pyproteonet.aggregation.partner_top_n_mean
Aggregates the values of partner molecule (e.g. peptides) into a molecule (e.g. protein) value by averaging the abundance values of the three highest abundant partner molecules. Partner molecule are determined according to a mapping.
- param dataset:
Dataset to run aggregation on.
- type dataset:
Dataset
- param molecule:
The molecule values should be aggregated for (e.g. protein).
- type molecule:
str
- param partner_column:
The columns of the partner molecule containing abundance values that should be aggregated.
- type partner_column:
str
- param mapping:
Either the name of the aggregated molecule (e.g. peptide) or the name of the mapping linking the molecule from above to a partner molecule (e.g. peptide-protein mapping)
- type mapping:
str
- param top_n:
Number of highest abundant partner molecules to consider. Defaults to 3.
- type top_n:
int, optional
- param only_unique:
Only consider unique peptides and ignore shared peptides. Defaults to True.
- type only_unique:
bool, optional
- param result_column:
If given aggregation results are stored in this alue column of the molecule. Defaults to None.
- type result_column:
Optional[str], optional
- param skip_if_less_than_n:
If True, the aggregation is skipped if there are less than n partner molecules and the aggregated values is set to NaN. Defaults to True.
- type skip_if_less_than_n:
bool, optional
- param is_log:
Wheter the input values are logarithmized. Defaults to False.
- type is_log:
bool, optional
- returns:
A pandas series with sample id and molecule id as multiindex containing the aggregated values.
- rtype:
pd.Series