enrichR module

class magine.enrichment.enrichr.Enrichr(verbose=False)

Bases: object

print_valid_libs()

Print a list of all available libraries EnrichR has to offer.

run(list_of_genes, gene_set_lib='GO_Biological_Process_2017')

Parameters

list_of_geneslist_like

List of genes using HGNC gene names

gene_set_libstr or list

Name of gene set library To print options use Enrichr.print_valid_libs

Returns

dfEnrichmentResult

Results from enrichR

Examples

>>> import pandas as pd
>>> pd.set_option('display.max_colwidth', 40)
>>> pd.set_option('precision', 3)
>>> e = Enrichr()
>>> df = e.run(['BAX', 'BCL2', 'CASP3', 'CASP8'],                         gene_set_lib='Reactome_2016')
>>> print(df[['term_name','combined_score']].head(5))
                                     term_name  combined_score
    0  intrinsic pathway for apoptosis hsa ...       11814.410
    1               apoptosis hsa r-hsa-109581        2365.141
    2  programmed cell death hsa r-hsa-5357801        2313.527
    3  caspase-mediated cleavage of cytoske...       10944.261
    4  caspase activation via extrinsic apo...        4245.542

run_samples(sample_lists, sample_ids, gene_set_lib='GO_Biological_Process_2017', save_name=None, create_html=False, out_dir=None, run_parallel=False, exp_data=None, pivot=False)

Run enrichment analysis on a list of samples.

Parameters

sample_listslist_like

List of lists of genes for enrichment analysis

sample_idslist

list of ids for the provided sample list

gene_set_libstr, list

Type of gene set, refer to Enrichr.print_valid_libs

save_namestr, optional

if provided it will save a file as a pivoted table with the term_ids vs sample_ids

create_htmlbool

Creates html of output with plots of species across sample

out_dirstr

If create_html, it will place all html plots into this directory

run_parallelbool

If create_html, it will create plots using multiprocessing

exp_datamagine.data.ExperimentalData

Must be provided if create_html=True

pivotbool

Returns

EnrichmentResult

Examples

>>> import pandas as pd
>>> import matplotlib.pyplot as plt
>>> from magine.enrichment.enrichr import Enrichr
>>> pd.set_option('display.max_colwidth', 40)
>>> pd.set_option('precision', 3)
>>> samples = [['BAX', 'BCL2', 'CASP3', 'CASP8'], ['ATR', 'ATM', 'TP53', 'CHEK1']]
>>> sample_ids = ['apoptosis', 'dna_repair']
>>> e = Enrichr()
>>> df = e.run_samples(samples, sample_ids, gene_set_lib='Reactome_2016')
>>> print(df[['term_name','combined_score']].head(5))
                                     term_name  combined_score
    0  intrinsic pathway for apoptosis hsa ...       11814.410
    1               apoptosis hsa r-hsa-109581        2365.141
    2  programmed cell death hsa r-hsa-5357801        2313.527
    3  caspase-mediated cleavage of cytoske...       10944.261
    4  caspase activation via extrinsic apo...        4245.542
>>> df.filter_multi(rank=10, inplace=True)
>>> df['term_name'] = df['term_name'].str.split('_').str.get(0)
>>> fig = df.sig.heatmap(figsize=(6, 6), linewidths=.05)

Using a ExperimentalData instance, we can run enrichR for all the databases using a simple wrapper around.

magine.enrichment.enrichr.run_enrichment_for_project(exp_data, project_name, databases=None, output_path=None)

Parameters

exp_datamagine.data.experimental_data.ExprerimentalData

project_namestr

databaseslist

output_pathstr

Location to save all individual enrichment output files created.

We also provide some tools to clean up and standardize enrichRs output.

Functions to cleanup enrichR term names

Note: this are in progress and not fully tested! Warning!

magine.enrichment.enrichr.clean_term_names(row)

magine.enrichment.enrichr.clean_lincs(df)

Cleans the lincs databases term_names from enrichR.

Parameters

df

magine.enrichment.enrichr.clean_drug_pert_geo(df)

magine.enrichment.enrichr.clean_tf_names(data)

Cleans transcription factors databases by removing everything after ‘_’.

Parameters

datapd.DataFrame

Download reference databases

magine.enrichment.enrichr.get_background_list(lib_name)

Return reference list for given gene referecen set

Parameters

lib_namestr