Batch collation

Batch collation#

We use this tutorial to batch collate a list of proteins.

Reminder of data

Please make sure that the build-in example dataset has been downloaded before you walk through the tutorial.

Example usage#

We can first define a few Pandas dataframes to store the information about 4 example proteins: 6cxh chain C, 3pux chain G, 5guf chain A, and 4kjs chain A.

Attribute

Description

prot_df

stores the information about the four chains

prot_pdbtm_df

stores the information about all chains of the protein PDBTM complexes that the four chains are derived from

prot_rcsb_df

stores the information about all chains of the protein RCSB complexes that the four chains are derived from

 
import pandas as pd

prot_df = pd.DataFrame(
    [['6cxh', 'C'],
     ['3pux', 'G'],
     ['5guf', 'A'],
     ['4kjs', 'A'],]
)

prot_pdbtm_df = pd.DataFrame(
    [['3pux', 'E'],
    ['3pux', 'F'],
    ['3pux', 'G'],
    ['3pux', 'A'],
    ['3pux', 'B'],
    ['5guf', 'A'],
    ['5guf', 'B'],
    ['6cxh', 'A'],
    ['6cxh', 'B'],
    ['6cxh', 'C'],
    ['6cxh', 'D'],
    ['6cxh', 'E'],
    ['6cxh', 'F'],
    ['6cxh', 'G'],
    ['6cxh', 'H'],
    ['4kjs', 'A'],
    ['4kjs', 'C'],
    ['4kjs', 'D'],]
)

prot_rcsb_df = pd.DataFrame(
    [['4kjs', 'A'],
     ['4kjs', 'B'],
     ['6cxh', 'A'],
     ['6cxh', 'B'],
     ['6cxh', 'C'],
     ['5guf', 'A'],
     ['3pux', 'E'],
     ['3pux', 'F'],
     ['3pux', 'G'],
     ['3pux', 'A'],
     ['3pux', 'B'],],
)

We can use the following codes to batch collate them. If you have a large number of protein complexes, the tmk.collate.batch should be very effective for your needs.

import tmkit as tmk

df, transformation_detection = tmk.collate.batch(
    prot_df=prot_df,
    prot_pdbtm_df=prot_pdbtm_df,
    prot_rcsb_df=prot_rcsb_df,
    pdb_rcsb_fp=pdb_rcsb_fp,
    pdb_pdbtm_fp=pdb_pdbtm_fp,
)
print(df)
print(transformation_detection)

Attributes#

Attribute

Description

pdb_rcsb_fp

path where protein complexes from RCSB are placed

pdb_pdbtm_fp

path where protein complexes from PDBTM are placed

prot_df

tab-delimiter Pandas dataframe containing protein names and protein chains in two columns, respectively

prot_pdbtm_df

tab-delimiter Pandas dataframe containing protein names and all of the chains of the protein (from PDBTM) in two columns, respectively

prot_rcsb_df

tab-delimiter Pandas dataframe containing protein names and all of the chains of the protein (from RCSB) in two columns, respectively

See also

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

Output#

      0  1     pdbtm   rcsb   diff source
0  6cxh  C  ABCDEFGH    ABC  FDHEG   rcsb
1  3pux  G     EFGAB  EFGAB          rcsb
2  5guf  A        AB      A      B   rcsb
3  4kjs  A       ACD     AB     CD   rcsb

File format#

It returns two output objects. protein collated df - It contains 6 columns:

Attribute

Description

0

protein name

1

chain name

pdbtm

DB

rcsb

DB

diff

different chains (pdbtm and rcsb)

source

a chain(s) in PDBTM exists in the chains in RCSB, which means if this chain(s) in PDBTM is transformed using the BIOMAT 350 records

If strategy='diff' is selected and values in column source are shown rcsb, which means all chains of a self.prot_df in PDBTM can be found in RCSB. strategy_dict stores the same or different chains between PDBTM and RCSB. throwback IS the path to the protein complex from PDBTM.

Contents