Contents

biotransformers.utils.compute_utils

biotransformers.utils.compute_utils

Module Contents

Classes

Mutation

register a mutation from a string

Functions

validate_position_str(position_str: str)

Checks positions str format

get_list_probs(mutation_list: List[Tuple[Mutation]], mutate_probs: SequenceProbsList, length_mutations: List[int]) → Tuple[List[List[float]], List[List[float]]]

This function build a list of mutate and native probabilities to compute

mutation_score(native_probs: List[float], mutate_probs: List[float]) → float

Compute mutate score based on Masked marginal probability

split_list(list_to_split: List, lengths_list: List) → List[List]

split a list in sublist

Attributes

ProbTuple

TokenProbsDict

SequenceProbsList

biotransformers.utils.compute_utils.ProbTuple
biotransformers.utils.compute_utils.TokenProbsDict
biotransformers.utils.compute_utils.SequenceProbsList
exception biotransformers.utils.compute_utils.InvalidPositionStringError

Bases: Exception

Raised when a position string is passed with incorrect format

biotransformers.utils.compute_utils.validate_position_str(position_str: str)

Checks positions str format

class biotransformers.utils.compute_utils.Mutation(mutation_str: str)

register a mutation from a string

Parameters

mutation (str) – string mutation format “A8U” -> “NativeIdMutant”

__repr__(self) → str

Return repr(self).

is_valid_mutation(self, sequence: str)

Check if mutation is valid for the sequence of AA :param sequence: protein sequence string :type sequence: str

biotransformers.utils.compute_utils.get_list_probs(mutation_list: List[Tuple[Mutation]], mutate_probs: SequenceProbsList, length_mutations: List[int]) → Tuple[List[List[float]], List[List[float]]]

This function build a list of mutate and native probabilities to compute the mutate_score. For each position in the mutate list, we catch the native probability and the mutate probability of this position. We do this for each sequence and return two lists : native_probs and mutate probs.

Parameters

  • mutation_list (List[Mutation]) – list with integer which are mutations

  • mutate_probs (List[Dict[Any]]) – probabilities for mutate sequence

  • length_mutations (List[int]) – length of indivual mutation for each sequence

biotransformers.utils.compute_utils.mutation_score(native_probs: List[float], mutate_probs: List[float]) → float

Compute mutate score based on Masked marginal probability Sum(log(p(xi=xi_mutate|x-M))-log(p(xi=xi_native|x-M))) over M (M s a mutation set)

Parameters

  • native_probs (List[ProbTuple]) – [description]

  • mutate_probs (List[ProbTuple]) – [description]

Returns

[description]

Return type

List[float]

biotransformers.utils.compute_utils.split_list(list_to_split: List, lengths_list: List) → List[List]

split a list in sublist

Parameters

  • list_to_split (List) – native list

  • lengths_list (List) – length of each sublist

Returns

List of sublist

Return type

[type]