@datagrok/proteomics 1.0.0 → 1.2.0

package/scripts/number_antibody.py

@@ -1,190 +0,0 @@
-#name: Antpack Antibody Numbering
-#description: Assigns antibody numbering (IMGT/Kabat/Chothia/AHo) using AntPack
-#language: python
-#environment: channels: [conda-forge, defaults], dependencies: [python=3.9, pip, pip: [antpack==0.3.8.6,pyside6,qt_material]]
-#meta.role: antibodyNumbering
-#input: dataframe df
-#input: column seqCol {semType: Macromolecule}
-#input: string scheme {choices: ["imgt", "kabat", "chothia", "aho"]} [Numbering scheme]
-#output: dataframe result
-
-import json
-import pandas as pd
-
-try:
-    from antpack import SingleChainAnnotator
-except ImportError:
-    raise ImportError(
-        'AntPack is not installed. Please install it with:\n'
-        '  pip install antpack==0.3.8.5\n'
-    )
-
-# Antpack github: https://gidthub.com/jlparkI/AntPack
-# While antpack is currently under non-commercial licence, the version used here (0.3.8.6 as specified in environment) is released unde GPL, allowing us to distribute this plugin with same GPL.
-
-# AntPack uses "martin" for modern Chothia numbering
-SCHEME_MAP = {
-    'imgt': 'imgt',
-    'kabat': 'kabat',
-    'chothia': 'martin',
-    'aho': 'aho',
-}
-
-# Region definitions per scheme, keyed by (scheme, chain_group).
-# chain_group is 'Heavy' or 'Light'.
-# Each region: (name, start_position, end_position).
-IMGT_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 26), ('CDR1', 27, 38), ('FR2', 39, 55),
-        ('CDR2', 56, 65), ('FR3', 66, 104), ('CDR3', 105, 117), ('FR4', 118, 128),
-    ],
-    'Light': [
-        ('FR1', 1, 26), ('CDR1', 27, 38), ('FR2', 39, 55),
-        ('CDR2', 56, 65), ('FR3', 66, 104), ('CDR3', 105, 117), ('FR4', 118, 127),
-    ],
-}
-
-KABAT_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 30), ('CDR1', 31, 35), ('FR2', 36, 49),
-        ('CDR2', 50, 65), ('FR3', 66, 94), ('CDR3', 95, 102), ('FR4', 103, 113),
-    ],
-    'Light': [
-        ('FR1', 1, 23), ('CDR1', 24, 34), ('FR2', 35, 49),
-        ('CDR2', 50, 56), ('FR3', 57, 88), ('CDR3', 89, 97), ('FR4', 98, 107),
-    ],
-}
-
-CHOTHIA_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 25), ('CDR1', 26, 32), ('FR2', 33, 51),
-        ('CDR2', 52, 56), ('FR3', 57, 94), ('CDR3', 95, 102), ('FR4', 103, 113),
-    ],
-    'Light': [
-        ('FR1', 1, 25), ('CDR1', 26, 32), ('FR2', 33, 49),
-        ('CDR2', 50, 52), ('FR3', 53, 90), ('CDR3', 91, 96), ('FR4', 97, 107),
-    ],
-}
-
-AHO_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 24), ('CDR1', 25, 40), ('FR2', 41, 55),
-        ('CDR2', 56, 78), ('FR3', 79, 108), ('CDR3', 109, 138), ('FR4', 139, 149),
-    ],
-    'Light': [
-        ('FR1', 1, 24), ('CDR1', 25, 40), ('FR2', 41, 55),
-        ('CDR2', 56, 78), ('FR3', 79, 108), ('CDR3', 109, 138), ('FR4', 139, 149),
-    ],
-}
-
-SCHEME_REGIONS = {
-    'imgt': IMGT_REGIONS,
-    'kabat': KABAT_REGIONS,
-    'chothia': CHOTHIA_REGIONS,
-    'aho': AHO_REGIONS,
-}
-
-
-def extract_sequence(raw_seq):
-    """Extract single-letter amino acid sequence from various formats."""
