations in one sequence and combine any that overlap, deallocating the unneeded locations. More...

BlastMaskLoc *	BlastMaskLocNew (Int4 total)
	Allocate memory for a BlastMaskLoc. More...

BlastMaskLoc *	BlastMaskLocDup (const BlastMaskLoc *mask_loc)
	Perform a deep copy of the BlastMaskLoc structure passed to this function. More...

BlastMaskLoc *	BlastMaskLocFree (BlastMaskLoc *mask_loc)
	Deallocate memory for a BlastMaskLoc structure as well as the BlastSeqLoc's pointed to. More...

Int2	BlastMaskLocDNAToProtein (BlastMaskLoc mask_loc, const BlastQueryInfo query_info)
	Given a BlastMaskLoc with an array of lists of DNA mask locations, substitutes that array by a new array of per-protein-frame mask location lists. More...

Int2	BlastMaskLocProteinToDNA (BlastMaskLoc mask_loc, const BlastQueryInfo query_info)
	Given a BlastMaskLoc with an array of lists of mask locations per protein frame, recalculates all mask offsets in terms of the DNA sequence. More...

Int2	BLAST_ComplementMaskLocations (EBlastProgramType program_number, const BlastQueryInfo query_info, const BlastMaskLoc mask_loc, BlastSeqLoc **complement_mask)
	This function takes the list of mask locations (i.e., regions that should not be searched or not added to lookup table) and makes up a set of SSeqRange*'s in the concatenated sequence built from a set of queries, that should be searched (that is, takes the complement). More...

Int2	BlastSetUp_Filter (EBlastProgramType program_number, Uint1 sequence, Int4 length, Int4 offset, const SBlastFilterOptions filter_options, BlastSeqLoc seqloc_retval, Blast_Message blast_message)
	Runs seg filtering functions, according to the filtering options, returns BlastSeqLoc*. More...

Int2	BlastSetUp_GetFilteringLocations (BLAST_SequenceBlk query_blk, const BlastQueryInfo query_info, EBlastProgramType program_number, const SBlastFilterOptions filter_options, BlastMaskLoc filter_out, Blast_Message *blast_message)
	Does preparation for filtering and then calls BlastSetUp_Filter. More...

void	Blast_MaskTheResidues (Uint1 buffer, Int4 length, Boolean is_na, const BlastSeqLoc mask_loc, Boolean reverse, Int4 offset)
	Masks the letters in buffer. More...

void	Blast_MaskUnsupportedAA (BLAST_SequenceBlk *seq, Uint1 min_invalid)
	Mask protein letters that are currently unsupported. More...

void	BlastSetUp_MaskQuery (BLAST_SequenceBlk query_blk, const BlastQueryInfo query_info, const BlastMaskLoc *filter_maskloc, EBlastProgramType program_number)
	Masks the sequence given a BlastMaskLoc. More...

Int2	BlastFilteringOptionsFromString (EBlastProgramType program_number, const char instructions, SBlastFilterOptions filtering_options, Blast_Message *blast_message)
	Produces SBlastFilterOptions from a string that has been traditionally supported in blast. More...

char *	BlastFilteringOptionsToString (const SBlastFilterOptions *filtering_options)
	Convert the filtering options structure to a string. More...

static NCBI_INLINE Boolean	BlastIsReverseStrand (Boolean is_na, Int4 context)
	Determines whether this is a nucleotide query and whether this a minus strand or not. More...

Variables
const Uint1	kNuclMask
	BLASTNA element used to mask bases in BLAST. More...

const Uint1	kProtMask
	NCBISTDAA element used to mask residues in BLAST. More...

Detailed Description

BLAST filtering functions.

Definition in file blast_filter.h.

Macro Definition Documentation

◆ REPEAT_MASK_LINK_VALUE

#define REPEAT_MASK_LINK_VALUE 5

Largest gap allowed to be filled between repeat mask intervals.

Definition at line 72 of file blast_filter.h.

◆ REPEATS_SEARCH_EVALUE

#define REPEATS_SEARCH_EVALUE 0.1

Repeats filtering default options.

Default e-value threshold, keep for C toolkit

Definition at line 57 of file blast_filter.h.

◆ REPEATS_SEARCH_FILTER_STRING

#define REPEATS_SEARCH_FILTER_STRING "F"

Default filter string - no filtering.

Definition at line 69 of file blast_filter.h.

