Scoring

Detailed Description

The functions in this module implement a pipeline stage in which all the seed hits currently in the ScoringQueues get "extended" and scored using DP alignment against the reference genome.

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink
• HitQueues::opposite_score
• HitQueues::opposite_loc
• HitQueues::opposite_sink

Modules
	ScoringDetail

Functions
void	nvbio::bowtie2::cuda::score_best (const uint32 band_len, const BestApproxScoringPipelineState< EditDistanceScoringScheme > &pipeline, const ParamsPOD &params)
void	nvbio::bowtie2::cuda::score_best (const uint32 band_len, const BestApproxScoringPipelineState< SmithWatermanScoringScheme<> > &pipeline, const ParamsPOD &params)
uint32	nvbio::bowtie2::cuda::score_all (const uint32 band_len, const AllMappingPipelineState< EditDistanceScoringScheme > &pipeline, const ParamsPOD &params, const uint32 buffer_offset, const uint32 buffer_size)
uint32	nvbio::bowtie2::cuda::score_all (const uint32 band_len, const AllMappingPipelineState< SmithWatermanScoringScheme<> > &pipeline, const ParamsPOD &params, const uint32 buffer_offset, const uint32 buffer_size)
void	nvbio::bowtie2::cuda::anchor_score_best (const uint32 band_len, const BestApproxScoringPipelineState< EditDistanceScoringScheme > &pipeline, const ParamsPOD &params)
void	nvbio::bowtie2::cuda::anchor_score_best (const uint32 band_len, const BestApproxScoringPipelineState< SmithWatermanScoringScheme<> > &pipeline, const ParamsPOD &params)
void	nvbio::bowtie2::cuda::opposite_score_best (const BestApproxScoringPipelineState< EditDistanceScoringScheme > &pipeline, const ParamsPOD &params)
void	nvbio::bowtie2::cuda::opposite_score_best (const BestApproxScoringPipelineState< SmithWatermanScoringScheme<> > &pipeline, const ParamsPOD &params)
template<typename scheme_type >
uint32	nvbio::bowtie2::cuda::score_all_t (const uint32 band_len, const AllMappingPipelineState< scheme_type > &pipeline, const ParamsPOD params, const uint32 buffer_offset, const uint32 buffer_size)
template<typename scheme_type >
void	nvbio::bowtie2::cuda::score_best_t (const uint32 band_len, const BestApproxScoringPipelineState< scheme_type > &pipeline, const ParamsPOD params)
template<typename scheme_type >
void	nvbio::bowtie2::cuda::opposite_score_best_t (const BestApproxScoringPipelineState< scheme_type > &pipeline, const ParamsPOD params)
template<typename scheme_type >
void	nvbio::bowtie2::cuda::anchor_score_best_t (const uint32 band_len, const BestApproxScoringPipelineState< scheme_type > &pipeline, const ParamsPOD params)

Function Documentation

void nvbio::bowtie2::cuda::anchor_score_best	(	const uint32	band_len,
		const BestApproxScoringPipelineState< EditDistanceScoringScheme > &	pipeline,
		const ParamsPOD &	params
	)

execute a batch of banded-alignment score calculations for the anchor mates, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

void nvbio::bowtie2::cuda::anchor_score_best	(	const uint32	band_len,
		const BestApproxScoringPipelineState< SmithWatermanScoringScheme<> > &	pipeline,
		const ParamsPOD &	params
	)

execute a batch of banded-alignment score calculations for the anchor mates, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

template<typename scheme_type >

void nvbio::bowtie2::cuda::anchor_score_best_t	(	const uint32	band_len,
		const BestApproxScoringPipelineState< scheme_type > &	pipeline,
		const ParamsPOD	params
	)

execute a batch of banded-alignment score calculations for the anchor mates, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

Definition at line 251 of file score_paired_inl.h.

