MyGeneExample - Gene list to DataFrame to CSV file

mygene_example.py

# -*- coding: utf-8 -*-
"""
Created on Fri Jan 13 18:18:52 2017

@author: sdhutchins

"""
#------------------------------------------------------------------------------
# Modules Used
#------------------------------------------------------------------------------
import mygene
import csv
import pandas as pd
import sys

# Import mygene.MyGeneInfo() search command
mg = mygene.MyGeneInfo()


#------------------------------------------------------------------------------
# Create a list of gene symbols/names for .csv file
#------------------------------------------------------------------------------
g = open('genes.csv')  # List of genes
genes_list = []   # Initialize a list of genes
genes_list.append('')
file2 = csv.reader(g)
for gene in file2:    # Format a list of genes
    genes = str(gene)
    genes = genes.replace("'", "")
    genes = genes.replace("[", "")
    genes = genes.replace("]", "")
    genes = genes.replace(" ", "_")
    genes_list.append(genes)
print(genes_list)

#------------------------------------------------------------------------------
# Set up Input to start command if gene list is correct
#------------------------------------------------------------------------------
"""
x = str(input('Is the input properly formatted? (Type Yes or No) '))
if x == 'Yes':
    print("\n" + "MyGene will start." + "\n")
else:
     raise SystemExit
"""
#------------------------------------------------------------------------------
# Use MyGene to get gene information
#------------------------------------------------------------------------------
"""
Call querymany method.
Scopes is your query, and it can be "entrezgene", "symbol" such as HTR1A, "mim" for omim id,
and "accession". FOr more, see: http://mygene.info/doc/query_service.html#available_fields
Input is "symbol" in this example. Scroll to the bottom of this script for a list of genes I used.

Set as_dataframe to True will return a pandas dataframe object
Set verbose to False as this will suppress the messages like "finished".
The resuls will be a list of dictionaries.
The dictionary contains the entrezid for the "entrezgene" field.
If you want the ensembl ids, use fields='ensembl.gene'

List of fields: http://mygene.info/metadata/fields
Fields can be set to 'all' for all fields to return.

There are also multiple species available or you can input the Taxonomy ID.

Examples:
entrez_ids = mg.querymany(genes_list, scopes='symbol,ensembl.gene', fields='entrezgene',
                          species='human', returnall=True, as_dataframe=True)

ensembl_ids = mg.querymany(genes_list, scopes='symbol', fields='ensembl.gene',
                          species='9606', returnall=True, as_dataframe=True)
"""
# This creates a dictionary of basic human gene information to be used later
basic_gene_info = mg.querymany(genes_list, scopes='symbol',
                          fields='symbol,name,entrezgene,summary',
                           species='human', returnall=True, as_dataframe=True,
                           size=1)

#------------------------------------------------------------------------------
# Use pandas to turn results of the mygene queries into dataframes
#------------------------------------------------------------------------------
"""
Use dict.keys() or basic_info.keys() to find out what the data keys are.
The data keys will be 'out' for output, 'missing' for any missing genes, 'dup' for any duplicates
Write the dataframe to a csv file using pandas (it saved as a dataframe).
Save the data as a .csv file.
Use df.drop to delete columns of the data you don't want.

Additional dictionary command:
To return a dictionary of MyGene.info metadata, use metadata = mg.metadata
"""
# Turn the dict into a pandas csv file
basic_gene_info['out'].to_csv('basic_gene_info.csv', sep=',', encoding='utf-8')
df = pd.read_csv('basic_gene_info.csv')
data = df
gene_info = pd.DataFrame(data)
gene_info.drop(data.columns[[1,2,6]], axis=1, inplace=True)

# Rename the columns
gene_info.rename(columns={'entrezgene': 'Entrez ID','summary':
    'Gene Summary','query': 'Gene Symbol','name': 'Gene Name'}, inplace=True)

gene_info.to_csv('basic_gene_info.csv', index=False)


"""
List of Genes I used (I saved them to a csv file - genes.csv in this example)
ADRA1A
ADRA1B
ADRA1D
ADRA2A
ADRA2B
CHRM1
CHRM2
CHRM3
CHRM5
CNR1
CNR2
DRD2
DRD3
GABBR2
HTR1A
HTR1D
HTR1F
HTR2A
HTR2B
HTR4
HTR5A
HTR7
OPRK1
OPRM1
"""

Collection of Data Using MyGene

Background Information

I'm currently working on a project in which the data consists of alignments, phylogenetic trees, and PAML analysis of over 100 genes across 60+ species. Lately, I've been trying to find a way to easily get the summary of those genes for a smaller but related project.

Notable Links:
MyGene Python Docs | MyGene Website | MyGene GitHub | Script's Inspiration | Tommy's Github