◆ REPEATS_SEARCH_GAP_EXTEND

#define REPEATS_SEARCH_GAP_EXTEND 1

Default gap extension cost.

Definition at line 62 of file blast_filter.h.

◆ REPEATS_SEARCH_GAP_OPEN

#define REPEATS_SEARCH_GAP_OPEN 2

Default gap opening cost.

Definition at line 61 of file blast_filter.h.

◆ REPEATS_SEARCH_MINSCORE

#define REPEATS_SEARCH_MINSCORE 26

Default score cutoff.

Definition at line 58 of file blast_filter.h.

◆ REPEATS_SEARCH_PENALTY

#define REPEATS_SEARCH_PENALTY -1

Default mismatch penalty.

Definition at line 59 of file blast_filter.h.

◆ REPEATS_SEARCH_REWARD

#define REPEATS_SEARCH_REWARD 1

Default match reward.

Definition at line 60 of file blast_filter.h.

◆ REPEATS_SEARCH_WORD_SIZE

#define REPEATS_SEARCH_WORD_SIZE 11

Default word size.

Definition at line 63 of file blast_filter.h.

◆ REPEATS_SEARCH_XDROP_FINAL

#define REPEATS_SEARCH_XDROP_FINAL 90

Default X-dropoff for gapped extension with traceback.

Definition at line 67 of file blast_filter.h.

◆ REPEATS_SEARCH_XDROP_UNGAPPED

#define REPEATS_SEARCH_XDROP_UNGAPPED 40

Default X-dropoff for ungapped extension.

Definition at line 65 of file blast_filter.h.

Function Documentation

◆ BLAST_ComplementMaskLocations()

Int2 BLAST_ComplementMaskLocations	(	EBlastProgramType	program_number,
		const BlastQueryInfo *	query_info,
		const BlastMaskLoc *	mask_loc,
		BlastSeqLoc **	complement_mask
	)

This function takes the list of mask locations (i.e., regions that should not be searched or not added to lookup table) and makes up a set of SSeqRange*'s in the concatenated sequence built from a set of queries, that should be searched (that is, takes the complement).

If all sequences in the query set are completely filtered, then an SSeqRange is created and both of its elements (left and right) are set to -1 to indicate this. If any of the mask_loc's is NULL, an SSeqRange for the full span of the respective query sequence is created.

Parameters

program_number	Type of BLAST program [in]
query_info	The query information structure [in]
mask_loc	All mask locations [in]
complement_mask	Linked list of SSeqRange*s in the concatenated sequence to be indexed in the lookup table . [out]

Definition at line 1017 of file blast_filter.c.

References ASSERT, BlastIsReverseStrand(), BlastSeqLocListReverse(), BlastSeqLocNew(), context, BlastQueryInfo::contexts, eBlastTypeBlastn, eBlastTypeMapping, FALSE, first(), BlastQueryInfo::first_context, BlastContextInfo::is_valid, SSeqRange::left, BlastSeqLoc::next, NULL, BlastContextInfo::query_length, BlastContextInfo::query_offset, SSeqRange::right, BlastMaskLoc::seqloc_array, BlastSeqLoc::ssr, and TRUE.

Referenced by BLAST_MainSetUp().

◆ Blast_MaskTheResidues()

void Blast_MaskTheResidues	(	Uint1 *	buffer,
		Int4	length,
		Boolean	is_na,
		const BlastSeqLoc *	mask_loc,
		Boolean	reverse,
		Int4	offset
	)

Masks the letters in buffer.

This is a low-level routine and takes a raw buffer which it assumes to be in ncbistdaa (protein) or blastna (nucleotide).

Parameters

buffer	the sequence to be masked (will be modified, cannot be NULL or undefined behavior will result).[in\|out]
length	length of the sequence to be masked . [in]
is_na	nucleotide if TRUE [in]
mask_loc	the BlastSeqLoc to use for masking [in]
reverse	minus strand if TRUE [in]
offset	how far along sequence is 1st residuse in buffer [in]

Definition at line 1307 of file blast_filter.c.

References ASSERT, buffer, kNuclMask, kProtMask, SSeqRange::left, BlastSeqLoc::next, offset, SSeqRange::right, and BlastSeqLoc::ssr.

Referenced by BlastSetUp_MaskQuery(), and s_DoSegSequenceData().

◆ Blast_MaskUnsupportedAA()

void Blast_MaskUnsupportedAA	(	BLAST_SequenceBlk *	seq,
		Uint1	min_invalid
	)

Mask protein letters that are currently unsupported.

