Tokenizers (depthcharge.tokenizers)

PeptideTokenizer(residues=None, replace_isoleucine_with_leucine=False, reverse=False, start_token=None, stop_token='$')

Bases: Tokenizer

A tokenizer for ProForma peptide sequences.

Parse and tokenize ProForma-compliant peptide sequences.

PARAMETER	DESCRIPTION
residues	Residues and modifications to add to the vocabulary beyond the standard 20 amino acids. TYPE: dict[str, float] DEFAULT: None
replace_isoleucine_with_leucine	Replace I with L residues, because they are isomeric and often indistinguishable by mass spectrometry. TYPE: bool DEFAULT: False
reverse	Reverse the sequence for tokenization, C-terminus to N-terminus. TYPE: bool DEFAULT: False
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

ATTRIBUTE	DESCRIPTION
residues	The residues and modifications and their associated masses. terminal modifcations are indicated by -. TYPE: SortedDict[str, float]
index	The mapping of residues and modifications to integer representations. TYPE: SortedDict{str, int}
reverse_index	The ordered residues and modifications where the list index is the integer representation for a token. TYPE: list[None | str]
start_token	The start token TYPE: str
stop_token	The stop token. TYPE: str
start_int	The integer representation of the start token TYPE: int
stop_int	The integer representation of the stop token. TYPE: int
padding_int	The integer used to represent padding. TYPE: int

Functions

calculate_precursor_ions(tokens, charges)

Calculate the m/z for precursor ions.

PARAMETER	DESCRIPTION
tokens	The tokens corresponding to the peptide sequence. TYPE: torch.Tensor of shape (n_sequences, len_seq)
charges	The charge state for each peptide. TYPE: torch.Tensor of shape (n_sequences,)

RETURNS	DESCRIPTION
Tensor	The monoisotopic m/z for each charged peptide.

detokenize(tokens, join=True, trim_start_token=True, trim_stop_token=True)

Retreive sequences from tokens.

PARAMETER	DESCRIPTION
tokens	The zero-padded tensor of integerized tokens to decode. TYPE: torch.Tensor of shape (n_sequences, max_length)
join	Join tokens into strings? TYPE: bool DEFAULT: True
trim_start_token	Remove the start token from the beginning of a sequence. TYPE: bool DEFAULT: True
trim_stop_token	Remove the stop token from the end of a sequence. TYPE: bool DEFAULT: True

RETURNS	DESCRIPTION
list[str] or list[list[str]]	The decoded sequences each as a string or list or strings.

from_massivekb(replace_isoleucine_with_leucine=False, reverse=False, start_token=None, stop_token='$') staticmethod

Create a tokenizer with the observed peptide modications.

Modifications are parsed from MassIVE-KB peptide strings and added to the vocabulary.

PARAMETER	DESCRIPTION
replace_isoleucine_with_leucine	Replace I with L residues, because they are isobaric and often indistinguishable by mass spectrometry. TYPE: bool DEFAULT: False
reverse	Reverse the sequence for tokenization, C-terminus to N-terminus. TYPE: bool DEFAULT: False
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

RETURNS	DESCRIPTION
MskbPeptideTokenizer	A tokenizer for peptides with the observed modifications.

from_proforma(sequences, replace_isoleucine_with_leucine=False, reverse=False, start_token=None, stop_token='$') classmethod

Create a tokenizer with the observed peptide modications.

Modifications are parsed from ProForma 2.0-compliant peptide strings and added to the vocabulary.

PARAMETER	DESCRIPTION
sequences	The peptides from which to parse modifications. TYPE: Iterable[str]
replace_isoleucine_with_leucine	Replace I with L residues, because they are isobaric and often indistinguishable by mass spectrometry. TYPE: bool DEFAULT: False
reverse	Reverse the sequence for tokenization, C-terminus to N-terminus. TYPE: bool DEFAULT: False
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

RETURNS	DESCRIPTION
PeptideTokenizer	A tokenizer for peptides with the observed modifications.

split(sequence)

Split a ProForma peptide sequence.

PARAMETER	DESCRIPTION
sequence	The peptide sequence. TYPE: str

RETURNS	DESCRIPTION
list[str]	The tokens that comprise the peptide sequence.

tokenize(sequences, add_start=False, add_stop=False, to_strings=False)

Tokenize the input sequences.

PARAMETER	DESCRIPTION
sequences	The sequences to tokenize. TYPE: Iterable[str] or str
add_start	Prepend the start token to the beginning of the sequence. TYPE: bool DEFAULT: False
add_stop	Append the stop token to the end of the sequence. TYPE: bool DEFAULT: False
to_strings	Return each as a list of token strings rather than a tensor. This is useful for debugging. TYPE: bool DEFAULT: False

RETURNS	DESCRIPTION
torch.tensor of shape (n_sequences, max_length) or list[list[str]]	Either a tensor containing the integer values for each token, padded with 0's, or the list of tokens comprising each sequence.

MoleculeTokenizer(selfies_vocab=None, start_token=None, stop_token='$')

Bases: Tokenizer

A tokenizer for small molecules.

