geney 1.3.73__py2.py3-none-any.whl → 1.3.75__py2.py3-none-any.whl

geney/Gene.py

@@ -90,7 +90,9 @@ class Gene:
         # Find gene data files in the configured organism MRNA path
         gene_files = list((config[organism]['MRNA_PATH'] / 'protein_coding').glob(f'*_{gene_name}.pkl'))
         if not gene_files:
-            raise FileNotFoundError(f"No files available for gene '{gene_name}'.")
+            print(f"No files available for gene '{gene_name}'.")
+            return None
+            # raise FileNotFoundError(f"No files available for gene '{gene_name}'.")
 
         # Load gene data from the first matching file
         data = unload_pickle(gene_files[0])
@@ -148,7 +150,8 @@ class Gene:
             return None #Transcript()
 
         if tid not in self.transcripts:
-            raise AttributeError(f"Transcript '{tid}' not found in gene '{self.gene_name}'.")
+            return None
+            # raise AttributeError(f"Transcript '{tid}' not found in gene '{self.gene_name}'.")
 
         return Transcript(self.transcripts[tid], organism=self.organism)
 

geney/oncosplice.py

@@ -1,6 +1,7 @@
 from Bio import pairwise2
 import re
 import hashlib
+from datetime import datetime
 from tqdm import tqdm
 import pandas as pd
 import numpy as np
@@ -282,6 +283,16 @@ def summarize_missplicing_event(pes, pir, es, ne, ir):
     else:
         return '-'
 
+def missense_effect(r, v):
+    if len(r.protein) != len(v.protein):
+        return '', ''
+
+    for i, (x, y) in enumerate(zip(r.protein, v.protein)):
+        if x != y:
+            aa_change = f'{x}>{y}'
+            codon_change = f'{r.orf[i*3:i*3+3]}>{v.orf[i*3:i*3+3]}'
+            return aa_change, codon_change
+    return '', ''
 
 # Annotating
 def OncospliceAnnotator(reference_transcript, variant_transcript, mut, ref_attributes=[], var_attributes=[]):
@@ -301,12 +312,15 @@ def OncospliceAnnotator(reference_transcript, variant_transcript, mut, ref_attri
     report['affected_intron'] = affected_intron
     report['mutation_distance_from_5'] = distance_from_5
     report['mutation_distance_from_3'] = distance_from_3
+    aa_c, c_c = missense_effect(reference_transcript, variant_transcript)
+    report['missense_effect'] = aa_c
+    report['codon_change'] = c_c
     return report
 
 
-def oncosplice(mut_id, splicing_threshold=0.5, protein_coding=True, cons_required=False, primary_transcript=False,
+def oncosplice(mut_id, splicing_threshold=0.5, protein_coding=True, primary_transcript=False,
                window_length=13, organism='hg38', splicing_engine=None, splicing_db=None, verbose=False,
-               tis_engine=None, tis_db=None):
+               tis_engine=None, target_transcripts=None):
 
     gene = Gene.from_file(mut_id.split(':')[0], organism=organism)
     reference_gene_proteins = {
@@ -315,10 +329,13 @@ def oncosplice(mut_id, splicing_threshold=0.5, protein_coding=True, cons_require
 
     mutations = [MutSeqMat.from_mutid(m) for m in mut_id.split('|')]
     if gene.rev:
-        mutations = [m.reverse_complement(inplace=True) for m in mutations[::-1]]
+        mutations = [m.reverse_complement() for m in mutations[::-1]]
 
     results = []
     for reference_transcript in tqdm(gene, desc=f'Processing {mut_id}...'):
+        if target_transcripts is not None and reference_transcript.transcript_id not in target_transcripts:
+            continue
+
         # if (cons_required and not reference_transcript.cons_available) or (
         #         protein_coding and not reference_transcript.transcript_biotype == 'protein_coding'):
         if protein_coding and not reference_transcript.transcript_biotype == 'protein_coding':
@@ -338,9 +355,6 @@ def oncosplice(mut_id, splicing_threshold=0.5, protein_coding=True, cons_require
             mutated_transcript.mutate(mutation, inplace=True)
 
