Merge branch 'main' of https://huggingface.co/datasets/ProteinMPNN/SAbDab

src/sabdab_downloader.py

@@ -0,0 +1,438 @@
+description='''
+
+
+SAbDab Download Script                         \\\    //
+The OPIG Antibody Database                      \\\  //
+Authors: James Dunbar and Konrad Krawczyk 2013.   ||
+Contributors: Jinwoo Leem                         ||
+Supervisor: Charlotte Deane
+
+Contact: opig@stats.ox.ac.uk
+
+In collaboration with:
+UCB: Jiye Shi, Terry Baker.
+Roche: Angelika Fuchs, Guy Georges.
+
+
+o This is a script that allows a user to download data from SAbDab.
+o It requires a csv summary file downloaded from the website (opig.stats.ox.ac.uk/webapps/sabdab)
+
+o This file should contain AT LEAST:
+
+    1. A header line with tab-separated fields as "pdb    Hchain    Lchain    model"
+    2. The pdb identifier, heavy chain, light chain and model id on new lines e.g.
+
+    pdb    Hchain    Lchain    model
+    12e8   H         L         0
+    12e8   P         M         0
+    1ahw   B         A         0
+    1ahw   E         D         0
+    .      .         .         .
+    .      .         .         .
+    .      .         .         .
+
+o Other fields will be ignored but may be included in the file.
+
+o The user must provide a directory in which the data should be downloaded to.
+o The type of data that should be downloaded should be specified using the command-line options.
+
+o Example useage:
+    To run on a linux command line type:
+
+    python sabdab_downloader.py -s summary_file.csv -o path/to/output/ --original_pdb
+
+    This will create a directory in  "path/to/output/" name sabdab_dataset.
+    It will contain a directory for each unique pdb code in the summary_file.csv .
+    The structure for each of these pdbs will be downloaded there.
+'''
+
+epilogue="""
+Copyright (C) 2013 James Dunbar
+"""
+
+
+import argparse, sys, os, urllib.request
+
+def getpdb(pdb_entry, out_path):
+    """
+    Get the PDB file from sabdab.
+    Check that it has successfully downloaded.
+    """
+    out_file = os.path.join( out_path, "%s.pdb"%pdb_entry)
+    try:
+        urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/structure/%s.pdb"%(pdb_entry,pdb_entry), out_file)
+    except Exception as e:
+        print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/structure/%s.pdb failed"%(pdb_entry,pdb_entry))
+        return False
+    if os.path.isfile(out_file):
+        Retrieved = open(out_file).read()
+        if not Retrieved.count("ATOM"):
+            print("Failed to retrieve PDB file from SAbDab")
+            os.remove(out_file)
+            return False
+        else:
+            return True
+    else:
+        return False
+
+def getchothpdb(pdb_entry, out_path):
+    """
+    Get the chothia PDB file from sabdab.
+    Check that it has successfully downloaded.
+    """
+    out_file = os.path.join( out_path, "%s.pdb"%pdb_entry)
+    try:
+        urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/structure/chothia/%s.pdb"%(pdb_entry,pdb_entry), out_file)
+    except Exception as e:
+        print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/structure/chothia/%s.pdb failed"%(pdb_entry,pdb_entry))
+        return False
+    if os.path.isfile(out_file):
+        Retrieved = open(out_file).read()
+        if not Retrieved.count("ATOM"):
+            print("Failed to retrieve PDB file from SAbDab")
+            os.remove(out_file)
+            return False
+        else:
+            return True
+    else:
+        return False
+
+def getsequence(entry, fab_list, out_path):
+    """
+    Get the sequence files
+    Check that they successfully download
+    Put them into the directory
+    """
+
+    out_file = os.path.join( out_path, "%s_raw.pdb"%entry)
+    try:
+        urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_raw.fa"%(entry,entry), out_file)
+    except Exception as e:
+        print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_raw.fa failed"%(entry,entry))
+        return False
+    if os.path.isfile(out_file):
+        Retrieved = open(out_file).read()
+        if not Retrieved.count(">%s"%entry):
+            print("Failed to retrieve sequence file from SAbDab.")
