Extracting Metadata

After data are loaded and optionally filtered, a metadata table can be created with the hicutils.core.metadata.make_metadata_table function.

Examples

In [1]:

import hicutils as hu

df = hu.io.read_directory('example_data_immunedb')

Generate a metadata table by replicate¶

In [2]:

hu.metadata.make_metadata_table(df, 'replicate_name')

Out[2]:

	uniques	copies	cdr3_num_nts	avg_v_identity	in_frame	clones
replicate_name
IgH_HPAP015_rep1_200p0ng	19195	21250	53.4805	0.960210	0.8759	10000
IgH_HPAP015_rep2_200p0ng	19803	21765	53.3733	0.957881	0.8635	10000
IgH_HPAP017_rep1_200p0ng	19731	22617	55.1513	0.973640	0.8507	10000
IgH_HPAP017_rep2_200p0ng	23410	27914	54.7887	0.971429	0.8530	10000

Generate a metadata table by subject/disease¶

In [3]:

pdf = (
    df
    .pipe(hu.pooling.pool_by, 'disease') # Pool on disease
    .pipe(hu.filters.filter_by_overall_copies, 2) # Filter on 2+ copy
)

hu.metadata.make_metadata_table(pdf, ['subject', 'disease'])

Out[3]:

		uniques	copies	cdr3_num_nts	avg_v_identity	in_frame	clones
subject	disease
HPAP015	T1D	27487	31504	52.444321	0.962745	0.888642	8109
HPAP017	Control	31356	38746	54.364948	0.974526	0.870252	8108

API Documentation

hicutils.core.metadata.make_metadata_table(df, pool)

Generates a metadata table from a pooled DataFrame.

Parameters

dfpd.DataFrame

The DataFrame to use for the metadata table.

poolstr

The pooling column to use for each row of the table.

Returns

A metadata table, indexed by pool.