Function: encode

def encode(sequences, mapping, default_value, padding_value=None, pad_type="after", pad_length=0, n_jobs=1) -> list[np.ndarray] | np.ndarray:

Description

Encode sequences according to a mapping defined by the user.

Parameters

Parameter	Type	Default	Description
sequences	any iterable containing str	Required	Sequences to encode, can also be dna-parser SequenceReader, pysam FastxFile or Biopython SeqIO.read objects.
mapping	dict[str, int/float] or dict[str, 1D array]	Required	Mappings used to encode the sequences. The mapping can map a character to a value or to a 1D list/numpy.ndarray
default_value	int/float/1D array	Required	Value to use if a character encountered in a sequence is not in mapping. Must be of the same type than the values in mapping
padding_value	int/float/1D array/None	None	Value used to pad the encoding. If Nonedefaults to default_value
pad_type	str	after	Whether to trim/pad at the front or at the back of the sequences. Options are before and after. This paramater is ignored if there is no padding
pad_length	int	0	Length for padding/trimming. -2 to pad to the longest sequence, -1 to trim to the shortest sequence, any positive integer for a fixed length.
n_jobs	int	1	number of threads used to encode the sequences in parallel. 0 to use all CPUs available.

Returns

Type	Description
list[numpy.ndarray] or numpy.ndarray	The encoded sequences. Returns a list of numpy arrays if there is no trimming/padding, otherwise returns a numpy array with the outermost dimension representing sequences.

Examples

import dna_parser as dps

sequences= ["agt-","acc-"]
mapping= {"a":0.2, "g":-0.1, "t":-0.2, "c":0.3}
encoding= dps.encode(sequences, mapping, default_value=-1.0)
print(encoding)

# Output:
#[array([ 0.2, -0.1, -0.2, -1. ]), array([ 0.2,  0.3,  0.3, -1. ])]


reader= dps.SequenceReader("path/to/file") # file with sequences agt, acc, attagc
mapping= {"a":[0,4], "g":[0,3], "t":[0,2], "c":[0,1]}
encoding= dps.encode(reader, mapping, default_value=[0,0], padding_value=[-1,-1], pad_type="after", pad_length=-2) #pad after the sequences to the longest one
print(encoding)
print(encoding.shape)

# Output:
#[[[ 0.  4.]
#  [ 0.  3.]
#  [ 0.  2.]
#  [-1. -1.]
#  [-1. -1.]
#  [-1. -1.]]
#
# [[ 0.  4.]
#  [ 0.  1.]
#  [ 0.  1.]
#  [-1. -1.]
#  [-1. -1.]
#  [-1. -1.]]
#
# [[ 0.  4.]
#  [ 0.  2.]
#  [ 0.  2.]
#  [ 0.  4.]
#  [ 0.  3.]
#  [ 0.  1.]]]
#  
#(3, 6, 2)