#include <counting_bloom_filter.hpp>

Public Member Functions
	KmerCountingBloomFilter ()

	KmerCountingBloomFilter (size_t bytes, unsigned hash_num, unsigned k)

	KmerCountingBloomFilter (const std::string &path)

	KmerCountingBloomFilter (const KmerCountingBloomFilter &)=delete

	KmerCountingBloomFilter (KmerCountingBloomFilter &&)=delete

KmerCountingBloomFilter &	operator= (const KmerCountingBloomFilter &)=delete

KmerCountingBloomFilter &	operator= (KmerCountingBloomFilter &&)=delete

void	insert (const char *seq, size_t seq_len)

void	insert (const std::string &seq)

void	insert (const uint64_t *hashes, T n=1)

void	insert (const std::vector< uint64_t > &hashes, T n=1)

void	remove (const char *seq, size_t seq_len)

void	remove (const std::string &seq)

void	remove (const uint64_t *hashes)

void	remove (const std::vector< uint64_t > &hashes)

void	clear (const char *seq, size_t seq_len)

void	clear (const std::string &seq)

void	clear (const uint64_t *hashes)

void	clear (const std::vector< uint64_t > &hashes)

uint64_t	contains (const char *seq, size_t seq_len) const

uint64_t	contains (const std::string &seq) const

T	contains (const uint64_t *hashes) const

T	contains (const std::vector< uint64_t > &hashes) const

T	contains_insert (const char *seq, size_t seq_len)

T	contains_insert (const std::string &seq)

T	contains_insert (const uint64_t *hashes, T n=1)

T	contains_insert (const std::vector< uint64_t > &hashes, T n=1)

T	insert_contains (const char *seq, size_t seq_len)

T	insert_contains (const std::string &seq)

T	insert_contains (const uint64_t *hashes, T n=1)

T	insert_contains (const std::vector< uint64_t > &hashes, T n=1)

T	insert_thresh_contains (const char *seq, size_t seq_len, T threshold)

T	insert_thresh_contains (const std::string &seq, const T threshold)

T	insert_thresh_contains (const uint64_t *hashes, const T threshold)

T	insert_thresh_contains (const std::vector< uint64_t > &hashes, const T threshold)

T	contains_insert_thresh (const char *seq, size_t seq_len, T threshold)

T	contains_insert_thresh (const std::string &seq, const T threshold)

T	contains_insert_thresh (const uint64_t *hashes, const T threshold)

T	contains_insert_thresh (const std::vector< uint64_t > &hashes, const T threshold)

size_t	get_bytes () const

uint64_t	get_pop_cnt (T threshold=1) const

double	get_occupancy (T threshold=1) const

unsigned	get_hash_num () const

double	get_fpr (T threshold=1) const

unsigned	get_k () const

const std::string &	get_hash_fn () const

CountingBloomFilter< T > &	get_counting_bloom_filter ()

void	save (const std::string &path)

Static Public Member Functions
static bool	is_bloom_file (const std::string &path)

Detailed Description

template<typename T>
class btllib::KmerCountingBloomFilter< T >

Counting Bloom filter data structure that stores k-mers. Provides KmerCountingBloomFilter8, KmerCountingBloomFilter16, and KmerCountingBloomFilter32 classes with corresponding bit-size counters.

Constructor & Destructor Documentation

◆ KmerCountingBloomFilter() [1/3]

template<typename T >

btllib::KmerCountingBloomFilter< T >::KmerCountingBloomFilter ( )

inline

Construct a dummy k-mer Bloom filter (e.g. as a default argument).

◆ KmerCountingBloomFilter() [2/3]

template<typename T >

btllib::KmerCountingBloomFilter< T >::KmerCountingBloomFilter	(	size_t	bytes,
		unsigned	hash_num,
		unsigned	k )

inline

Construct an empty k-mer Counting Bloom filter of given size.

Parameters

bytes	Filter size in bytes.
hash_num	Number of hash values per element.
k	K-mer size.

◆ KmerCountingBloomFilter() [3/3]

template<typename T >

btllib::KmerCountingBloomFilter< T >::KmerCountingBloomFilter ( const std::string & path )

inlineexplicit

Load a k-mer Counting Bloom filter from a file.

