API Reference: Similarity

This document describes the similarity calculation functions in wtv.similarity.

`dot_product_distance`

Calculate the dot product distance between two vectors.

Parameters

Parameter	Type	Description
`p`	np.ndarray	First vector
`q`	np.ndarray	Second vector

Returns

float: Dot product distance between the two vectors (0 to 1)

Example

import numpy as np
from wtv.similarity import dot_product_distance

p = np.array([1, 2, 3])
q = np.array([4, 5, 6])
score = dot_product_distance(p, q)
print(f"Similarity score: {score}")

`weighted_dot_product_distance`

Calculate the weighted dot product distance with fragment ratio consideration.

Parameters

Parameter	Type	Description
`compare_df`	pd.DataFrame	DataFrame with two columns representing vectors to compare
`fr_factor`	float	Factor used in the FR calculation

Returns

float: Composite score based on weighted dot product distance and fragment ratio

`calculate_similarity`

Calculate similarity scores between a target compound and all compounds in a DataFrame.

Parameters

Parameter	Type	Description
`target_name`	str	Name of the target compound
`df`	pd.DataFrame	DataFrame containing compound data
`fr_factor`	float	Factor used in the weighted dot product distance calculation

Returns

pd.DataFrame: DataFrame containing similarity scores for each compound

`calculate_average_score_and_difference_count`

Calculate the average similarity score and difference count for a targeted compound.

Parameters

Parameter	Type	Description
`targeted_compound`	str	Name of the targeted compound
`ion_combination`	list	List of ion combinations for similarity calculation
`df`	pd.DataFrame	DataFrame containing compound data
`similarity_threshold`	float	Similarity threshold for considering compounds as similar
`fr_factor`	float	Factor used in the weighted dot product distance calculation

Returns

pd.DataFrame: DataFrame containing difference counts and average similarity scores

`calculate_combination_score`

Calculate the combination score for ion combinations.

Parameters

Parameter	Type	Description
`combination_df`	pd.DataFrame	DataFrame containing ion combinations
`targeted_compound`	str	Name of the targeted compound
`temp_df`	pd.DataFrame	Temporary DataFrame containing compound data
`prefer_mz_threshold`	float	Preferred m/z threshold

Returns

pd.DataFrame: DataFrame with combination scores added

`calculate_solo_compound_combination_score`

Calculate the combination score for solo compounds.

Parameters

Parameter	Type	Description
`matrix_1`	pd.DataFrame	DataFrame containing solo compound data
`prefer_mz_threshold`	float	Preferred m/z threshold

Returns

pd.DataFrame: DataFrame with combination scores added and sorted by score