-    if not raw_seq or not isinstance(raw_seq, str):
-        return ''
-    s = raw_seq.strip()
-    s = s.replace('-', '').replace('.', '')
-    valid = set('ACDEFGHIKLMNPQRSTVWY')
-    return ''.join(c for c in s.upper() if c in valid)
-
-
-# Map user-facing scheme to AntPack scheme name
-antpack_scheme = SCHEME_MAP.get(scheme, 'imgt')
-
-# Create the annotator once for all sequences
-annotator = SingleChainAnnotator(chains=['H', 'K', 'L'], scheme=antpack_scheme)
-
-# Extract sequences from the input column
-col_name = seqCol
-sequences = []
-for i in range(df.shape[0]):
-    raw = df[col_name].iloc[i]
-    sequences.append(extract_sequence(str(raw) if raw is not None else ''))
-
-# Process each sequence
-position_names_list = []
-chain_types = []
-region_annotations_list = []
-numbering_results = []
-numbering_maps = []
-
-regions_def = SCHEME_REGIONS.get(scheme, IMGT_REGIONS)
-
-for seq in sequences:
-    if not seq:
-        position_names_list.append('')
-        chain_types.append('')
-        region_annotations_list.append('[]')
-        numbering_results.append('')
-        numbering_maps.append('')
-        continue
-
-    numbering, pct_identity, chain_type_code, err = annotator.analyze_seq(seq)
-
-    if err or pct_identity < 0.5:
-        position_names_list.append('')
-        chain_types.append('')
-        region_annotations_list.append('[]')
-        numbering_results.append('')
-        numbering_maps.append('')
-        continue
-
-    # Determine chain type
-    if chain_type_code == 'H':
-        chain_type = 'Heavy'
-    elif chain_type_code in ('K', 'L'):
-        chain_type = 'Light'
-    else:
-        chain_type = 'Unknown'
-
-    # Build position names, numbering detail, and position-to-character-index map.
-    # numbering is a list with the same length as the input sequence;
-    # each element is a position code string or '-' for residues outside the variable region.
-    pos_names = []
-    numbering_detail = []
-    pos_to_char_idx = {}
-    for i, pos_code in enumerate(numbering):
-        if pos_code != '-':
-            pos_names.append(pos_code)
-            numbering_detail.append({'position': pos_code, 'aa': seq[i]})
-            pos_to_char_idx[pos_code] = i
-
-    position_names_list.append(', '.join(pos_names))
-    chain_types.append(chain_type)
-
-    # Build region annotations
-    chain_key = chain_type if chain_type in regions_def else 'Heavy'
-    region_defs = regions_def.get(chain_key, [])
-    annotations = []
-    for region_name, start, end in region_defs:
-        annotations.append({
-            'id': f'{scheme}-{chain_type}-{region_name}'.lower(),
-            'name': region_name,
-            'description': f'{region_name} ({scheme.upper()} {start}-{end})',
-            'start': str(start),
-            'end': str(end),
-            'visualType': 'region',
-            'category': 'structure',
-            'sourceScheme': scheme.upper(),
-            'autoGenerated': True,
-        })
-
-    region_annotations_list.append(json.dumps(annotations))
-    numbering_results.append(json.dumps(numbering_detail))
-    numbering_maps.append(json.dumps(pos_to_char_idx))
-
-# Build result DataFrame
-result = pd.DataFrame({
-    'position_names': position_names_list,
-    'chain_type': chain_types,
-    'annotations_json': region_annotations_list,
-    'numbering_detail': numbering_results,
-    'numbering_map': numbering_maps,
-})

package/scripts/number_antibody_abnumber.py

@@ -1,177 +0,0 @@
-#name: Number Antibody Sequences (AbNumber)
-#description: [Legacy] Assigns antibody numbering (IMGT/Kabat/Chothia/AHo) using AbNumber
-#language: python
-#environment: channels: [conda-forge, bioconda, defaults], dependencies: [python=3.9, pip, hmmer=3.3.2, pip: [abnumber]]
-#input: dataframe df
-#input: column seqCol {semType: Macromolecule}
-#input: string scheme {choices: ["imgt", "kabat", "chothia", "aho"]} [Numbering scheme]
-#output: dataframe result
-
-import json
-import pandas as pd
-
-try:
-    from abnumber import Chain
-except ImportError:
-    raise ImportError(
-        'AbNumber is not installed. Please install it with:\n'
-        '  conda install -c bioconda abnumber\n'
-        'or:\n'
-        '  pip install abnumber\n'
-        'HMMER must also be available on PATH.'
-    )
-
-# Region definitions per scheme, keyed by (scheme, chain_group).
-# chain_group is 'Heavy' or 'Light'.
-# Each region: (name, start_position, end_position).