+            os.remove(out_file)
+            return False
+    else:
+        return False
+
+    for fab in fab_list:
+        Hchain = fab[1].upper()
+        if Hchain!="NA":
+            out_file = os.path.join( out_path, "%s_%s_VH.fa"%(entry,Hchain) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_%s_VH.fa"%(entry,entry,Hchain), out_file)
+            except Exception as e:
+                Hchain = Hchain.lower()
+                out_file = os.path.join( out_path, "%s_%s_VH.fa"%(entry,Hchain) )
+                try:
+                    urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_%s_VH.fa"%(entry,entry,Hchain), out_file)
+                except Exception as e:
+                    print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_%s_VH.fa failed"%(entry,entry,Hchain))
+                    return False
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count(">%s"%entry):
+                    print("Failed to retrieve sequence file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+
+        Lchain = fab[2].upper()
+        if Lchain!="NA":
+            out_file = os.path.join( out_path, "%s_%s_VL.fa"%(entry,Lchain) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_%s_VL.fa"%(entry,entry,Lchain), out_file)
+            except Exception as e:
+                Lchain = Lchain.lower()
+                out_file = os.path.join( out_path, "%s_%s_VL.fa"%(entry,Lchain) )
+                try:
+                    urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_%s_VL.fa"%(entry,entry,Lchain), out_file)
+                except Exception as e:
+                    print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/sequences/%s_%s_VL.fa failed"%(entry,entry,Lchain))
+                    return False
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count(">%s"%entry):
+                    print("Failed to retrieve sequence file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+
+    return True
+
+def getannotation(entry, fab_list, out_path):
+    """
+    Get the annotation files for the antibody sequences.
+    These are for the variable region of the sequences only.
+    """
+    for fab in fab_list:
+        Hchain = fab[1].upper()
+        if Hchain!="NA":
+            out_file = os.path.join( out_path, "%s_%s_VH.ann"%(entry,Hchain) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/annotation/%s_%s_VH.ann"%(entry,entry,Hchain), out_file)
+            except Exception as e:
+                Hchain = Hchain.lower()
+                out_file = os.path.join( out_path, "%s_%s_VH.ann"%(entry,Hchain) )
+                try:
+                    urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/annotation/%s_%s_VH.ann"%(entry,entry,Hchain), out_file)
+                except Exception as e:
+                    print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/annotation/%s_%s_VH.ann failed"%(entry,entry,Hchain))
+                    return False
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count("H3"):
+                    print("Failed to retrieve annotation file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+
+        Lchain = fab[2].upper()
+        if Lchain!="NA":
+            out_file = os.path.join( out_path, "%s_%s_VL.ann"%(entry,Lchain) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/annotation/%s_%s_VL.ann"%(entry,entry,Lchain), out_file)
+            except Exception as e:
+                Lchain = Lchain.lower()
+                out_file = os.path.join( out_path, "%s_%s_VL.ann"%(entry,Lchain) )
+                try:
+                    urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/annotation/%s_%s_VL.ann"%(entry,entry,Lchain), out_file)
+                except Exception as e:
+                    print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/annotation/%s_%s_VL.ann failed"%(entry,entry,Lchain))
+                    return False
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count("L3"):
+                    print("Failed to retrieve annotation file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+    return True
+
+def getabangle(entry, fab_list, out_path):
+    """
+    Get the orientation angles for any of the fabs in the pdb.
+    A non-paired antibody chain e.g VHH will have NA as the other chain identifier.
+    """
+    for fab in fab_list:
+        if "NA" in fab:
+            continue
+        else:
+            out_file = os.path.join( out_path, "%s.abangle"%(entry) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/abangle/%s.abangle"%(entry,entry), out_file)
+            except Exception as e:
+                print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/abangle/%s.abangle failed"%(entry,entry))
+                return False
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count(entry):
+                    print("Failed to retrieve abangle file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+            return True
+    return True
+
+def getimgt(entry, fab_list, out_path):
+    """
+    Get the imgt files for the antibody sequences.