         reference_transcript.generate_mature_mrna().generate_protein()
-        # if ('*' in reference_transcript.protein[:-1]):
-        #     print(f"> Errors in reference transcript {reference_transcript.transcript_id}")
-        #     continue
 
         if len(reference_transcript.protein) < window_length:
             print(f"> Window length issue {reference_transcript.transcript_id}")
@@ -415,15 +429,173 @@ def oncosplice(mut_id, splicing_threshold=0.5, protein_coding=True, cons_require
             results.append(report)
 
     if len(results) == 0:
-        print("Nothing...")
-        return None
+        # print("Nothing...")
+        return pd.DataFrame()
 
     return pd.DataFrame(results)[
         ['mut_id', 'transcript_id', 'isoform_id', 'primary_transcript', 'missplicing', 'full_missplicing',
          'exon_changes', 'splicing_codes', 'affected_exon', 'affected_intron', 'mutation_distance_from_5',
-         'mutation_distance_from_3', 'reference_resemblance', 'oncosplice_score', 'percentile',
+         'mutation_distance_from_3', 'missense_effect', 'codon_change', 'reference_resemblance', 'oncosplice_score', 'percentile',
          'isoform_prevalence', 'reference_protein', 'variant_protein', 'splicing_engine']]
 
+
+def process_splicing_path(new_boundaries, reference_transcript, mutated_transcript,
+                          current_mutations, missplicing, mut_id, transcript_id,
+                          splicing_engine, window_length, start_time):
+    """
+    Processes a single alternative splicing path and returns an annotation report.
+    """
+    # Update acceptors and donors
+    mutated_transcript.acceptors = new_boundaries['acceptors']
+    mutated_transcript.donors = new_boundaries['donors']
+    mutated_transcript.generate_mature_mrna().generate_protein()
+
+    # Align reference and mutated proteins
+    alignment = get_logical_alignment(reference_transcript.protein, mutated_transcript.protein)
+    deleted, inserted = find_indels_with_mismatches_as_deletions(alignment.seqA, alignment.seqB)
+    modified_positions = find_modified_positions(len(reference_transcript.protein), deleted, inserted)
+
+    # Compute conservation impact
+    temp_cons = np.convolve(reference_transcript.cons_vector * modified_positions,
+                            np.ones(window_length)) / window_length
+    affected_cons_scores = max(temp_cons)
+
+    # Compute percentile of conservation change
+    sorted_cons = sorted(reference_transcript.cons_vector)
+    percentile = (
+            sorted_cons.index(next(x for x in sorted_cons if x >= affected_cons_scores)) / len(sorted_cons)
+    )
+
+    # Generate annotation report
+    report = OncospliceAnnotator(reference_transcript, mutated_transcript, current_mutations[0])
+    report.update({
+        'mut_id': mut_id,
+        'transcript_id': transcript_id,
+        'splicing_engine': splicing_engine if splicing_engine is not None else 'None',
+        'oncosplice_score': affected_cons_scores,
+        'percentile': percentile,
+        'isoform_id': short_hash_of_list(mutated_transcript.exons),
+        'isoform_prevalence': new_boundaries['path_weight'],
+        'full_missplicing': missplicing.aberrant_splicing,
+        'missplicing': missplicing.max_delta,
+        'execution_time': start_time,
+        'status': 'Success'
+    })
+    return report
+
+
+def oncosplice_df(row, splicing_threshold=0.5, window_length=13,
+                  organism='hg38', splicing_engine='spliceai'):
+    """
+    Process a given mutation-transcript pair to analyze alternative splicing events
+    and their oncogenic potential.
+
+    Ensures that every `mut_id` and `transcript_id` is included in the output,
+    even if processing fails, with an appropriate failure status.
+
+    Parameters:
+    - row (pd.Series): A row from a DataFrame containing `mut_id` and `transcript_id`
+    - splicing_threshold (float): The threshold for splicing disruption detection
+    - window_length (int): The window size for conservation analysis
+    - organism (str): Genome assembly version
+    - splicing_engine (str or None): The splicing prediction engine to use