This routine is used to make the core ignore letters within protein sequences that cannot (yet) be correctly handled

Parameters

seq	Protein sequence to be masked (ncbistdaa format required). Letters whose numerical value exceeds a cutoff are converted into kProtMask values [in\|out]
min_invalid	The first ncbistdaa value that is considered invalid. All sequence letters with numerical value >= this number are masked [in]

Definition at line 1336 of file blast_filter.c.

References i, kProtMask, BLAST_SequenceBlk::length, and BLAST_SequenceBlk::sequence.

Referenced by LookupTableWrapInit_MT().

◆ BlastFilteringOptionsFromString()

Int2 BlastFilteringOptionsFromString	(	EBlastProgramType	program_number,
		const char *	instructions,
		SBlastFilterOptions **	filtering_options,
		Blast_Message **	blast_message
	)

Produces SBlastFilterOptions from a string that has been traditionally supported in blast.

Parameters

program_number	Type of BLAST program [in]
instructions	the string describing the filtering to be done [in]
filtering_options	the structure to be filled in [out]
blast_message	optional field for error messages [out]

Returns: zero on success

Definition at line 436 of file blast_filter.c.

References Blast_MessageWrite(), buffer, calloc(), SRepeatFilterOptions::database, SWindowMaskerOptions::database, dbname(), eBlastSevError, eBlastTypeBlastn, eBlastTypeMapping, eEmpty, FALSE, SSegOptions::hicut, kBlastMessageNoContext, SDustOptions::level, SDustOptions::linker, SSegOptions::locut, NULL, NULLB, s_LoadOptionsToBuffer(), s_ParseDustOptions(), s_ParseRepeatOptions(), s_ParseSegOptions(), s_ParseWindowMaskerOptions(), SBlastFilterOptionsNew(), SDustOptionsFree(), SDustOptionsNew(), sfree, SRepeatFilterOptionsFree(), SRepeatFilterOptionsNew(), SSegOptionsFree(), SSegOptionsNew(), strcasecmp, SWindowMaskerOptionsFree(), SWindowMaskerOptionsNew(), SWindowMaskerOptions::taxid, TRUE, SDustOptions::window, and SSegOptions::window.

Referenced by BLAST_FillQuerySetUpOptions(), BLAST_MainSetUp(), BOOST_AUTO_TEST_CASE(), s_DoSegSequenceData(), and CBlastOptionsLocal::SetFilterString().

◆ BlastFilteringOptionsToString()

char* BlastFilteringOptionsToString ( const SBlastFilterOptions * filtering_options )

Convert the filtering options structure to a string.

Parameters

filtering_options filtering options structure, assumed to be correctly filled in [in]

Returns: malloc'd string containing filtering options or NULL if there is not enough memory to create the return value. Caller must free() return value

See also: TAutoCharPtr it is recommended that the caller store the return value in a TAutoCharPtr if working in C++ (defined in blast_aux.hpp)

Definition at line 321 of file blast_filter.c.

References buffer, calloc(), SRepeatFilterOptions::database, SWindowMaskerOptions::database, SBlastFilterOptions::dustOptions, SSegOptions::hicut, kDustLevel, kDustLinker, kDustWindow, kSegHicut, kSegLocut, kSegWindow, SDustOptions::level, SDustOptions::linker, SSegOptions::locut, NULL, SBlastFilterOptions::repeatFilterOptions, s_SafeStrCat(), SBlastFilterOptionsMaskAtHash(), SBlastFilterOptions::segOptions, strdup, SWindowMaskerOptions::taxid, SDustOptions::window, SSegOptions::window, and SBlastFilterOptions::windowMaskerOptions.

Referenced by BOOST_AUTO_TEST_CASE(), and CBlastOptionsLocal::GetFilterString().

◆ BlastIsReverseStrand()

static NCBI_INLINE Boolean BlastIsReverseStrand	(	Boolean	is_na,
		Int4	context
	)

static

Determines whether this is a nucleotide query and whether this a minus strand or not.

Parameters

is_na	the query is nucleotide
context	offset in the QueryInfo array

Returns: TRUE if this is minus strand

Definition at line 325 of file blast_filter.h.

References context.

Referenced by BLAST_ComplementMaskLocations(), BlastSetUp_MaskQuery(), and s_GetFilteringLocationsForOneContext().

