Pdbe API

query = "ligand_name:axitinib"

query = "molecule_name:kinase AND resolution:[0 TO 2.0]"

query = "revision_date:[20240101 TO 20241231]"

import json
from datetime import datetime

def check_structure_revision(pdb_id):
    """Check if structure was revised and when"""
    url = f"https://www.ebi.ac.uk/pdbe/api/pdb/entry/summary/{pdb_id}"
    result = coot.coot_get_url_as_string_py(url)
    
    if isinstance(result, bytes):
        result = result.decode('utf-8')
    
    data = json.loads(result)
    
    entry = data[pdb_id][0]
    release = entry['release_date']
    revision = entry['revision_date']
    
    # Convert to datetime for comparison
    release_dt = datetime.strptime(release, "%Y%m%d")
    revision_dt = datetime.strptime(revision, "%Y%m%d")
    
    days_diff = (revision_dt - release_dt).days
    years_diff = days_diff / 365.25
    
    print(f"PDB {pdb_id}:")
    print(f"  Released: {release}")
    print(f"  Revised: {revision}")
    print(f"  Time since release: {years_diff:.1f} years")
    
    if days_diff > 30:
        print(f"  WARNING: Structure revised {days_diff} days after release")
        return True
    return False

# Example usage
check_structure_revision("4wa9")

import json

def check_ligand_revision(comp_id):
    """Check when a ligand was last revised"""
    url = f"https://www.ebi.ac.uk/pdbe/api/pdb/compound/summary/{comp_id}"
    result = coot.coot_get_url_as_string_py(url)
    
    if isinstance(result, bytes):
        result = result.decode('utf-8')
    
    data = json.loads(result)
    
    compound = data[comp_id][0]
    print(f"Compound {comp_id} ({compound['name']}):")
    print(f"  Created: {compound['creation_date']}")
    print(f"  Revised: {compound['revision_date']}")
    
    if compound['creation_date'] != compound['revision_date']:
        print(f"  WARNING: Ligand definition was revised")
        return True
    return False

# Example usage
check_ligand_revision("AXI")

import gzip
import json

def download_and_load_structure(pdb_id):
    """Download and load a structure from RCSB"""
    url = f"https://files.rcsb.org/download/{pdb_id}.cif.gz"
    
    print(f"Downloading {pdb_id}...")
    compressed = coot.coot_get_url_as_string_py(url)
    
    # Check if it's an error response (HTML)
    if isinstance(compressed, str) and compressed.startswith("<!DOCTYPE"):
        print(f"ERROR: Could not download {pdb_id}")
        return None
    
    decompressed = gzip.decompress(compressed)
    imol = coot.read_coordinates_as_string(decompressed.decode('utf-8'), pdb_id)
    print(f"Loaded as molecule {imol}")
    
    return imol

def compare_species_structures(pdb_id1, pdb_id2):
    """Download two structures and superpose them"""
    # Download both structures
    imol1 = download_and_load_structure(pdb_id1)
    imol2 = download_and_load_structure(pdb_id2)
    
    if imol1 is None or imol2 is None:
        print("Failed to download one or both structures")
        return
    
    # Superpose (using CA atoms from chain A, residues 240-400)
    print(f"\nSuperposing {pdb_id2} onto {pdb_id1}...")
    sel1 = "//A/240-400/CA"
    sel2 = "//A/240-400/CA"
    result = coot.superpose_with_atom_selection(imol1, imol2, sel1, sel2, 0)
    
    if result >= 0:
        print(f"Success! Structures superposed.")
    else:
        print("Superposition failed!")
    
    return imol1, imol2

# Example: Compare human and mouse ABL1
# First find structures using Solr
query = "molecule_name:*ABL1*"
url = f"https://www.ebi.ac.uk/pdbe/search/pdb/select?q={query}&wt=json&rows=100&fl=pdb_id,organism_scientific_name"
result = coot.coot_get_url_as_string_py(url)

if isinstance(result, bytes):
   result = result.decode('utf-8')

data = json.loads(result)

# Find human and mouse structures
human_pdbs = []
mouse_pdbs = []
for doc in data['response']['docs']:
   org = doc.get('organism_scientific_name', ['Unknown'])[0]
   if 'Homo sapiens' in org:
      human_pdbs.append(doc['pdb_id'])
   elif 'Mus musculus' in org:
      mouse_pdbs.append(doc['pdb_id'])

print(f"Human ABL1 structures: {len(human_pdbs)}")
print(f"Mouse ABL1 structures: {len(mouse_pdbs)}")

