Retrieve

Retrieve#

Accessing both the sequence and structure of a transmembrane protein is crucial. Typically, sequences are provided in the FASTA format, while structural data is stored in the Protein Data Bank (PDB) format.

In addition to these two standard formats, PDBTM introduced the Extensible Markup Language (XML) file, specifically designed to document transmembrane protein topologies in a structured and accessible manner.

In TMKit, the module that allows users to download protein files from different sources is tmkit.seq.

RCSB PDB file#

We can retrieve a RCSB PDB file. First, we need to specify all proteins of interest in a Pandas dataframe. In TMKit, the Pandas dataframe of proteins is recognised.

import pandas as pd

prot_series = pd.Series(['6e3y', '6rfq', '6t0b'])

You can save the files in ./data/pdb/. Then, please put the following code.

import tmkit as tmk

tmk.seq.retrieve_pdb_from_rcsb(
    prot_series=prot_series,
    sv_fp='./data/pdb/',
)

Attribute

Description

prot_series

data series of proteins

sv_fp

path to save the RCSB PDB file to be downloaded

Please see here for better understanding the file-naming system.

Output#

===>No.1 protein name: 6e3y
Downloading PDB structure '6e3y'...
======>successfully downloaded!
===>No.2 protein name: 6rfq
Downloading PDB structure '6rfq'...
======>successfully downloaded!
===>No.3 protein name: 6t0b
Downloading PDB structure '6t0b'...
======>successfully downloaded!

Retrieve a PDBTM PDB file#

Similarly, we can retrieve a PDBTM PDB file. Specifying all proteins of interest in a Pandas dataframe.

import pandas as pd

prot_series = pd.Series(['6e3y', '6rfq', '6t0b'])

You can save the files in ./data/pdb/. Then, putting the following code.

import tmkit as tmk

tmk.seq.retrieve_pdb_from_pdbtm(
    prot_series=prot_series,
    sv_fp='./data/pdb/pdbtm/',
)

Attribute

Description

prot_series

data series of proteins

sv_fp

path to save the RCSB PDB file to be downloaded

Please see here for better understanding the file-naming system.

Output#

===>No.1 protein name: 6e3y
======>successfully downloaded!
===>No.2 protein name: 6rfq
======>successfully downloaded!
===>No.3 protein name: 6t0b
======>successfully downloaded!

Retrieve a PDBTM XML file#

Similarly, we can retrieve a PDBTM PDB file. As introduced in the PDBTM official manual, there are many records in the PDBTM database, including

Attribute

Description

pdb_id

PDB code

ch_id

chain ID

type

topologies

title

TITLE section of PDB file

numtm

number of transmembrane segments

seq

sequence

n_ifh

number of interfacial helices

n_loop

number of loops

source

SOURCE section of PDB file

class

HEADER section of PDB file

keyword

keyword

creation

date of creation

lmod_date

date of last modification

lmod_descr

description of last mod

These records help researchers understand and screen transmembrane proteins. All records are placed in the XML file of a PDB protein file. In these records, <REGION> represents the topology of a protein, as shown in the table below.

Attribute

Description

1

Side1

2

Side2

B

Beta-strand

H

alpha-helix

C

coil

I

membrane-inside

L

membrane-loop

F

interfacial helix

U

unknown

Also, we can start specifying all proteins of interest in a Pandas dataframe.

import pandas as pd

prot_series = pd.Series(['6e3y', '6rfq', '6t0b'])

You can save the files in ./data/xml/. Then, putting the following code.

import tmkit as tmk

tmk.seq.retrieve_xml_from_pdbtm(
    prot_series=prot_series,
    sv_fp='./data/xml/',
)

Attribute

Description

prot_series

data series of proteins

sv_fp

path to save the RCSB PDB file to be downloaded

Please see here for better understanding the file-naming system.

Output#

===>No.1 protein name: 6e3y
======>successfully downloaded!
===>No.2 protein name: 6rfq
======>successfully downloaded!
===>No.3 protein name: 6t0b
======>successfully downloaded!

Retrieve a AlphaFold2 PDB file#

Since the emerging AlphaFold2[1] technology has swept the whole protein field, with a profound impact on the development of both experiment- and computation-driven structural studies, we added a few its related functions to TMKit ( we are also continuing to release more of those.

Also, we can start specifying all proteins of interest in a Pandas dataframe. Differently, we need to put the UniProt accession codes of the proteins, because they usually do not have determined structures in the PDB.

import pandas as pd

prot_series = pd.Series(['P63092', 'Q9B6E8', 'P07256', 'P63027'])

You can save the files in ./data/pdb/. Then, putting the following code.

import tmkit as tmk

tmk.seq.retrieve_pdb_alphafold(
    prot_series=prot_series,
    sv_fp='./data/pdb/',
)

Attribute

Description

prot_series

data series of proteins

sv_fp

path to save the RCSB PDB file to be downloaded

Please see here for better understanding the file-naming system.

Output#

===>No.0 protein name: P63092
======>successfully downloaded!
===>No.1 protein name: Q9B6E8
======>successfully downloaded!
===>No.2 protein name: P07256
======>successfully downloaded!
===>No.3 protein name: P63027
======>successfully downloaded!
Contents