◆ BlastMaskLocDNAToProtein()

Int2 BlastMaskLocDNAToProtein	(	BlastMaskLoc *	mask_loc,
		const BlastQueryInfo *	query_info
	)

Given a BlastMaskLoc with an array of lists of DNA mask locations, substitutes that array by a new array of per-protein-frame mask location lists.

Parameters

mask_loc	Mask locations structure. This structure can have either masks for all frames in nucleotide coordinates (e.g.: the results of translating protein masks to nucleotide) or a single mask per query (i.e.:location NUM_FRAMES*query_index). In the latter case, this mask will be used for all frames. [in\|out]
query_info	Query information structure, containing contexts data [in]

Note: This function does NOT take into consideration the strands requested to be searched, which is INCONSISTENT with what the C++ API does.

Definition at line 806 of file blast_filter.c.

References ASSERT, BLAST_ContextToFrame(), BlastQueryInfoGetQueryLength(), BlastSeqLocFree(), BlastSeqLocNew(), CODON_LENGTH, context, BlastQueryInfo::contexts, eBlastTypeBlastx, BlastQueryInfo::last_context, SSeqRange::left, BlastSeqLoc::next, NULL, NUM_FRAMES, BlastQueryInfo::num_queries, BlastContextInfo::query_length, SSeqRange::right, BlastMaskLoc::seqloc_array, BlastSeqLoc::ssr, and BlastMaskLoc::total_size.

Referenced by BOOST_AUTO_TEST_CASE().

◆ BlastMaskLocDup()

BlastMaskLoc* BlastMaskLocDup ( const BlastMaskLoc * mask_loc )

Perform a deep copy of the BlastMaskLoc structure passed to this function.

Parameters

mask_loc Source masking location structure [in]

Returns: Deep copy of its argument, or NULL if the argument was NULL or if not enough memory was available

Definition at line 770 of file blast_filter.c.

References BlastMaskLocNew(), BlastSeqLocListDup(), NULL, BlastMaskLoc::seqloc_array, and BlastMaskLoc::total_size.

◆ BlastMaskLocFree()

BlastMaskLoc* BlastMaskLocFree ( BlastMaskLoc * mask_loc )

Deallocate memory for a BlastMaskLoc structure as well as the BlastSeqLoc's pointed to.

Parameters

mask_loc the object to be deleted [in]

Returns: NULL pointer

Definition at line 789 of file blast_filter.c.

References BlastSeqLocFree(), NULL, BlastMaskLoc::seqloc_array, sfree, and BlastMaskLoc::total_size.

Referenced by BLAST_MainSetUp(), BlastSequenceBlkFree(), and BOOST_AUTO_TEST_CASE().

◆ BlastMaskLocNew()

BlastMaskLoc* BlastMaskLocNew ( Int4 total )

Allocate memory for a BlastMaskLoc.

Parameters

total number of contexts for which SSeqLocs should be allocated (result of number of queries * number of contexts for given program) [in]

Returns: Pointer to the allocated BlastMaskLoc structure.

Definition at line 760 of file blast_filter.c.

References calloc(), BlastMaskLoc::seqloc_array, and BlastMaskLoc::total_size.

Referenced by BlastMaskLocDup(), BlastSetUp_GetFilteringLocations(), BOOST_AUTO_TEST_CASE(), SetupQueries_OMF(), and x_TestGetSeqLocInfoVector().

◆ BlastMaskLocProteinToDNA()

Int2 BlastMaskLocProteinToDNA	(	BlastMaskLoc *	mask_loc,
		const BlastQueryInfo *	query_info
	)

Given a BlastMaskLoc with an array of lists of mask locations per protein frame, recalculates all mask offsets in terms of the DNA sequence.

Parameters

mask_loc	Mask locations structure [in\|out]
query_info	Query information structure, containing contexts data [in]

Definition at line 892 of file blast_filter.c.

References ASSERT, BLAST_ContextToFrame(), BlastQueryInfoGetQueryLength(), CODON_LENGTH, eBlastTypeBlastx, BlastQueryInfo::last_context, SSeqRange::left, BlastSeqLoc::next, NUM_FRAMES, BlastQueryInfo::num_queries, SSeqRange::right, BlastMaskLoc::seqloc_array, BlastSeqLoc::ssr, and BlastMaskLoc::total_size.

Referenced by BLAST_MainSetUp(), and BOOST_AUTO_TEST_CASE().