Parameters

path	Filepath to load from.

Member Function Documentation

◆ clear() [1/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::clear	(	const char *	seq,
		size_t	seq_len )

inline

Set the counts of a sequence's k-mers to zero in the filter.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.

◆ clear() [2/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::clear ( const std::string & seq )

inline

Set the counts of a sequence's k-mers to zero in the filter.

Parameters

seq	Sequence to k-merize.

◆ clear() [3/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::clear ( const std::vector< uint64_t > & hashes )

inline

Set the counts of a sequence's k-mers to zero in the filter.

Parameters

hashes Integer vector of the k-mer's hash values.

◆ clear() [4/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::clear ( const uint64_t * hashes )

inline

Set the counts of a sequence's k-mers to zero in the filter.

Parameters

hashes Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.

◆ contains() [1/4]

template<typename T >

uint64_t btllib::KmerCountingBloomFilter< T >::contains	(	const char *	seq,
		size_t	seq_len ) const

inline

Query the counts of k-mers of a sequence.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.

Returns: The sum of counts of seq's k-mers found in the filter.

◆ contains() [2/4]

template<typename T >

uint64_t btllib::KmerCountingBloomFilter< T >::contains ( const std::string & seq ) const

inline

Query the counts of k-mers of a sequence.

Parameters

seq	Sequence to k-merize.

Returns: The sum of counts of seq's k-mers found in the filter.

◆ contains() [3/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains ( const std::vector< uint64_t > & hashes ) const

inline

Get a k-mer's count.

Parameters

hashes Integer vector of k-mer's hash values.

Returns: The count of the queried k-mer.

◆ contains() [4/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains ( const uint64_t * hashes ) const

inline

Get a k-mer's count.

Parameters

hashes Integer array of k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.

Returns: The count of the queried k-mer.

◆ contains_insert() [1/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert	(	const char *	seq,
		size_t	seq_len )

inline

Get the counts of a sequence's k-mers and then increment the counts.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.

Returns: The sum of counts of the queried k-mers before insertion.

◆ contains_insert() [2/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert ( const std::string & seq )

inline

Get the counts of a sequence's k-mers and then increment the counts.

Parameters

seq	Sequence to k-merize.

Returns: The sum of counts of the queried k-mers before insertion.

◆ contains_insert() [3/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert	(	const std::vector< uint64_t > &	hashes,
		T	n = 1 )

inline

Get the count of a k-mer and then increment the count.

Parameters

hashes	Integer vector of the k-mer's hash values.
n	Increment value

Returns: The count of the queried k-mer before insertion.

◆ contains_insert() [4/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert	(	const uint64_t *	hashes,
		T	n = 1 )

inline

Get the count of a k-mer and then increment the count.

Parameters

hashes	Integer array of the k-mers's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.
n	Increment value

Returns: The count of the queried k-mer before insertion.

◆ contains_insert_thresh() [1/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert_thresh	(	const char *	seq,
		size_t	seq_len,
		T	threshold )

inline

Get the counts of a sequence's k-mer's and then increment the counts if they are not above the threshold.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.
threshold	The threshold.

Returns: The sum of counts of the queried k-mers before insertion.

◆ contains_insert_thresh() [2/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert_thresh	(	const std::string &	seq,
		const T	threshold )

inline

Get the counts of a sequence's k-mer's and then increment the counts if they are not above the threshold.

Parameters

seq	Sequence to k-merize.
threshold	The threshold.

Returns: The sum of counts of the queried k-mers before insertion.

◆ contains_insert_thresh() [3/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert_thresh	(	const std::vector< uint64_t > &	hashes,
		const T	threshold )

inline

Get the count of a k-mer and then increment the count if it's not above the threshold.

Parameters

hashes	Integer vector of the k-mer's hash values.
threshold	The threshold.

Returns: The count of the queried k-mer before insertion.

◆ contains_insert_thresh() [4/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::contains_insert_thresh	(	const uint64_t *	hashes,
		const T	threshold )

inline

Get the count of a k-mer and then increment the count if it's not above the threshold.

