Function: atomic_encoding

def atomic_encoding(sequences, pad_type= "after", pad_length= 0, n_jobs= 1) -> list[np.ndarray] | np.ndarray:

Description

Encodes sequences converting each nucleotide to its atomic number. Returns a list of 1D numpy arrays if there is no trimming/padding, otherwise returns a 2D numpy array with rows representing sequences.

• A= 70
• C= 58
• G= 78
• T/U= 66
• Other characters or gaps = 0

See the About section for more details.

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.
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 1D numpy arrays if there is no trimming/padding, otherwise returns a 2D numpy array with rows representing sequences.

Examples

import dna_parser as dps

sequences= ["agt","acc"]
encoding= dps.atomic_encoding(sequences)
print(encoding)

# Output:
#[array([70, 78, 66], dtype=int32), array([70, 58, 58], dtype=int32)]


reader= dps.SequenceReader("path/to/file") # file with sequences agt, acc, attagc
encoding= dps.atomic_encoding(reader, pad_type="after", pad_length=-2) #pad after the sequences to the longest one
print(encoding)
print(encoding.shape)

# Output:
#[[70 78 66  0  0  0]
# [70 58 58  0  0  0]
# [70 66 66 70 78 58]]
#
#(3, 6)