-IMGT_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 26), ('CDR1', 27, 38), ('FR2', 39, 55),
-        ('CDR2', 56, 65), ('FR3', 66, 104), ('CDR3', 105, 117), ('FR4', 118, 128),
-    ],
-    'Light': [
-        ('FR1', 1, 26), ('CDR1', 27, 38), ('FR2', 39, 55),
-        ('CDR2', 56, 65), ('FR3', 66, 104), ('CDR3', 105, 117), ('FR4', 118, 127),
-    ],
-}
-
-KABAT_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 30), ('CDR1', 31, 35), ('FR2', 36, 49),
-        ('CDR2', 50, 65), ('FR3', 66, 94), ('CDR3', 95, 102), ('FR4', 103, 113),
-    ],
-    'Light': [
-        ('FR1', 1, 23), ('CDR1', 24, 34), ('FR2', 35, 49),
-        ('CDR2', 50, 56), ('FR3', 57, 88), ('CDR3', 89, 97), ('FR4', 98, 107),
-    ],
-}
-
-CHOTHIA_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 25), ('CDR1', 26, 32), ('FR2', 33, 51),
-        ('CDR2', 52, 56), ('FR3', 57, 94), ('CDR3', 95, 102), ('FR4', 103, 113),
-    ],
-    'Light': [
-        ('FR1', 1, 25), ('CDR1', 26, 32), ('FR2', 33, 49),
-        ('CDR2', 50, 52), ('FR3', 53, 90), ('CDR3', 91, 96), ('FR4', 97, 107),
-    ],
-}
-
-AHO_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 24), ('CDR1', 25, 40), ('FR2', 41, 55),
-        ('CDR2', 56, 78), ('FR3', 79, 108), ('CDR3', 109, 138), ('FR4', 139, 149),
-    ],
-    'Light': [
-        ('FR1', 1, 24), ('CDR1', 25, 40), ('FR2', 41, 55),
-        ('CDR2', 56, 78), ('FR3', 79, 108), ('CDR3', 109, 138), ('FR4', 139, 149),
-    ],
-}
-
-SCHEME_REGIONS = {
-    'imgt': IMGT_REGIONS,
-    'kabat': KABAT_REGIONS,
-    'chothia': CHOTHIA_REGIONS,
-    'aho': AHO_REGIONS,
-}
-
-
-def extract_sequence(raw_seq):
-    """Extract single-letter amino acid sequence from various formats."""
-    if not raw_seq or not isinstance(raw_seq, str):
-        return ''
-    s = raw_seq.strip()
-    s = s.replace('-', '').replace('.', '')
-    valid = set('ACDEFGHIKLMNPQRSTVWY')
-    return ''.join(c for c in s.upper() if c in valid)
-
-
-def number_single_sequence(seq, scheme_name):
-    """Number a single sequence using AbNumber's Chain object.
-
-    Returns (chain, error_msg) where chain is an abnumber.Chain or None.
-    """
-    if not seq:
-        return None, 'empty sequence'
-    try:
-        chain = Chain(seq, scheme=scheme_name)
-        return chain, None
-    except Exception as e:
-        return None, str(e)
-
-
-# Extract sequences from the input column
-col_name = seqCol
-sequences = []
-for i in range(df.shape[0]):
-    raw = df[col_name].iloc[i]
-    sequences.append(extract_sequence(str(raw) if raw is not None else ''))
-
-# Process each sequence
-position_names_list = []
-chain_types = []
-region_annotations_list = []
-numbering_results = []
-
-regions_def = SCHEME_REGIONS.get(scheme, IMGT_REGIONS)
-
-for seq in sequences:
-    chain, err = number_single_sequence(seq, scheme)
-
-    if chain is None:
-        position_names_list.append('')
-        chain_types.append('')
-        region_annotations_list.append('[]')
-        numbering_results.append('')
-        continue
-
-    # Determine chain type
-    ct = chain.chain_type  # 'H', 'K', or 'L'
-    if ct == 'H':
-        chain_type = 'Heavy'
-    elif ct in ('K', 'L'):
-        chain_type = 'Light'
-    else:
-        chain_type = 'Unknown'
-
-    # Build position names from numbering
-    pos_names = []
-    numbering_detail = []
-    for position, aa in chain:
-        pos_name = str(position.number)
-        if position.letter and position.letter.strip():
-            pos_name += position.letter.strip()
-        pos_names.append(pos_name)
-        numbering_detail.append({'position': pos_name, 'aa': aa})
-
-    position_names_list.append(', '.join(pos_names))
-
