How can I set a Specific chain name for a Structure?(BioStrcuture Package)

Recently, I was working on a project dealing with a large number of coordinate files, which required me to convert each chain in some coordinate files with MMCIF/PDBXML/MMTF/PDB format into PDB format. Some of the chains in MMCIF files have long chain names, which makes the writepdb function down due to chain name too long.

How can I set a Specific chain name for a Structure?

I have found a function in biopython to solve this problem:

set_chain_info ( self , chain_id , chain_name , num_groups )

Set the chain information.

Parameters

  • chain_id – the asym chain id from mmCIF
  • chain_name – the auth chain id from mmCIF
  • num_groups – the number of groups this chain has.

But I want to solve this problem in Julia, my script is as follows. How can I solve this problem?

using BioStructures #BioStructures Module to read and write pdb files.

"Module to split a Structure to many single chain pdb file."
function splitStructure(Structure,pdbid::AbstractString,library::AbstractString,singlechainDB::AbstractString)
    for model in Structure
        modelname = modelnumber(model)
        for chain in model
            chainname = chainid(chain)
            try
                writepdb("$singlechainDB/$pdbid-$modelname-$chainname.pdb",chain)
            catch e0
                writemmtf("$singlechainDB/$pdbid-$modelname-$chainname.cif",chain)
                println("$pdbid-$modelname-$chainname has been saved as cif")
            end
        end
    end
end

"Module to split a pdb file to many single chain pdb file."
function getStructuretospilt(pdbid::AbstractString,library::AbstractString,singlechainDB::AbstractString)
    try
        Structure = retrievepdb(pdbid, dir=library)
    catch e1
        try    
            file = downloadpdb(pdbid,dir=library,format=MMCIF)
            Structure = read(file, MMCIF)
        catch e2
            file = downloadpdb(pdbid,dir=library,format=MMTF)
            Structure = read(file, MMTF)
        end
    end
    splitStructure(Structure,pdbid,library,singlechainDB)
    Structure = nothing
end

"Module to download a pdb."
function downloadastructure(pdbid::AbstractString,library::AbstractString)
    try
        downloadpdb(pdbid,dir=library,format=PDB)
    catch e1
        try
            downloadpdb(pdbid,dir=library,format=MMCIF)
        catch e2
            try
                downloadpdb(pdbid,dir=library,format=MMTF)
            catch e3    
                downloadpdb(pdbid,dir=library,format=PDBXML)
            end
        end
    end
end

function main()
    print("What's path of your pdb library?\n")
    library=readline(stdin)
    print("What's path of your pdb list file?\n")
    pdblist=readline(stdin)
    print("What's path of your output pdb library?\n")
    singlechainDB=readline(stdin)
    existpdblist = String[]
    for file in readdir(library)
        filename = split(basename(file),".")[1]
        push!(existpdblist,filename)
    end
    for pdbline in readlines(pdblist)
        Threads.@threads for pdbid in split(pdbline,",")
            findfirst(isequal(pdbid),existpdblist) == nothing || getStructuretospilt(pdbid,library,singlechainDB)
        end
    end
end

main()

Thank you for your reading~

1 Like

You cannot do this currently in BioStructures.jl, sorry, though I’ve had a few requests for it so I’ll try and get it implemented. Follow this issue for more: https://github.com/BioJulia/BioStructures.jl/issues/22

Just a quick note that writemmtf("$singlechainDB/$pdbid-$modelname-$chainname.cif",chain) saves to MMTF format, not mmCIF as stated.

1 Like

thank you for your reply, BioStructures help me a lot, thank you for your contribution for it!

Yes, it’s a bug in my script, thank you for pointing it out~

1 Like

Hi, do you Remember my question about editing chain name for a Structure? I found a solution in PDBTools.jl, Hope this is useful for BioStructures.jl.

1 Like

but it still can’t solve the problem that some chain of MMCIF or MMTF has too long chain name to writing to pdb file…

That was basically why they invented the mmCIF format, to make up for that deficiency in PDB files.

I guess in the 1970s they never considered they would run out of single letter chain IDs within a PDB entry. Structural biology has come on a long way!

1 Like

yeah, mmCIF is more suitable than PDB for storing huge complex structures, however, most of the computational biology software use PDB format as the first choice…