void nvbio::bowtie2::cuda::opposite_score_best	(	const BestApproxScoringPipelineState< EditDistanceScoringScheme > &	pipeline,
		const ParamsPOD &	params
	)

execute a batch of full-DP alignment score calculations for the opposite mates, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc
• HitQueues::score
• HitQueues::sink

outputs:

• HitQueues::opposite_score
• HitQueues::opposite_loc
• HitQueues::opposite_sink

void nvbio::bowtie2::cuda::opposite_score_best	(	const BestApproxScoringPipelineState< SmithWatermanScoringScheme<> > &	pipeline,
		const ParamsPOD &	params
	)

execute a batch of full-DP alignment score calculations for the opposite mates, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc
• HitQueues::score
• HitQueues::sink

outputs:

• HitQueues::opposite_score
• HitQueues::opposite_loc
• HitQueues::opposite_sink

template<typename scheme_type >

void nvbio::bowtie2::cuda::opposite_score_best_t	(	const BestApproxScoringPipelineState< scheme_type > &	pipeline,
		const ParamsPOD	params
	)

execute a batch of full-DP alignment score calculations for the opposite mates, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc
• HitQueues::score
• HitQueues::sink

outputs:

• HitQueues::opposite_score
• HitQueues::opposite_loc
• HitQueues::opposite_sink

uint32 nvbio::bowtie2::cuda::score_all	(	const uint32	band_len,
		const AllMappingPipelineState< EditDistanceScoringScheme > &	pipeline,
		const ParamsPOD &	params,
		const uint32	buffer_offset,
		const uint32	buffer_size
	)

execute a batch of single-ended banded-alignment score calculations, all-mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

Parameters

band_len	alignment band length
pipeline	all mapping pipeline
params	alignment params
buffer_offset	ring buffer offset
buffer_size	ring buffer size

Returns

number of valid alignments

uint32 nvbio::bowtie2::cuda::score_all	(	const uint32	band_len,
		const AllMappingPipelineState< SmithWatermanScoringScheme<> > &	pipeline,
		const ParamsPOD &	params,
		const uint32	buffer_offset,
		const uint32	buffer_size
	)

execute a batch of single-ended banded-alignment score calculations, all-mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

Parameters

band_len	alignment band length
pipeline	all mapping pipeline
params	alignment params
buffer_offset	ring buffer offset
buffer_size	ring buffer size

Returns

number of valid alignments

template<typename scheme_type >

uint32 nvbio::bowtie2::cuda::score_all_t	(	const uint32	band_len,
		const AllMappingPipelineState< scheme_type > &	pipeline,
		const ParamsPOD	params,
		const uint32	buffer_offset,
		const uint32	buffer_size
	)

execute a batch of single-ended banded-alignment score calculations, all-mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

Parameters

band_len	alignment band length
pipeline	all mapping pipeline
params	alignment params
buffer_offset	ring buffer offset
buffer_size	ring buffer size

Returns

number of valid alignments

Definition at line 239 of file score_all_inl.h.

void nvbio::bowtie2::cuda::score_best	(	const uint32	band_len,
		const BestApproxScoringPipelineState< EditDistanceScoringScheme > &	pipeline,
		const ParamsPOD &	params
	)

execute a batch of single-ended banded-alignment score calculations, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

void nvbio::bowtie2::cuda::score_best	(	const uint32	band_len,
		const BestApproxScoringPipelineState< SmithWatermanScoringScheme<> > &	pipeline,
		const ParamsPOD &	params
	)

execute a batch of single-ended banded-alignment score calculations, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

template<typename scheme_type >

void nvbio::bowtie2::cuda::score_best_t	(	const uint32	band_len,
		const BestApproxScoringPipelineState< scheme_type > &	pipeline,
		const ParamsPOD	params
	)

execute a batch of single-ended banded-alignment score calculations, best mapping

inputs:

• HitQueues::seed
• HitQueues::loc

outputs:

• HitQueues::score
• HitQueues::sink

Definition at line 213 of file score_best_inl.h.