-    chain_types.append(chain_type)
-
-    # Build region annotations
-    chain_key = chain_type if chain_type in regions_def else 'Heavy'
-    region_defs = regions_def.get(chain_key, [])
-    annotations = []
-    for region_name, start, end in region_defs:
-        annotations.append({
-            'id': f'{scheme}-{chain_type}-{region_name}'.lower(),
-            'name': region_name,
-            'description': f'{region_name} ({scheme.upper()} {start}-{end})',
-            'start': str(start),
-            'end': str(end),
-            'visualType': 'region',
-            'category': 'structure',
-            'sourceScheme': scheme.upper(),
-            'autoGenerated': True,
-        })
-
-    region_annotations_list.append(json.dumps(annotations))
-    numbering_results.append(json.dumps(numbering_detail))
-
-# Build result DataFrame
-result = pd.DataFrame({
-    'position_names': position_names_list,
-    'chain_type': chain_types,
-    'annotations_json': region_annotations_list,
-    'numbering_detail': numbering_results,
-})

package/scripts/number_antibody_anarci.py

@@ -1,200 +0,0 @@
-#name: Number Antibody Sequences (ANARCI)
-#description: [Legacy] Assigns antibody numbering (IMGT/Kabat/Chothia/AHo) using ANARCI directly
-#language: python
-#environment: channels: [conda-forge, bioconda, defaults], dependencies: [python=3.9, pip, hmmer=3.3.2, anarci]
-#input: dataframe df
-#input: column seqCol {semType: Macromolecule}
-#input: string scheme {choices: ["imgt", "kabat", "chothia", "aho"]} [Numbering scheme]
-#output: dataframe result
-
-import json
-import pandas as pd
-
-try:
-    from anarci import anarci, number
-except ImportError:
-    raise ImportError(
-        'ANARCI is not installed. Please install it with:\n'
-        '  conda install -c bioconda anarci hmmer\n'
-        'or:\n'
-        '  pip install anarci\n'
-        'HMMER must also be available on PATH.'
-    )
-
-# IMGT region boundaries (used for all schemes after numbering)
-IMGT_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 26), ('CDR1', 27, 38), ('FR2', 39, 55),
-        ('CDR2', 56, 65), ('FR3', 66, 104), ('CDR3', 105, 117), ('FR4', 118, 128),
-    ],
-    'Light': [
-        ('FR1', 1, 26), ('CDR1', 27, 38), ('FR2', 39, 55),
-        ('CDR2', 56, 65), ('FR3', 66, 104), ('CDR3', 105, 117), ('FR4', 118, 127),
-    ],
-}
-
-KABAT_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 30), ('CDR1', 31, 35), ('FR2', 36, 49),
-        ('CDR2', 50, 65), ('FR3', 66, 94), ('CDR3', 95, 102), ('FR4', 103, 113),
-    ],
-    'Light': [
-        ('FR1', 1, 23), ('CDR1', 24, 34), ('FR2', 35, 49),
-        ('CDR2', 50, 56), ('FR3', 57, 88), ('CDR3', 89, 97), ('FR4', 98, 107),
-    ],
-}
-
-CHOTHIA_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 25), ('CDR1', 26, 32), ('FR2', 33, 51),
-        ('CDR2', 52, 56), ('FR3', 57, 94), ('CDR3', 95, 102), ('FR4', 103, 113),
-    ],
-    'Light': [
-        ('FR1', 1, 25), ('CDR1', 26, 32), ('FR2', 33, 49),
-        ('CDR2', 50, 52), ('FR3', 53, 90), ('CDR3', 91, 96), ('FR4', 97, 107),
-    ],
-}
-
-AHO_REGIONS = {
-    'Heavy': [
-        ('FR1', 1, 24), ('CDR1', 25, 40), ('FR2', 41, 55),
-        ('CDR2', 56, 78), ('FR3', 79, 108), ('CDR3', 109, 138), ('FR4', 139, 149),
-    ],
-    'Light': [
-        ('FR1', 1, 24), ('CDR1', 25, 40), ('FR2', 41, 55),
-        ('CDR2', 56, 78), ('FR3', 79, 108), ('CDR3', 109, 138), ('FR4', 139, 149),
-    ],
-}
-
-SCHEME_REGIONS = {
-    'imgt': IMGT_REGIONS,
-    'kabat': KABAT_REGIONS,
-    'chothia': CHOTHIA_REGIONS,
-    'aho': AHO_REGIONS,
-}
-
-
-def extract_sequence(raw_seq):
-    """Extract single-letter amino acid sequence from various formats."""