◆ BlastSeqLocAppend()

BlastSeqLoc* BlastSeqLocAppend	(	BlastSeqLoc **	head,
		BlastSeqLoc *	node
	)

Appends the BlastSeqLoc to the list of BlastSeqLoc-s pointed to by head.

Parameters

head	Pointer to the head of the linked list of BlastSeqLoc-s [in]
node	Pointer to the node to be added to the list. If this is NULL, this function does nothing. [in]

Returns: pointer to the second argument to this function (i.e.: tail of the list)

Definition at line 621 of file blast_filter.c.

References head, NULL, and tmp.

Referenced by BlastSeqLocListDup(), BlastSeqLocNew(), and s_GetFilteringLocationsForOneContext().

◆ BlastSeqLocCombine()

void BlastSeqLocCombine	(	BlastSeqLoc **	mask_loc,
		Int4	link_value
	)

Go through all mask locations in one sequence and combine any that overlap, deallocating the unneeded locations.

Parameters

mask_loc	The list of masks to be merged (in place) [in\|out]
link_value	Largest gap size between locations for which they should be linked together [in]

Definition at line 972 of file blast_filter.c.

References ASSERT, BlastSeqLocNodeFree(), i, SSeqRange::left, MAX, BlastSeqLoc::next, NULL, SSeqRange::right, s_BlastSeqLocListToArrayOfPointers(), s_SeqRangeSortByStartPosition(), sfree, and BlastSeqLoc::ssr.

Referenced by BOOST_AUTO_TEST_CASE(), s_FillMaskLocFromBlastResults(), and s_GetFilteringLocationsForOneContext().

◆ BlastSeqLocFree()

BlastSeqLoc* BlastSeqLocFree ( BlastSeqLoc * loc )

Deallocate all BlastSeqLoc objects in a chain.

Parameters

loc	object to be freed [in]

Returns: NULL pointer returned.

Definition at line 737 of file blast_filter.c.

References BlastSeqLocNodeFree(), BlastSeqLoc::next, and NULL.

Referenced by AascanTestFixture::AascanTestFixture(), BlastKmerGetKmerSet(), BlastMaskLocDNAToProtein(), BlastMaskLocFree(), BlastMBLookupTableDestruct(), BlastNaHashLookupTableDestruct(), BlastNaLookupTableDestruct(), BlastSmallNaLookupTableDestruct(), BOOST_AUTO_TEST_CASE(), CompressedAascanTestFixture::CompressedAascanTestFixture(), CSetupFactory::CreateScoreBlock(), CScore_SegPct::Get(), CSegMasker::operator()(), s_DoSegSequenceData(), s_FillMaskLocFromBlastResults(), s_GetInitialWordParameters(), s_SegsToBlastSeqLoc(), LinkHspTestFixture::setupCutoffScores(), TestFixture::TearDownQuery(), CPhiblastTestFixture::x_FindQueryOccurrences(), CMakeProfileDBApp::x_RPSUpdateLookup(), AalookupTestFixture::~AalookupTestFixture(), CBlastQueryFilteredFrames::~CBlastQueryFilteredFrames(), CBlastSeqLocWrap::~CBlastSeqLocWrap(), CompressedAalookupTestFixture::~CompressedAalookupTestFixture(), CTracebackTestFixture::~CTracebackTestFixture(), and NtlookupTestFixture::~NtlookupTestFixture().

◆ BlastSeqLocListDup()

BlastSeqLoc* BlastSeqLocListDup ( BlastSeqLoc * head )

Make a deep copy of the linked list of BlastSeqLoc-s pointed to by its argument.

Parameters

head	head of the linked list [in]

Returns: NULL on NULL input or memory allocation failure, else a copy of the list and its contents

Definition at line 747 of file blast_filter.c.

References BlastSeqLocAppend(), head, NULL, and s_BlastSeqLocNodeDup().

Referenced by BlastMaskLocDup().

◆ BlastSeqLocNew()

BlastSeqLoc* BlastSeqLocNew	(	BlastSeqLoc **	head,
		Int4	from,
		Int4	to
	)

Create and initialize a new sequence interval.

Parameters

head	existing BlastSeqLoc to append onto, if *head is NULL then it will be set to new BlastSeqLoc, may be NULL [in\|out]
from	Start of the interval [in]
to	End of the interval [in]

Returns: Pointer to the allocated BlastSeqLoc structure (i.e.: tail of the list).

