comparison formatter.py @ 4:966b4134ad12 draft

planemo upload for repository https://github.com/RECETOX/galaxytools/tree/master/tools/matchms commit 5661cf2406e0616d7b2f4bee1b57ec43716088de

3:574c6331e9db

import click
from pandas import DataFrame, read_csv, to_numeric


def create_long_table(data: DataFrame, value_id: str) -> DataFrame:
    """Convert the table from compact into long format.
    See DataFrame.melt(...).

    filtered = data[data['score'] > t_score]
    filtered = filtered[filtered['matches'] > t_matches]
    return filtered


def load_data(scores_filename: str, matches_filename: str) -> DataFrame:
    """Load data from filenames and join on compound id.

    Args:
        scores_filename (str): Path to scores table.
        matches_filename (str): Path to matches table.

    Returns:
        DataFrame: Joined dataframe on compounds containing scores an matches in long format.
    """
    matches = read_csv(matches_filename, sep="\t", index_col=0, header=0).apply(to_numeric)
    scores = read_csv(scores_filename, sep="\t", index_col=0, header=0).apply(to_numeric)

    scores_long = create_long_table(scores, 'score')
    matches_long = create_long_table(matches, 'matches')

    combined = join_df(matches_long, scores_long, on=['compound'], how='inner')
    return combined


@click.group()
@click.option('--sf', 'scores_filename', type=click.Path(exists=True), required=True)
@click.option('--mf', 'matches_filename', type=click.Path(exists=True), required=True)
@click.option('--o', 'output_filename', type=click.Path(writable=True), required=True)
@click.pass_context
def cli(ctx, scores_filename, matches_filename, output_filename):
    ctx.ensure_object(dict)
    ctx.obj['data'] = load_data(scores_filename, matches_filename)
    pass


@cli.command()
@click.option('--st', 'scores_threshold', type=float, required=True)

def get_top_k_data(ctx, k):
    result = get_top_k_matches(ctx.obj['data'], k)
    return result


@cli.resultcallback()
def write_output(result: DataFrame, scores_filename, matches_filename, output_filename):
    result = result.reset_index().rename(columns={'level_0': 'query', 'compound': 'reference'})
    result.to_csv(output_filename, sep="\t", index=False)


if __name__ == '__main__':

4:966b4134ad12

import click
from matchms.importing import scores_from_json
from pandas import DataFrame


def create_long_table(data: DataFrame, value_id: str) -> DataFrame:
    """Convert the table from compact into long format.
    See DataFrame.melt(...).

    filtered = data[data['score'] > t_score]
    filtered = filtered[filtered['matches'] > t_matches]
    return filtered


def scores_to_dataframes(scores):
    """Unpack scores from matchms.scores into two dataframes of scores and matches.

    Args:
        scores (matchms.scores): matchms.scores object.

    Returns:
        DataFrame: Scores
        DataFrame: Matches
    """
    query_names = [spectra.metadata['compound_name'] for spectra in scores.queries]
    reference_names = [spectra.metadata['compound_name'] for spectra in scores.references]

    dataframe_scores = DataFrame(data=[entry["score"] for entry in scores.scores], index=reference_names, columns=query_names)
    dataframe_matches = DataFrame(data=[entry["matches"] for entry in scores.scores], index=reference_names, columns=query_names)

    return dataframe_scores, dataframe_matches


def load_data(scores_filename: str) -> DataFrame:
    """Load data from filenames and join on compound id.

    Args:
        scores_filename (str): Path to json file with serialized scores.

    Returns:
        DataFrame: Joined dataframe on compounds containing scores and matches in long format.
    """
    scores = scores_from_json(scores_filename)
    scores, matches = scores_to_dataframes(scores)

    scores_long = create_long_table(scores, 'score')
    matches_long = create_long_table(matches, 'matches')

    combined = join_df(matches_long, scores_long, on=['compound'], how='inner')
    return combined


@click.group()
@click.option('--sf', 'scores_filename', type=click.Path(exists=True), required=True)
@click.option('--o', 'output_filename', type=click.Path(writable=True), required=True)
@click.pass_context
def cli(ctx, scores_filename, output_filename):
    ctx.ensure_object(dict)
    ctx.obj['data'] = load_data(scores_filename)
    pass


@cli.command()
@click.option('--st', 'scores_threshold', type=float, required=True)

def get_top_k_data(ctx, k):
    result = get_top_k_matches(ctx.obj['data'], k)
    return result


@cli.result_callback()
def write_output(result: DataFrame, scores_filename, output_filename):
    result = result.reset_index().rename(columns={'level_0': 'query', 'compound': 'reference'})
    result.to_csv(output_filename, sep="\t", index=False)


if __name__ == '__main__':