-    if not raw_seq or not isinstance(raw_seq, str):
-        return ''
-    s = raw_seq.strip()
-    # If it looks like FASTA single-letter (only AA chars and gaps)
-    s = s.replace('-', '').replace('.', '')
-    # Filter to valid amino acid characters
-    valid = set('ACDEFGHIKLMNPQRSTVWY')
-    return ''.join(c for c in s.upper() if c in valid)
-
-
-def number_sequences(sequences, scheme_name):
-    """Number a list of sequences using ANARCI."""
-    # Prepare input for ANARCI
-    seq_list = [(f'seq_{i}', seq) for i, seq in enumerate(sequences) if seq]
-
-    if not seq_list:
-        return [], [], []
-
-    # Run ANARCI
-    numbered_seqs, alignment_details, hit_tables = anarci(
-        seq_list,
-        scheme=scheme_name,
-        output=False,
-        allow=set(['H', 'K', 'L']),
-    )
-
-    return numbered_seqs, alignment_details, hit_tables
-
-
-# Extract sequences from the input column
-col_name = seqCol
-sequences = []
-for i in range(df.shape[0]):
-    raw = df[col_name].iloc[i]
-    sequences.append(extract_sequence(str(raw) if raw is not None else ''))
-
-# Number all sequences
-numbered_seqs, alignment_details, hit_tables = number_sequences(sequences, scheme)
-
-# Process results
-position_names_list = []
-chain_types = []
-region_annotations_list = []
-numbering_results = []
-
-regions_def = SCHEME_REGIONS.get(scheme, IMGT_REGIONS)
-
-for idx in range(len(sequences)):
-    if numbered_seqs[idx] is None or len(numbered_seqs[idx]) == 0:
-        position_names_list.append('')
-        chain_types.append('')
-        region_annotations_list.append('[]')
-        numbering_results.append('')
-        continue
-
-    # Get the best hit (first domain)
-    numbering = numbered_seqs[idx][0]
-    chain_info = alignment_details[idx][0]
-
-    if numbering is None or chain_info is None:
-        position_names_list.append('')
-        chain_types.append('')
-        region_annotations_list.append('[]')
-        numbering_results.append('')
-        continue
-
-    # Extract chain type
-    chain_type_code = chain_info['chain_type']
-    if chain_type_code == 'H':
-        chain_type = 'Heavy'
-    elif chain_type_code in ('K', 'L'):
-        chain_type = 'Light'
-    else:
-        chain_type = 'Unknown'
-
-    # Build position names from numbering
-    pos_names = []
-    for (pos_num, insertion_code), aa in numbering:
-        if aa != '-':
-            pos_name = str(pos_num)
-            if insertion_code != ' ':
-                pos_name += insertion_code
-            pos_names.append(pos_name)
-
-    position_names_list.append(', '.join(pos_names))
-    chain_types.append(chain_type)
-
-    # Build region annotations
-    chain_key = chain_type if chain_type in regions_def else 'Heavy'
-    region_defs = regions_def.get(chain_key, [])
-    annotations = []
-    for region_name, start, end in region_defs:
-        region_type = 'CDR' if 'CDR' in region_name else 'FR'
-        annotations.append({
-            'id': f'{scheme}-{chain_type}-{region_name}'.lower(),
-            'name': region_name,
-            'description': f'{region_name} ({scheme.upper()} {start}-{end})',
-            'start': str(start),
-            'end': str(end),
-            'visualType': 'region',
-            'category': 'structure',
-            'sourceScheme': scheme.upper(),
-            'autoGenerated': True,
-        })
-
-    region_annotations_list.append(json.dumps(annotations))
-
-    # Full numbering as JSON for debugging/reference
-    numbering_json = json.dumps([
-        {'position': f'{pos_num}{insertion_code.strip()}', 'aa': aa}
-        for (pos_num, insertion_code), aa in numbering
-        if aa != '-'
-    ])
-    numbering_results.append(numbering_json)
-
-# Build result DataFrame
-result = pd.DataFrame({
-    'position_names': position_names_list,
-    'chain_type': chain_types,
-    'annotations_json': region_annotations_list,
-    'numbering_detail': numbering_results,
-})