Parameters

hashes	Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.
threshold	The threshold.

Returns: The count of the queried k-mer before insertion.

◆ get_bytes()

template<typename T >

size_t btllib::KmerCountingBloomFilter< T >::get_bytes ( ) const

inline

Get filter size in bytes.

◆ get_counting_bloom_filter()

template<typename T >

CountingBloomFilter< T > & btllib::KmerCountingBloomFilter< T >::get_counting_bloom_filter ( )

inline

Get a reference to the underlying vanilla Counting Bloom filter.

◆ get_fpr()

template<typename T >

double btllib::KmerCountingBloomFilter< T >::get_fpr ( T threshold = 1 ) const

inline

Get the query false positive rate for elements with count >= threshold.

Parameters

threshold The threshold.

Returns: The false positive rate.

◆ get_hash_fn()

template<typename T >

const std::string & btllib::KmerCountingBloomFilter< T >::get_hash_fn ( ) const

inline

Get the name of the hash function used.

◆ get_hash_num()

template<typename T >

unsigned btllib::KmerCountingBloomFilter< T >::get_hash_num ( ) const

inline

Get the number of hash values per element.

◆ get_k()

template<typename T >

unsigned btllib::KmerCountingBloomFilter< T >::get_k ( ) const

inline

Get the k-mer size used.

◆ get_occupancy()

template<typename T >

double btllib::KmerCountingBloomFilter< T >::get_occupancy ( T threshold = 1 ) const

inline

Get the fraction of the filter occupied by >0 counters.

◆ get_pop_cnt()

template<typename T >

uint64_t btllib::KmerCountingBloomFilter< T >::get_pop_cnt ( T threshold = 1 ) const

inline

Get population count, i.e. the number of counters >0 in the filter.

◆ insert() [1/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::insert	(	const char *	seq,
		size_t	seq_len )

inline

Insert a sequence's k-mers into the filter.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.

◆ insert() [2/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::insert ( const std::string & seq )

inline

Insert a sequence's k-mers into the filter.

Parameters

seq	Sequence to k-merize.

◆ insert() [3/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::insert	(	const std::vector< uint64_t > &	hashes,
		T	n = 1 )

inline

Insert a k-mer into the filter.

Parameters

hashes	Integer vector of the k-mer's hash values.
n	Increment value

◆ insert() [4/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::insert	(	const uint64_t *	hashes,
		T	n = 1 )

inline

Insert a k-mer into the filter.

Parameters

hashes	Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.
n	Increment value

◆ insert_contains() [1/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_contains	(	const char *	seq,
		size_t	seq_len )

inline

Increment the counts of a sequence's k-mers and then return the counts.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.

Returns: The sum of counts of the queried k-mers after insertion.

◆ insert_contains() [2/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_contains ( const std::string & seq )

inline

Increment the counts of a sequence's k-mers and then return the counts.

Parameters

seq	Sequence to k-merize.

Returns: The sum of counts of the queried k-mers after insertion.

◆ insert_contains() [3/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_contains	(	const std::vector< uint64_t > &	hashes,
		T	n = 1 )

inline

Increment a k-mer's count and then return the count.

Parameters

hashes	Integer vector of the k-mer's hash values.
n	Increment value

Returns: The count of the queried k-mer after insertion.

◆ insert_contains() [4/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_contains	(	const uint64_t *	hashes,
		T	n = 1 )

inline

Increment a k-mer's count and then return the count.

Parameters

hashes	Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.
n	Increment value

Returns: The count of the queried k-mer after insertion.

◆ insert_thresh_contains() [1/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_thresh_contains	(	const char *	seq,
		size_t	seq_len,
		T	threshold )

inline

Increment the counts of a sequence's k-mers if they are not above the threshold and then return the counts.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.
threshold	The threshold.

Returns: The sum of counts of the queried k-mers after insertion.

◆ insert_thresh_contains() [2/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_thresh_contains	(	const std::string &	seq,
		const T	threshold )

inline

Increment the counts of a sequence's k-mers if they are not above the threshold and then return the counts.