+    """
+    for fab in fab_list:
+        Hchain = fab[1].upper()
+        if Hchain!="NA":
+            out_file = os.path.join( out_path, "%s_%s_H.ann"%(entry,Hchain) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/imgt/%s_%s_H.imgt"%(entry,entry,Hchain), out_file)
+            except Exception as e:
+                Hchain = Hchain.lower()
+                out_file = os.path.join( out_path, "%s_%s_H.ann"%(entry,Hchain) )
+                try:
+                    urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/imgt/%s_%s_H.imgt"%(entry,entry,Hchain), out_file)
+                except Exception as e:
+                    print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/imgt/%s_%s_H.imgt failed"%(entry,entry,Hchain))
+                    return False
+            
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count("gene_type"):
+                    print("Failed to retrieve imgt file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+
+        Lchain = fab[2].upper()
+        if Lchain!="NA":
+            out_file = os.path.join( out_path, "%s_%s_L.ann"%(entry,Lchain) )
+            try:
+                urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/imgt/%s_%s_L.imgt"%(entry,entry,Lchain), out_file)
+            except Exception as e:
+                Lchain = Lchain.lower()
+                out_file = os.path.join( out_path, "%s_%s_L.ann"%(entry,Lchain) )
+                try:
+                    urllib.request.urlretrieve("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/imgt/%s_%s_L.imgt"%(entry,entry,Lchain), out_file)
+                except Exception as e:
+                    print("https://opig.stats.ox.ac.uk/webapps/abdb/entries/%s/imgt/%s_%s_L.imgt failed"%(entry,entry,Lchain))
+                    return False
+            if os.path.isfile(out_file):
+                Retrieved = open(out_file).read()
+                if not Retrieved.count("gene_type"):
+                    print("Failed to retrieve imgt file from SAbDab.")
+                    os.remove(out_file)
+                    return False
+            else:
+                return False
+
+    return True
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(prog="sabdab_downloader", description=description, epilog=epilogue,formatter_class=argparse.RawDescriptionHelpFormatter)
+    parser.add_argument( '--summary_file','-s',type=str, help="A tab-separated csv downloaded from SAbDab - https://opig.stats.ox.ac.uk/webapps/sabdab-sabpred/sabdab.", dest="summary_file")
+    parser.add_argument( '--output_path','-o',type=str, help="The path to the output directory.", dest="output_path")
+    parser.add_argument( '--original_pdb',action="store_true", help="Download the pdb structure(s).", dest="original_pdb")
+    parser.add_argument( '--chothia_pdb', action="store_true", help="Download the chothia re-numbered pdb structure(s).", dest="chothia_pdb")
+    parser.add_argument( '--sequences',action="store_true", help="Download the sequence information.", dest="sequence")
+    parser.add_argument( '--annotation',action="store_true", help="Download the chothia numbered sequence information.", dest="annotation")
+    parser.add_argument( '--abangle',action="store_true", help="Download the abangle angles.", dest="abangle")
+    parser.add_argument( '--imgt',action="store_true", help="Download the imgt information for the structure.", dest="imgt")
+
+    args= parser.parse_args()
+
+    if len(sys.argv)<2:
+        parser.print_help()
+        sys.exit(0)
+
+    #####################
+    #  Check the inputs #
+    #####################
+
+    if not args.summary_file:
+        print("No summary file found.")
+        sys.exit(1)
+    if not args.output_path:
+        print("No output path given.")
+        sys.exit(1)
+
+    if not (args.original_pdb or args.chothia_pdb or args.sequence or args.annotation or args.abangle or args.imgt):
+        print("No requested data type given. Please choose at least one.")
+
+    if not os.path.exists(args.output_path):
+        print("Output path does not exist.")
+        sys.exit(1)
+
+    if not os.path.isdir(args.output_path):
+        print("Output path is not a directory.")