+
+    Returns:
+    - pd.DataFrame: Processed results of the oncosplice analysis
+    - Run "pd.concat(applied_function_output.to_list(), ignore_index=True)" to build result dataframe
+    """
+
+    start_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")  # Log function start time
+    mut_id, transcript_id = row.mut_id, row.transcript_id
+
+    # Default response template (to ensure all IDs are included)
+    base_result = {
+        'mut_id': mut_id,
+        'transcript_id': transcript_id,
+        'status': 'Success',  # Will be updated if an issue occurs
+        'execution_time': start_time
+    }
+
+    # Load gene and transcript
+    gene = Gene.from_file(mut_id.split(':')[0], organism=organism)
+    if gene is None:
+        base_result['status'] = 'Gene not found'
+        return pd.DataFrame([base_result])
+
+    mutations = [MutSeqMat.from_mutid(m) for m in mut_id.split('|')]
+    reference_transcript = gene.transcript(transcript_id)
+
+    if reference_transcript is None:
+        base_result['status'] = 'Transcript not found'
+        return pd.DataFrame([base_result])
+
+    # Generate the reference transcript
+    reference_transcript = (
+        reference_transcript.generate_pre_mrna()
+        .generate_mature_mrna()
+        .generate_protein()
+    )
+
+    # Skip non-protein-coding transcripts
+    if reference_transcript.transcript_biotype != 'protein_coding':
+        base_result['status'] = 'Non-protein-coding transcript'
+        return pd.DataFrame([base_result])
+
+    # Filter mutations relevant to this transcript
+    current_mutations = [m for m in mutations if m in reference_transcript]
+    if not current_mutations:
+        base_result['status'] = 'No relevant mutations'
+        return pd.DataFrame([base_result])
+
+    # Define mutation center
+    center = np.mean([m.indices[0] for m in current_mutations]) // 1
+
+    # Clone and apply mutations
+    mutated_transcript = reference_transcript.clone()
+    for mutation in current_mutations:
+        mutated_transcript.mutate(mutation, inplace=True)
+
+    # Adjust window length if necessary
+    window_length = min(window_length, len(reference_transcript.protein))
+
+    # Transform conservation vector
+    reference_transcript.cons_vector = transform_conservation_vector(
+        reference_transcript.cons_vector, window=window_length
+    )
+
+    # Identify missplicing events
+    if splicing_engine is None:
+        missplicing = Missplicing()
+    else:
+        missplicing = find_transcript_missplicing_seqs(
+            reference_transcript.pre_mrna.get_context(center, context=7500, padding='N'),
+            mutated_transcript.pre_mrna.get_context(center, context=7500, padding='N'),
+            reference_transcript.donors, reference_transcript.acceptors,
+            threshold=splicing_threshold, engine=splicing_engine
+        )
+
+    # Generate alternative splicing paths
+    alternative_splicing_paths = develop_aberrant_splicing(reference_transcript, missplicing)
+    results = [
+        process_splicing_path(new_boundaries, reference_transcript, mutated_transcript,
+                              current_mutations, missplicing, mut_id, transcript_id,
+                              splicing_engine, window_length, start_time)
+        for new_boundaries in alternative_splicing_paths
+    ]
+
+    # Return results as DataFrame
+    if not results:
+        base_result['status'] = 'No alternative splicing detected'
+        return pd.DataFrame([base_result])
+
+    return pd.DataFrame(results)[
+        ['mut_id', 'transcript_id', 'isoform_id', 'primary_transcript', 'missplicing', 'full_missplicing',
+         'exon_changes', 'splicing_codes', 'affected_exon', 'affected_intron', 'mutation_distance_from_5',
+         'mutation_distance_from_3', 'missense_effect', 'codon_change', 'oncosplice_score', 'percentile',
+         'isoform_prevalence', 'reference_protein', 'variant_protein', 'splicing_engine', 'execution_time', 'status']
+    ]
+
+
 #
 # import asyncio
 # async def oncosplice_prototype(mut_id, splicing_threshold=0.5, protein_coding=True, primary_transcript=False,