Definition at line 608 of file blast_filter.c.

References BlastSeqLocAppend(), calloc(), head, SSeqRange::left, NULL, SSeqRange::right, and BlastSeqLoc::ssr.

Referenced by AascanTestFixture::AascanTestFixture(), CBlastQueryFilteredFrames::AddSeqLoc(), BLAST_ComplementMaskLocations(), BlastMaskLocDNAToProtein(), BOOST_AUTO_TEST_CASE(), CompressedAascanTestFixture::CompressedAascanTestFixture(), CSeqLoc2BlastSeqLoc(), NtlookupTestFixture::debruijnInit(), CompressedAalookupTestFixture::GetSeqBlk(), AalookupTestFixture::GetSeqBlk(), s_BlastSeqLocNodeDup(), s_GetInitialWordParameters(), s_RangeVector2BlastSeqLoc(), s_SeqAlignToBlastSeqLoc(), s_SeqLocListInvert(), LinkHspTestFixture::setupCutoffScores(), NtlookupTestFixture::SetUpQuery(), TestFixture::SetUpQuery(), CPhiblastTestFixture::x_FindQueryOccurrences(), CMakeProfileDBApp::x_RPSUpdateLookup(), and x_TestGetSeqLocInfoVector().

◆ BlastSeqLocNodeFree()

BlastSeqLoc* BlastSeqLocNodeFree ( BlastSeqLoc * node )

Deallocate a single BlastSeqLoc structure and its contents, without following its next pointer.

Parameters

node	structure to deallocate [in]

Returns: NULL

Definition at line 727 of file blast_filter.c.

References NULL, sfree, and BlastSeqLoc::ssr.

Referenced by BlastSeqLoc_RestrictToInterval(), BlastSeqLocCombine(), and BlastSeqLocFree().

◆ BlastSeqLocReverse()

void BlastSeqLocReverse	(	BlastSeqLoc *	masks,
		Int4	query_length
	)

Converts reverse strand coordinates to forward strand in place.

Parameters

masks	BlastSeqLoc to be reversed [in\|out]
query_length	length of query [in]

Definition at line 1173 of file blast_filter.c.

References SSeqRange::left, BlastSeqLoc::next, SSeqRange::right, and BlastSeqLoc::ssr.

Referenced by s_GetFilteringLocationsForOneContext().

◆ BlastSetUp_Filter()

Int2 BlastSetUp_Filter	(	EBlastProgramType	program_number,
		Uint1 *	sequence,
		Int4	length,
		Int4	offset,
		const SBlastFilterOptions *	filter_options,
		BlastSeqLoc **	seqloc_retval,
		Blast_Message **	blast_message
	)

Runs seg filtering functions, according to the filtering options, returns BlastSeqLoc*.

Should combine all SeqLocs so they are non-redundant.

Parameters

program_number	Type of BLAST program [in]
sequence	The sequence or part of the sequence to be filtered [in]
length	Length of the (sub)sequence [in]
offset	Offset into the full sequence [in]
filter_options	specifies how filtering is to be done [in]
seqloc_retval	Resulting locations for filtered region. [out]
blast_message	error messages on error [out]

Returns: zero on success

Definition at line 1122 of file blast_filter.c.

References ASSERT, FilterQueriesForMapping(), SSegOptions::hicut, SegParameters::hicut, SSegOptions::locut, SegParameters::locut, NULL, offset, SegParameters::overlaps, SBlastFilterOptions::readQualityOptions, SBlastFilterOptionsValidate(), SBlastFilterOptions::segOptions, SegParametersFree(), SegParametersNewAa(), SeqBufferSeg(), TRUE, SSegOptions::window, and SegParameters::window.

Referenced by BOOST_AUTO_TEST_CASE(), s_DoSegSequenceData(), and s_GetFilteringLocationsForOneContext().

◆ BlastSetUp_GetFilteringLocations()

Int2 BlastSetUp_GetFilteringLocations	(	BLAST_SequenceBlk *	query_blk,
		const BlastQueryInfo *	query_info,
		EBlastProgramType	program_number,
		const SBlastFilterOptions *	filter_options,
		BlastMaskLoc **	filter_out,
		Blast_Message **	blast_message
	)

Does preparation for filtering and then calls BlastSetUp_Filter.

