agglovar.align.score

Alignment routines.

Attributes

`AFFINE_SCORE_GAP`	Default affine gap scores.
`AFFINE_SCORE_MATCH`	Default match score.
`AFFINE_SCORE_MISMATCH`	Default mismatch score.
`AFFINE_SCORE_TS`	Default template switch score.
`DEFAULT_ALIGN_SCORE_MODEL`	Default alignment score model.

Classes

`AffineScoreModel`	Affine score model with default values modeled on minimap2 (2.26) default parameters.
`ScoreModel`	Score model interface.

Functions

`get_affine_by_params`(→ AffineScoreModel)	Parse a string to get alignment parameters from it.
`get_score_model`(→ ScoreModel)	Get score model from a string of alignment parameters.

Module Contents

class agglovar.align.score.AffineScoreModel(match: float = AFFINE_SCORE_MATCH, mismatch: float = AFFINE_SCORE_MISMATCH, affine_gap: Iterable[tuple[float, float]] = AFFINE_SCORE_GAP, template_switch: float = AFFINE_SCORE_TS)

Bases: ScoreModel

Affine score model with default values modeled on minimap2 (2.26) default parameters.

Variables:

match – Match score.
mismatch – Mismatch penalty.
affine_gap – An iterable containing tuples of two elements (gap open, gap extend).
template_switch – Template switch penalty. If none, defaults to 2x the penalty of a 50 bp gap.

__eq__(other: Any) → bool

Determine if this scoremodel is the same as another.

Parameters:: other – Other object.
Returns:: True if this scoremodel is equivalent to other.

__repr__(): Get a string representation of this score model.

gap(n: int = 1) → float

Score gap (insertion or deletion).

Compute all affine alignment gap score for each affine segment and return the highest value (least negative).

Parameters:: n – Size of gap.
Returns:: Gap score.

match(n: int = 1) → float

Score match.

Parameters:: n – Number of matching bases.
Returns:: Match score.

mismatch(n: int = 1) → float

Score mismatch.

Parameters:: n – Number of mismatched bases.
Returns:: Mismatch score.

mismatch_model() → Self

Get a mismatch model.

Returns:: A copy of this score model that does not penalize gaps. Used for computing the score of mismatches.

model_param_string() → str

Get score parameter string.

Returns:: A parameter string representing this score model.

score_operations(op_arr: numpy.ndarray) → float

Sum scores for affine models using a A vectorized implementation.

Parameters:: op_arr – Array of alignment operations (op_code: first column, op_len: second column).
Returns:: Sum of scores for each CIGAR operation.

template_switch() → float

Score a template switch.

Returns:: Score for a single template switch.

score_affine_gap: tuple

score_match: float

score_mismatch: float

score_template_switch: float

class agglovar.align.score.ScoreModel

Bases: abc.ABC

Score model interface.

abstractmethod __eq__(other: Any) → bool

Determine if this scoremodel is the same as another.

Parameters:: other – Other object.
Returns:: True if this scoremodel is equivalent to other.

abstractmethod __repr__(): Get a string representation of this score model.

abstractmethod gap(n: int = 1) → float

Score gap (insertion or deletion).

Compute all affine alignment gap score for each affine segment and return the highest value (least negative).

Parameters:: n – Size of gap.
Returns:: Gap score.

abstractmethod match(n: int = 1) → float

Score match.

Parameters:: n – Number of matching bases.
Returns:: Match score.

abstractmethod mismatch(n: int = 1) → float

Score mismatch.

Parameters:: n – Number of mismatched bases.
Returns:: Mismatch score.

abstractmethod mismatch_model() → Self

Get a mismatch model.

Returns:: A copy of this score model that does not penalize gaps. Used for computing the score of mismatches.

abstractmethod model_param_string() → str

Get score parameter string.

Returns:: A parameter string representing this score model.

score(op_code: int | str, op_len: int) → float

Score an alignment operation.

Parameters:

op_code – Operation code. Can be a symbol (e.g. “=”, “X”, “I”, etc.) or the numeric operation code.
op_len – Operation length.

Returns:

Score for this operation or 0.0 if the operation is not scored (S, H, N, and P operations).

score_operations(op_arr: numpy.ndarray) → float

Sum scores for affine models using a vectorized implementation.

Parameters:: op_arr – Array of alignment operations (op_code: first column, op_len: second column).
Returns:: Sum of scores for each CIGAR operation.

template_switch() → float

Score a template switch.

Returns:: Score for a single template switch.

agglovar.align.score.get_affine_by_params(param_string: str) → AffineScoreModel

Parse a string to get alignment parameters from it.

Parameters:: param_string – Parameter string.
Returns:: A configured AffineScoreModel.

agglovar.align.score.get_score_model(param_string: str = None) → ScoreModel

Get score model from a string of alignment parameters.

Parameters:: param_string – Parameter string. Can be None or an empty string (default model is used). If the string is an instance of ScoreModel, then the ScoreModel object is returned. Otherwise, the string is parsed and a score model object is returned.
Returns:: A ScoreModel object.

agglovar.align.score.AFFINE_SCORE_GAP = ((4.0, 2.0), (24.0, 1.0)): Default affine gap scores.

agglovar.align.score.AFFINE_SCORE_MATCH = 2.0: Default match score.

agglovar.align.score.AFFINE_SCORE_MISMATCH = 4.0: Default mismatch score.

agglovar.align.score.AFFINE_SCORE_TS = None: Default template switch score.

agglovar.align.score.DEFAULT_ALIGN_SCORE_MODEL = 'affine::match=2.0,mismatch=4.0,gap=': Default alignment score model.