geney/spliceai_utils.py

@@ -1,9 +1,6 @@
-
 #### SpliceAI Modules
 from keras.models import load_model
-# from pkg_resources import resource_filename
 from importlib import resources
-# from spliceai.utils import one_hot_encode
 import numpy as np
 import tensorflow as tf
 import sys
@@ -17,18 +14,25 @@ else:
 # tf.config.threading.set_intra_op_parallelism_threads(1)
 # tf.config.threading.set_inter_op_parallelism_threads(1)
 
-if sys.platform == 'darwin':
-    sai_paths = ('models/spliceai{}.h5'.format(x) for x in range(1, 6))
-    # sai_models = [load_model(resource_filename('spliceai', x)) for x in sai_paths]
-    sai_models = [load_model(resources.files('spliceai').joinpath(f)) for f in sai_paths]
-else:
-    sai_paths = ['/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai1.h5',
-                 '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai2.h5',
-                 '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai3.h5',
-                 '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai4.h5',
-                 '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai5.h5']
-
-    sai_models = [load_model(f) for f in sai_paths]
+# List the model filenames relative to the spliceai package.
+model_filenames = [f"models/spliceai{i}.h5" for i in range(1, 6)]
+
+# Load each model using the package resources.
+sai_models = [load_model(resources.files("spliceai").joinpath(filename))
+              for filename in model_filenames]
+
+# if sys.platform == 'darwin':
+#     sai_paths = ('models/spliceai{}.h5'.format(x) for x in range(1, 6))
+#     # sai_models = [load_model(resource_filename('spliceai', x)) for x in sai_paths]
+#     sai_models = [load_model(resources.files('spliceai').joinpath(f)) for f in sai_paths]
+# else:
+#     sai_paths = ['/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai1.h5',
+#                  '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai2.h5',
+#                  '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai3.h5',
+#                  '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai4.h5',
+#                  '/tamir2/nicolaslynn/home/miniconda3/lib/python3.10/site-packages/spliceai/models/spliceai5.h5']
+
+    # sai_models = [load_model(f) for f in sai_paths]
 
 
 def one_hot_encode(seq):

geney/splicing_utils.py

@@ -221,29 +221,34 @@ def find_transcript_splicing(transcript, engine: str = 'spliceai') -> Tuple[Dict
     return donor_probs, acceptor_probs
 
 
-def find_transcript_missplicing(mut_id, transcript=None, threshold=0.5, engine='spliceai', organism='hg38', db=None, force_recompute=False):
+def missplicing_df(mut_id, **kwargs):
+    return find_transcript_missplicing(mut_id, **kwargs).max_delta
+
+
+def find_transcript_missplicing(mut_id, transcript=None, threshold=0.5, engine='spliceai', organism='hg38'):
     gene = Gene.from_file(mut_id.split(':')[0], organism=organism)
     reference_transcript = gene.transcript(transcript) if transcript is not None else gene.transcript()
-
-    if db is not None:
-        cached_data = db.get_mutation_data(engine, gene.gene_name, mut_id, reference_transcript.transcript_id)
-        if cached_data and not force_recompute:
-            return Missplicing(cached_data)
+    if reference_transcript is None:
+        return Missplicing()
+    #
+    # if db is not None:
+    #     cached_data = db.get_mutation_data(engine, gene.gene_name, mut_id, reference_transcript.transcript_id)
+    #     if cached_data and not force_recompute:
+    #         return Missplicing(cached_data)
 