Parameters

seq	Sequence to k-merize.
threshold	The threshold.

Returns: The sum of counts of the queried k-mers after insertion.

◆ insert_thresh_contains() [3/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_thresh_contains	(	const std::vector< uint64_t > &	hashes,
		const T	threshold )

inline

Increment a k-mer's count if it's not above the threshold and then return the count.

Parameters

hashes	Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.
threshold	The threshold.

Returns: The count of the queried k-mer after insertion.

◆ insert_thresh_contains() [4/4]

template<typename T >

T btllib::KmerCountingBloomFilter< T >::insert_thresh_contains	(	const uint64_t *	hashes,
		const T	threshold )

inline

Increment a k-mer's count if it's not above the threshold and then return the count.

Parameters

hashes	Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.
threshold	The threshold.

Returns: The count of the queried k-mer after insertion.

◆ is_bloom_file()

template<typename T >

static bool btllib::KmerCountingBloomFilter< T >::is_bloom_file ( const std::string & path )

inlinestatic

Check whether the file at the given path is a saved k-mer counting Bloom filter.

Parameters

path	Filepath to check.

◆ remove() [1/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::remove	(	const char *	seq,
		size_t	seq_len )

inline

Decrease the counts of a sequence's k-mers from the filter.

Parameters

seq	Sequence to k-merize.
seq_len	Length of seq.

◆ remove() [2/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::remove ( const std::string & seq )

inline

Decrease the counts of a sequence's k-mers from the filter.

Parameters

seq	Sequence to k-merize.

◆ remove() [3/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::remove ( const std::vector< uint64_t > & hashes )

inline

Decrease the counts of a sequence's k-mers from the filter.

Parameters

hashes Integer vector of the k-mer's hash values.

◆ remove() [4/4]

template<typename T >

void btllib::KmerCountingBloomFilter< T >::remove ( const uint64_t * hashes )

inline

Decrease the counts of a sequence's k-mers from the filter.

Parameters

hashes Integer array of the k-mer's hash values. Array size should equal the hash_num argument used when the Bloom filter was constructed.

◆ save()

template<typename T >

void btllib::KmerCountingBloomFilter< T >::save ( const std::string & path )

inline

Save the Bloom filter to a file that can be loaded in the future.

Parameters

path	Filepath to store filter at.

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

include/btllib/counting_bloom_filter.hpp
include/btllib/counting_bloom_filter-inl.hpp

Public Member Functions

Static Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ KmerCountingBloomFilter() [1/3]

◆ KmerCountingBloomFilter() [2/3]

◆ KmerCountingBloomFilter() [3/3]

Member Function Documentation

◆ clear() [1/4]

◆ clear() [2/4]

◆ clear() [3/4]

◆ clear() [4/4]

◆ contains() [1/4]

◆ contains() [2/4]

◆ contains() [3/4]

◆ contains() [4/4]

◆ contains_insert() [1/4]

◆ contains_insert() [2/4]

◆ contains_insert() [3/4]

◆ contains_insert() [4/4]

◆ contains_insert_thresh() [1/4]

◆ contains_insert_thresh() [2/4]

◆ contains_insert_thresh() [3/4]

◆ contains_insert_thresh() [4/4]

◆ get_bytes()

◆ get_counting_bloom_filter()

◆ get_fpr()

◆ get_hash_fn()

◆ get_hash_num()

◆ get_k()

◆ get_occupancy()

◆ get_pop_cnt()

◆ insert() [1/4]

◆ insert() [2/4]

◆ insert() [3/4]

◆ insert() [4/4]

◆ insert_contains() [1/4]

◆ insert_contains() [2/4]

◆ insert_contains() [3/4]

◆ insert_contains() [4/4]

◆ insert_thresh_contains() [1/4]

◆ insert_thresh_contains() [2/4]

◆ insert_thresh_contains() [3/4]

◆ insert_thresh_contains() [4/4]

◆ is_bloom_file()

◆ remove() [1/4]

◆ remove() [2/4]

◆ remove() [3/4]

◆ remove() [4/4]

◆ save()