+        sys.exit(1)
+
+    if not os.access(args.output_path, os.W_OK):
+        print("Output path is not writable.")
+        sys.exit(1)
+
+    # Set up output directory
+    output_path = os.path.join(args.output_path,"sabdab_dataset")
+    try:
+        os.mkdir(output_path)
+    except OSError:
+        print("A 'sabdab_dataset' already exists in the output directory. Please rename it or provide a new output directory.")
+        sys.exit(1)
+
+    # Get the summary data    
+    try:
+        with open(args.summary_file,'r') as input_file:
+            lines = input_file.readlines()
+            header = lines[0].strip().split("\t")[:4]
+            if header != ["pdb", "Hchain", "Lchain", "model"]:
+                raise IndexError
+            data={}
+            for line in lines[1:]:
+                if not line.strip(): continue
+                entry = line.strip().split("\t")[:4]
+                if len(entry) < 4 and not entry[0].isalnum():
+                    raise IndexError
+                try:
+                    data[entry[0].lower()].append(entry)
+                except KeyError:
+                    data[entry[0].lower()] = [entry]
+    except IOError:
+        print("Could not open summary file.")
+        sys.exit(1)
+    except IndexError:
+        print("Summary file in incorrect format.")
+        sys.exit(1)
+
+    for pdb_entry in data:
+        print("Getting data for %s"%pdb_entry)
+        got_data=False
+        pdb_entry_dir = os.path.join(output_path, pdb_entry)
+        os.mkdir(pdb_entry_dir)
+        if args.original_pdb or args.chothia_pdb:
+            struc_out_path = os.path.join(pdb_entry_dir,"structure")
+            os.mkdir(struc_out_path)
+            if args.original_pdb:
+                if getpdb(pdb_entry, struc_out_path):
+                    got_data=True
+                else:
+                    print(f"removed {struc_out_path}")
+                    # os.rmdir(struc_out_path)
+
+            if args.chothia_pdb:
+                choth_struc_out_path = os.path.join(struc_out_path,"chothia")
+                os.mkdir(choth_struc_out_path)
+                if getchothpdb(pdb_entry, choth_struc_out_path):
+                    got_data=True
+                else:
+                    print(f"removed {choth_struc_out_path}")
+                    # os.rmdir(choth_struc_out_path)
+
+        if args.sequence:
+            seq_out_path = os.path.join(pdb_entry_dir,"sequence")
+            os.mkdir(seq_out_path)
+            if getsequence(pdb_entry, data[pdb_entry] , seq_out_path):
+                got_data=True
+            else:
+                print(f"removed {seq_out_path}")
+                # os.rmdir(seq_out_path)
+
+        if args.annotation:
+            annotation_out_path = os.path.join(pdb_entry_dir,"annotation")
+            os.mkdir(annotation_out_path)
+            if getannotation(pdb_entry, data[pdb_entry] , annotation_out_path):
+                got_data=True
+            else:
+                print(f"removed {annotation_out_path}")
+                # os.rmdir(annotation_out_path)
+
+        if args.abangle:
+            abangle_out_path = os.path.join(pdb_entry_dir,"abangle")
+            os.mkdir(abangle_out_path)
+            if getabangle(pdb_entry, data[pdb_entry] , abangle_out_path):
+                got_data=True
+            else:
+                print(f"removed {abangle_out_path}")
+                # os.rmdir(abangle_out_path)
+
+        if args.imgt:
+            imgt_out_path = os.path.join(pdb_entry_dir,"imgt")
+            os.mkdir(imgt_out_path)
+            if getimgt(pdb_entry, data[pdb_entry] , imgt_out_path):
+                got_data=True
+            else:
+                print(f"removed {imgt_out_path}")
+                # os.rmdir(imgt_out_path)
+
+        if not got_data:
+            print(f"deleted {pdb_entry}")
+            # os.rmdir(pdb_entry_dir)
+
+# Failed or failed or deleted or removed