     variant_transcript = reference_transcript.clone()
     mutations = [MutSeqMat.from_mutid(m) for m in mut_id.split('|')]
     mutations = [m for m in mutations if m in reference_transcript]
     if len(mutations) == 0:
-        return Missplicing({'missed_acceptors': {}, 'missed_donors': {}, 'discovered_acceptors': {}, 'discovered_donors': {}})
+        return Missplicing()
 
     center = int(np.mean([m.indices[0] for m in mutations]))
     for mutation in mutations:
         variant_transcript.mutate(mutation, inplace=True)
 
     missplicing = find_transcript_missplicing_seqs(reference_transcript.pre_mrna.get_context(center, 7500, padding='N'), variant_transcript.pre_mrna.get_context(center, 7500, padding='N'), reference_transcript.donors, reference_transcript.acceptors, threshold=threshold, engine=engine)
-    if db is not None:
-        db.store_mutation_data(engine, gene.gene_name, mut_id, reference_transcript.transcript_id, missplicing.missplicing)
-
+    # if db is not None:
+    #     db.store_mutation_data(engine, gene.gene_name, mut_id, reference_transcript.transcript_id, missplicing.missplicing)
     return missplicing
 
     # from functools import reduce
@@ -650,7 +655,7 @@ def process_pairwise_epistasis_explicit(mid, engine='spliceai'):
 
 
 class Missplicing:
-    def __init__(self, splicing_dict=None, threshold=0.5):
+    def __init__(self, splicing_dict={'missed_acceptors': {}, 'missed_donors': {}, 'discovered_acceptors': {}, 'discovered_donors': {}}, threshold=0.5):
         """
         Initialize a Missplicing object.
 

geney/tcga_utils.py

@@ -1,4 +1,3 @@
-
 import pandas as pd
 import random
 from pathlib import Path

{geney-1.3.73.dist-info → geney-1.3.75.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: geney
-Version: 1.3.73
+Version: 1.3.75
 Summary: A Python package for gene expression modeling.
 Home-page: https://github.com/nicolaslynn/geney
 Author: Nicolas Lynn