Parameters

query_blk	sequence to be filtered [in]
query_info	info on sequence to be filtered [in]
program_number	one of blastn,blastp,blastx,etc. [in]
filter_options	specifies how filtering is to be done [in]
filter_out	resulting locations for filtered region. [out]
blast_message	message that needs to be sent back to user.

Definition at line 1262 of file blast_filter.c.

References ASSERT, BLAST_GetNumberOfContexts(), Blast_MessageWrite(), BlastMaskLocNew(), context, eBlastSevError, BlastQueryInfo::first_context, BlastQueryInfo::last_context, NULL, BlastQueryInfo::num_queries, and s_GetFilteringLocationsForOneContext().

Referenced by BLAST_MainSetUp(), and BOOST_AUTO_TEST_CASE().

◆ BlastSetUp_MaskQuery()

void BlastSetUp_MaskQuery	(	BLAST_SequenceBlk *	query_blk,
		const BlastQueryInfo *	query_info,
		const BlastMaskLoc *	filter_maskloc,
		EBlastProgramType	program_number
	)

Masks the sequence given a BlastMaskLoc.

Parameters

query_blk	sequence to be filtered [in]
query_info	info on sequence to be filtered [in]
filter_maskloc	Locations to filter [in]
program_number	one of blastn,blastp,blastx,etc. [in]

Definition at line 1350 of file blast_filter.c.

References ASSERT, Blast_MaskTheResidues(), BlastIsReverseStrand(), BlastMemDup(), buffer, context, BlastQueryInfo::contexts, eBlastTypeBlastn, eBlastTypeMapping, FALSE, BlastQueryInfo::first_context, BlastContextInfo::is_valid, BlastQueryInfo::last_context, BLAST_SequenceBlk::nomask_allocated, NULL, BlastContextInfo::query_length, BlastContextInfo::query_offset, BlastMaskLoc::seqloc_array, BLAST_SequenceBlk::sequence, BLAST_SequenceBlk::sequence_nomask, BLAST_SequenceBlk::sequence_start, BLAST_SequenceBlk::sequence_start_nomask, BlastMaskLoc::total_size, and TRUE.

Referenced by BLAST_MainSetUp(), and BOOST_AUTO_TEST_CASE().

Variable Documentation

◆ kNuclMask

const Uint1 kNuclMask

extern

BLASTNA element used to mask bases in BLAST.

Definition at line 38 of file blast_filter.c.

Referenced by Blast_MaskTheResidues(), and x_AreAllBasesMasked().

◆ kProtMask

const Uint1 kProtMask

extern

NCBISTDAA element used to mask residues in BLAST.

Definition at line 39 of file blast_filter.c.

Referenced by Blast_MaskTheResidues(), Blast_MaskUnsupportedAA(), and CPsiBlastInputClustalW::x_ValidateQueryInMsa().

Macros

Functions

Variables

Detailed Description

Macro Definition Documentation

◆ REPEAT_MASK_LINK_VALUE

◆ REPEATS_SEARCH_EVALUE

◆ REPEATS_SEARCH_FILTER_STRING

◆ REPEATS_SEARCH_GAP_EXTEND

◆ REPEATS_SEARCH_GAP_OPEN

◆ REPEATS_SEARCH_MINSCORE

◆ REPEATS_SEARCH_PENALTY

◆ REPEATS_SEARCH_REWARD

◆ REPEATS_SEARCH_WORD_SIZE

◆ REPEATS_SEARCH_XDROP_FINAL

◆ REPEATS_SEARCH_XDROP_UNGAPPED

Function Documentation

◆ BLAST_ComplementMaskLocations()

◆ Blast_MaskTheResidues()

◆ Blast_MaskUnsupportedAA()

◆ BlastFilteringOptionsFromString()

◆ BlastFilteringOptionsToString()

◆ BlastIsReverseStrand()

◆ BlastMaskLocDNAToProtein()

◆ BlastMaskLocDup()

◆ BlastMaskLocFree()

◆ BlastMaskLocNew()

◆ BlastMaskLocProteinToDNA()

◆ BlastSeqLocAppend()

◆ BlastSeqLocCombine()

◆ BlastSeqLocFree()

◆ BlastSeqLocListDup()

◆ BlastSeqLocNew()

◆ BlastSeqLocNodeFree()

◆ BlastSeqLocReverse()

◆ BlastSetUp_Filter()

◆ BlastSetUp_GetFilteringLocations()

◆ BlastSetUp_MaskQuery()

Variable Documentation

◆ kNuclMask

◆ kProtMask