# Compare first human and mouse structures
if human_pdbs and mouse_pdbs:
   compare_species_structures(human_pdbs[0], mouse_pdbs[0])

import json

def find_related_structures(protein_name, ligand_name=None):
    """Find structures containing specific protein-ligand combination"""
    if ligand_name:
        query = f'molecule_name:*{protein_name}* AND chem_comp_id:{ligand_name}'
    else:
        query = f'molecule_name:*{protein_name}*'
    
    url = f"https://www.ebi.ac.uk/pdbe/search/pdb/select?q={query}&wt=json&rows=50&fl=pdb_id,title,resolution,organism_scientific_name"
    
    result = coot.coot_get_url_as_string_py(url)
    
    if isinstance(result, bytes):
        result = result.decode('utf-8')
    
    data = json.loads(result)
    
    print(f"Found {data['response']['numFound']} structures")
    for doc in data['response']['docs']:
        org = doc.get('organism_scientific_name', ['N/A'])
        if isinstance(org, list):
            org = org[0] if org else 'N/A'
        
        print(f"  {doc['pdb_id']}: {doc.get('title', 'N/A')[:60]}")
        print(f"    Resolution: {doc.get('resolution', 'N/A')} Å")
        print(f"    Organism: {org}")

# Example usage
find_related_structures("ABL1", "STI")  # ABL1 with imatinib

import json

def safe_pdbe_query(url):
    """Safely query PDBe API with error handling"""
    try:
        result = coot.coot_get_url_as_string_py(url)
        if not result or result == "":
            print(f"Empty response from {url}")
            return None
        
        # Handle bytes response
        if isinstance(result, bytes):
            result = result.decode('utf-8')
        
        # Check for HTML error pages
        if result.startswith("<!DOCTYPE") or result.startswith("<html"):
            print(f"Received HTML error page instead of JSON")
            print(result[:200])
            return None
        
        data = json.loads(result)
        return data
    
    except json.JSONDecodeError as e:
        print(f"JSON parsing error: {e}")
        print(f"Response was: {result[:200]}...")
        return None
    
    except Exception as e:
        print(f"Error querying PDBe API: {e}")
        return None

# Example usage
data = safe_pdbe_query("https://www.ebi.ac.uk/pdbe/api/pdb/entry/summary/4wa9")
if data:
    print("Success!")

result = coot.coot_get_url_as_string_py(url)

if isinstance(result, bytes):
    # Binary data - might be gzipped
    if result.startswith(b'\x1f\x8b'):  # gzip magic bytes
        import gzip
        decompressed = gzip.decompress(result)
        content = decompressed.decode('utf-8')
    else:
        content = result.decode('utf-8')
else:
    # Already a string
    content = result

# Instead of json.loads(), parse line by line or handle errors
try:
    data = json.loads(result)
except json.JSONDecodeError:
    # Try alternative parsing or just display raw result
    print("Could not parse JSON, raw response:")
    print(result[:1000])

try:
    result = coot.coot_get_url_as_string_py(url)
except Exception as e:
    print(f"Error fetching URL: {e}")

import urllib.parse

query = "molecule_name:\"Protein kinase\" AND resolution:[0 TO 2.0]"
encoded = urllib.parse.quote(query)
url = f"https://www.ebi.ac.uk/pdbe/search/pdb/select?q={encoded}&wt=json"

import time

pdb_ids = ["4wa9", "2hyy", "1iep"]
for pdb_id in pdb_ids:
    data = safe_pdbe_query(f"https://www.ebi.ac.uk/pdbe/api/pdb/entry/summary/{pdb_id}")
    # Process data...
    time.sleep(0.5)  # Wait 500ms between requests

# Human kinases with resolution < 2Å from 2024
query = 'release_year:2024 AND organism_scientific_name:"Homo sapiens" AND molecule_name:*kinase* AND resolution:[* TO 2.0]'

# ABL1 from human OR mouse
query = 'molecule_name:*ABL1* AND (organism_scientific_name:"Homo sapiens" OR organism_scientific_name:"Mus musculus")'

import json

def structure_quality_report(imol):
    """Generate quality report using PDBe API data"""
    
    # Get PDB ID from molecule
    pdb_file = coot.molecule_name(imol)
    # Extract PDB ID from filename (assumes format like "pdb4wa9.ent" or "4wa9")
    import re
    match = re.search(r'(\d\w{3})', pdb_file.lower())
    if not match:
        print("Could not extract PDB ID from filename")
        return
    
    pdb_id = match.group(1)
    
    # Get structure info
    data = safe_pdbe_query(f"https://www.ebi.ac.uk/pdbe/api/pdb/entry/summary/{pdb_id}")
    if not data:
        return
    
    entry = data[pdb_id][0]
    
    print("=" * 60)
    print(f"STRUCTURE QUALITY REPORT: {pdb_id.upper()}")
    print("=" * 60)
    print(f"Title: {entry['title']}")
    print(f"Method: {entry['experimental_method']}")
    print(f"Released: {entry['release_date']}")
    print(f"Revised: {entry['revision_date']}")
    
    # Check for significant revisions
    if entry['revision_date'] != entry['release_date']:
        from datetime import datetime
        release = datetime.strptime(entry['release_date'], "%Y%m%d")
        revision = datetime.strptime(entry['revision_date'], "%Y%m%d")
        days = (revision - release).days
        print(f"\n⚠️  STRUCTURE REVISED {days} days after release")
        print("   Check PDBe for revision details")
    
    print("=" * 60)

# Usage: structure_quality_report(0)