{geney-1.3.73.dist-info → geney-1.3.75.dist-info}/RECORD

@@ -1,5 +1,5 @@
 geney/Fasta_segment.py,sha256=99HxNGNh_MfdVW6hhtlb1vOn7eSmT7oFoEfHDFMxG8w,11275
-geney/Gene.py,sha256=oAdwuguD1qWMTrS158ApW4s3MRcb1ZJlinzVeCZcYwE,6966
+geney/Gene.py,sha256=2eznhavDoO4rktisy3U0uz480OsktppEwH5pxC2qoog,7083
 geney/SeqMats.py,sha256=9-eJnfU2w3LGc0XvVvFEO_QrBneTkC6xkZKDfTcEw5o,19282
 geney/Transcript.py,sha256=CpfxYkuCwFILozrtLuiWnlr1mRnMKn4o84HVJislgYs,14499
 geney/__init__.py,sha256=eBdDl42N6UhcYeZDjOnv199Z88fI5_8Y6xW8447OKXM,755
@@ -11,21 +11,21 @@ geney/graphic_utils.py,sha256=oMsBpB9YeEn96gGpKh4MmtagJffWZbk-xPrIwHvkFhA,11016
 geney/gtex_utils.py,sha256=asL2lHyU5KsbWpV096vkf1Ka7hSo_RRfZqw7p5nERmE,1919
 geney/immune_utils.py,sha256=ZRni5ttrhpYBnmNr0d0ZatIbNPYs4nmQuoUO00SpsS4,5271
 geney/mutation_utils.py,sha256=C_kv2MB_L8LlhX3W2ooXjJ3uDoJ8zX1WeDtZKoBZJkI,1547
-geney/oncosplice.py,sha256=YZvAnbe8gj9fPvs2DldeQpqhhe_QR9xBLe_0tcm9tdg,24793
+geney/oncosplice.py,sha256=OptwyMhTaL5ptmdPxyuC2Iwm4feQf0n1ZjN0o9hwZ1w,32051
 geney/pangolin_utils.py,sha256=9jdBXlOcRaUdfi-UpUxHA0AkTMZkUF-Lt7HVZ1nEm3s,2973
 geney/power_utils.py,sha256=MehZFUdkJ2EFUot709yPEDxSkXmH5XevMebX2HD768A,7330
 geney/seqmat_utils.py,sha256=wzb3PX5it5bpIFQvcxyzlxfhoJTbHHbsjg0rzh05iVs,19753
-geney/spliceai_utils.py,sha256=tVY0T6F6l3fNoaktpn7Kq0oH5ZM0ThFYt9nPi_lfakw,3077
-geney/splicing_utils.py,sha256=mAiZXBzb0IzZ_O0CIhxR36q8H9xqPXJOUk2fU1UTqc8,47313
+geney/spliceai_utils.py,sha256=V5DccGjm2GZQINSVxJYPoN8iYQzU4gE9tINy6gmHjOM,3304
+geney/splicing_utils.py,sha256=UkG2YphjLNUYsv3o3RGUTW1ScHbEMOLL2M_7WbgDVME,47466
 geney/survival_utils.py,sha256=KnAzEviMuXh6SnVXId9PgsFLSbgkduTvYoIthxN7FPA,6886
-geney/tcga_utils.py,sha256=D_BNHm-D_K408dlcJm3hzH2c6QNFjQsKvUcOPiQRk7g,17612
+geney/tcga_utils.py,sha256=uJhVnTbTysj0XrEw_YeDKRSLexsqgBLYQdhl7_hnr64,17611
 geney/tis_utils.py,sha256=la0CZroaKe5RgAyFd4Bf_DqQncklWgAY2823xVst98o,7813
 geney/utils.py,sha256=KBdwNIywo7INVEQEsuIXauEJobvReE9TXAi5qqXanSI,2775
 geney/translation_initiation/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 geney/translation_initiation/tis_utils.py,sha256=AF3siFjuQH-Rs44EV-80zHdbxRMvN4woLFSHroWIETc,4448
 geney/translation_initiation/resources/kozak_pssm.json,sha256=pcd0Olziutq-6H3mFWDCD9cujQ_AlZO-iiOvBl82hqE,1165
 geney/translation_initiation/resources/tis_regressor_model.joblib,sha256=IXb4DUDhJ5rBDKcqMk9zE3ECTZZcdj7Jixz3KpoZ7OA,2592025
-geney-1.3.73.dist-info/METADATA,sha256=zBDPXEkIQIf58meB91wGEvGGPfe1rlR7-XZX9toMKdM,990
-geney-1.3.73.dist-info/WHEEL,sha256=AHX6tWk3qWuce7vKLrj7lnulVHEdWoltgauo8bgCXgU,109
-geney-1.3.73.dist-info/top_level.txt,sha256=O-FuNUMb5fn9dhZ-dYCgF0aZtfi1EslMstnzhc5IIVo,6
-geney-1.3.73.dist-info/RECORD,,
+geney-1.3.75.dist-info/METADATA,sha256=RJ7URjz5xQLQy1P8fISap3pWB7qdvknUp-c2Zjgtk9I,990
+geney-1.3.75.dist-info/WHEEL,sha256=AHX6tWk3qWuce7vKLrj7lnulVHEdWoltgauo8bgCXgU,109
+geney-1.3.75.dist-info/top_level.txt,sha256=O-FuNUMb5fn9dhZ-dYCgF0aZtfi1EslMstnzhc5IIVo,6
+geney-1.3.75.dist-info/RECORD,,