Tokenize SMILES and SELFIES representations of small molecules. SMILES are internally converted to SELFIES representations.

PARAMETER	DESCRIPTION
selfies_vocab	The SELFIES tokens to be considered. TYPE: Iterable[str] DEFAULT: None
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

ATTRIBUTE	DESCRIPTION
index	The mapping of residues and modifications to integer representations. TYPE: SortedDict{str, int}
reverse_index	The ordered residues and modifications where the list index is the integer representation for a token. TYPE: list[None | str]
start_token	The start token TYPE: str
stop_token	The stop token. TYPE: str
start_int	The integer representation of the start token TYPE: int
stop_int	The integer representation of the stop token. TYPE: int
padding_int	The integer used to represent padding. TYPE: int

Functions

detokenize(tokens, join=True, trim_start_token=True, trim_stop_token=True)

Retreive sequences from tokens.

PARAMETER	DESCRIPTION
tokens	The zero-padded tensor of integerized tokens to decode. TYPE: torch.Tensor of shape (n_sequences, max_length)
join	Join tokens into strings? TYPE: bool DEFAULT: True
trim_start_token	Remove the start token from the beginning of a sequence. TYPE: bool DEFAULT: True
trim_stop_token	Remove the stop token from the end of a sequence. TYPE: bool DEFAULT: True

RETURNS	DESCRIPTION
list[str] or list[list[str]]	The decoded sequences each as a string or list or strings.

from_selfies(selfies, start_token=None, stop_token='$') classmethod

Learn the vocabulary from SELFIES strings.

PARAMETER	DESCRIPTION
selfies	Create a vocabulary from all unique tokens in these SELFIES strings. TYPE: Iterable[str] | str
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

RETURNS	DESCRIPTION
MoleculeTokenizer	The tokenizer restricted to the vocabulary present in the input SMILES strings.

from_smiles(smiles, start_token=None, stop_token='$') classmethod

Learn the vocabulary from SMILES strings.

PARAMETER	DESCRIPTION
smiles	Create a vocabulary from all unique tokens in these SMILES strings. TYPE: Iterable[str] | str
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

RETURNS	DESCRIPTION
MoleculeTokenizer	The tokenizer restricted to the vocabulary present in the input SMILES strings.

split(sequence)

Split a SMILES or SELFIES string into SELFIES tokens.

PARAMETER	DESCRIPTION
sequence	The SMILES or SELFIES string representing a molecule. TYPE: str

RETURNS	DESCRIPTION
List[str]	The SELFIES tokens representing the molecule.

tokenize(sequences, add_start=False, add_stop=False, to_strings=False)

Tokenize the input sequences.

PARAMETER	DESCRIPTION
sequences	The sequences to tokenize. TYPE: Iterable[str] or str
add_start	Prepend the start token to the beginning of the sequence. TYPE: bool DEFAULT: False
add_stop	Append the stop token to the end of the sequence. TYPE: bool DEFAULT: False
to_strings	Return each as a list of token strings rather than a tensor. This is useful for debugging. TYPE: bool DEFAULT: False

RETURNS	DESCRIPTION
torch.tensor of shape (n_sequences, max_length) or list[list[str]]	Either a tensor containing the integer values for each token, padded with 0's, or the list of tokens comprising each sequence.

Tokenizer(tokens, start_token=None, stop_token='$')

Bases: ABC

An abstract base class for Depthcharge tokenizers.

PARAMETER	DESCRIPTION
tokens	The tokens to consider. TYPE: Sequence[str]
start_token	The start token to use. TYPE: str DEFAULT: None
stop_token	The stop token to use. TYPE: str DEFAULT: '$'

Functions

detokenize(tokens, join=True, trim_start_token=True, trim_stop_token=True)

Retreive sequences from tokens.

PARAMETER	DESCRIPTION
tokens	The zero-padded tensor of integerized tokens to decode. TYPE: torch.Tensor of shape (n_sequences, max_length)
join	Join tokens into strings? TYPE: bool DEFAULT: True
trim_start_token	Remove the start token from the beginning of a sequence. TYPE: bool DEFAULT: True
trim_stop_token	Remove the stop token from the end of a sequence. TYPE: bool DEFAULT: True

RETURNS	DESCRIPTION
list[str] or list[list[str]]	The decoded sequences each as a string or list or strings.

split(sequence) abstractmethod

Split a sequence into the constituent string tokens.

tokenize(sequences, add_start=False, add_stop=False, to_strings=False)

Tokenize the input sequences.

PARAMETER	DESCRIPTION
sequences	The sequences to tokenize. TYPE: Iterable[str] or str
add_start	Prepend the start token to the beginning of the sequence. TYPE: bool DEFAULT: False
add_stop	Append the stop token to the end of the sequence. TYPE: bool DEFAULT: False
to_strings	Return each as a list of token strings rather than a tensor. This is useful for debugging. TYPE: bool DEFAULT: False

RETURNS	DESCRIPTION
torch.tensor of shape (n_sequences, max_length) or list[list[str]]	Either a tensor containing the integer values for each token, padded with 0's, or the list of tokens comprising each sequence.