uniprotparser

UniProt Database Web Parser Project

TLDR: This parser can be used to parse UniProt accession id and obtain related data from the UniProt web database.

To use:

python -m pip install uniprotparser

or

python3 -m pip install uniprotparser

With version 1.2.0, we have exposed to and from mapping parameters for UniProt API where you can indicate which database you want to map to and from.

from uniprotparser import get_from_fields, get_to_fields

#to get all available fields to map from

from_fields = get_from_fields()
print(from_fields)

#to get all available fields to map to
to_fields = get_to_fields()
print(to_fields)

These parameters can be passed to the parse method of the UniprotParser class as follow

from uniprotparser.betaparser import UniprotParser

parser = UniprotParser()
for p in parser.parse(ids=["P06493"], to_key="UniProtKB", from_key="UniProtKB_AC-ID"):
    print(p)

With version 1.1.0, a simple CLI interface has been added to the package.

Usage: uniprotparser [OPTIONS]

Options:
  -i, --input FILENAME   Input file containing a list of accession ids
  -o, --output FILENAME  Output file
  --help                 Show this message and exit.

With version 1.0.5, support for asyncio through aiohttp has been added to betaparser. Usage can be seen as follow

from uniprotparser.betaparser import UniprotParser
from io import StringIO
import asyncio
import pandas as pd

async def main():
    example_acc_list = ["Q99490", "Q8NEJ0", "Q13322", "P05019", "P35568", "Q15323"]
    parser = UniprotParser()
    df = []
    #Yield result for 500 accession ids at a time
    async for r in parser.parse_async(ids=example_acc_list):
        df.append(pd.read_csv(StringIO(r), sep="\t"))
    
    #Check if there were more than one result and consolidate them into one dataframe
    if len(df) > 0:
        df = pd.concat(df, ignore_index=True)
    else:
        df = df[0]

asyncio.run(main())

With version 1.0.2, support for the new UniProt REST API have been added under betaparser module of the package.

In order to utilize this new module, you can follow the example bellow

from uniprotparser.betaparser import UniprotParser
from io import StringIO

import pandas as pd
example_acc_list = ["Q99490", "Q8NEJ0", "Q13322", "P05019", "P35568", "Q15323"]
parser = UniprotParser()
df = []
#Yield result for 500 accession ids at a time
for r in parser.parse(ids=example_acc_list):
    df.append(pd.read_csv(StringIO(r), sep="\t"))

#Check if there were more than one result and consolidate them into one dataframe
if len(df) > 0:
    df = pd.concat(df, ignore_index=True)
else:
    df = df[0]

To parse UniProt accession with legacy API

from uniprotparser.parser import UniprotSequence

protein_id = "seq|P06493|swiss"

acc_id = UniprotSequence(protein_id, parse_acc=True)

#Access ACCID
acc_id.accession

#Access isoform id
acc_id.isoform

To get additional data from UniProt online database

from uniprotparser.parser import UniprotParser
from io import StringIO
#Install pandas first to handle tabulated data
import pandas as pd

protein_accession = "P06493"

parser = UniprotParser([protein_accession])

#To get tabulated data
result = []
for i in parser.parse("tab"):
    tab_data = pd.read_csv(i, sep="\t")
    last_column_name = tab_data.columns[-1]
    tab_data.rename(columns={last_column_name: "query"}, inplace=True)
    result.append(tab_data)
fin = pd.concat(result, ignore_index=True)

#To get fasta sequence
with open("fasta_output.fasta", "wt") as fasta_output:
    for i in parser.parse():
        fasta_output.write(i)