geney 1.2.5__py2.py3-none-any.whl → 1.2.6__py2.py3-none-any.whl

geney/oncosplice.py

@@ -545,7 +545,7 @@ class Transcript:
     def generate_mature_mrna_pos(self, reset=True):
         if reset:
             pre_seq_pos, pre_indices_pos = self.generate_pre_mrna_pos()
-            self.pre_mrna, _ = self.__pos2sense(pre_seq_pos, pre_indices_pos)
+            self.pre_mrna, self.pre_indices = self.__pos2sense(pre_seq_pos, pre_indices_pos)
         else:
             pre_seq_pos, pre_indices_pos = self.__sense2pos(self.pre_mrna, self.pre_indices)
 
@@ -725,7 +725,17 @@ def pang_one_hot_encode(seq):
     seq = np.asarray(list(map(int, list(seq))))
     return IN_MAP[seq.astype('int8')]
 
+
+def get_pos_seq_indices(t):
+    seq, indices = t.pre_mrna, t.pre_indices
+    if t.rev:
+        return reverse_complement(seq), indices[::-1]
+    else:
+        return seq, indices
+
+
 def pangolin_predict_probs(true_seq, models):
+    # print(f"Running pangolin on: {true_seq}")
     model_nums = [0, 2, 4, 6]
     INDEX_MAP = {0: 1, 1: 2, 2: 4, 3: 5, 4: 7, 5: 8, 6: 10, 7: 11}
 
@@ -754,42 +764,36 @@ def pangolin_predict_probs(true_seq, models):
     acceptor_probs = [splicing_pred[i] * acceptor_dinucleotide[i] for i in range(len(true_seq))]
     return donor_probs[5000:-5000], acceptor_probs[5000:-5000]
 
-def run_spliceai_transcript(mutations, transcript_data, sai_mrg_context=5000, min_coverage=2500, sai_threshold=0.5, engine='spliceai'):
+
+def find_transcript_missplicing(mutations, ref_transcript, var_transcript, context=7500, threshold=0.5,
+                                engine='spliceai'):
     positions = mutations.positions
     end_positions = [m.start + len(m.ref) for m in mutations.variants]
     positions.extend(end_positions)
+    center = int(np.mean(positions) // 1)
+
+    seq_start_pos, seq_end_pos = center - context, center + context
+    transcript_start, transcript_end, rev = ref_transcript.transcript_lower, ref_transcript.transcript_upper, ref_transcript.rev
+
+    # Generate reference sequence data
+    ref_seq, ref_indices = get_pos_seq_indices(ref_transcript)
+    center_index = ref_indices.index(center)
+    start_cutoff = ref_indices.index(seq_start_pos) if seq_start_pos in ref_indices else 0
+    end_cutoff = ref_indices.index(seq_end_pos) if seq_end_pos in ref_indices else len(ref_indices)
+    start_pad, end_pad = max(0, context - (center_index - start_cutoff)), max(0, context - (end_cutoff - center_index))
+    ref_seq = 'N' * start_pad + ref_seq[start_cutoff:end_cutoff] + 'N' * end_pad
+    ref_indices = [-1] * start_pad + ref_indices[start_cutoff:end_cutoff] + [-1] * end_pad
+
+    # Generate mutation sequence data
+    mut_seq, mut_indices = get_pos_seq_indices(var_transcript)
+    start_cutoff = mut_indices.index(seq_start_pos) if seq_start_pos in mut_indices else 0
+    end_cutoff = mut_indices.index(seq_end_pos) if seq_end_pos in mut_indices else len(mut_indices)
+    start_pad, end_pad = max(0, context - (center_index - start_cutoff)), max(0, context - (end_cutoff - center_index))
+    mut_seq = 'N' * start_pad + mut_seq[start_cutoff:end_cutoff] + 'N' * end_pad
+    mut_indices = [-1] * start_pad + mut_indices[start_cutoff:end_cutoff] + [-1] * end_pad
 
-    seq_start_pos = min(positions) - sai_mrg_context - min_coverage
-    seq_end_pos = max(positions) + sai_mrg_context + min_coverage
-
-    fasta_obj = Fasta_segment()
-    ref_seq, ref_indices = fasta_obj.read_segment_endpoints(
-        config_setup[transcript_data.organism]['CHROM_SOURCE'] / f'chr{mutations.chrom}.fasta',
-        seq_start_pos,
-        seq_end_pos)
-
-    transcript_start, transcript_end, rev = transcript_data.transcript_lower, transcript_data.transcript_upper, transcript_data.rev
-
-    # visible_donors = np.intersect1d(transcript_data.donors, ref_indices)
-    # visible_acceptors = np.intersect1d(transcript_data.acceptors, ref_indices)
-
-    start_pad = ref_indices.index(transcript_start) if transcript_start in ref_indices else 0
-    end_cutoff = ref_indices.index(transcript_end) if transcript_end in ref_indices else len(ref_indices)
-    end_pad = len(ref_indices) - end_cutoff
-    ref_seq = 'N' * start_pad + ref_seq[start_pad:end_cutoff] + 'N' * end_pad
-    ref_indices = [-1] * start_pad + ref_indices[start_pad:end_cutoff] + [-1] * end_pad
-    mut_seq, mut_indices = ref_seq, ref_indices
-
-    for mut in mutations:
-        mut_seq, mut_indices = generate_mut_variant(seq=mut_seq, indices=mut_indices, mut=mut)
-
-    ref_indices = ref_indices[sai_mrg_context:-sai_mrg_context]
-    mut_indices = mut_indices[sai_mrg_context:-sai_mrg_context]
     copy_mut_indices = mut_indices.copy()
 
-    visible_donors = np.intersect1d(transcript_data.donors, ref_indices)
-    visible_acceptors = np.intersect1d(transcript_data.acceptors, ref_indices)
-
     if rev:
         ref_seq = reverse_complement(ref_seq)
         mut_seq = reverse_complement(mut_seq)
@@ -801,21 +805,28 @@ def run_spliceai_transcript(mutations, transcript_data, sai_mrg_context=5000, mi
         mut_seq_probs_temp = sai_predict_probs(mut_seq, sai_models)
         ref_seq_acceptor_probs, ref_seq_donor_probs = ref_seq_probs_temp[0, :], ref_seq_probs_temp[1, :]
         mut_seq_acceptor_probs, mut_seq_donor_probs = mut_seq_probs_temp[0, :], mut_seq_probs_temp[1, :]
+        ref_indices, mut_indices = ref_indices[5000:-5000], mut_indices[5000:-5000]
+
 
     elif engine == 'pangolin':
-        ref_seq_donor_probs, ref_seq_acceptor_probs = pangolin_predict_probs(ref_seq, pangolin_models=pang_models)
-        mut_seq_donor_probs, mut_seq_acceptor_probs = pangolin_predict_probs(mut_seq, pangolin_models=pang_models)
+        ref_seq_donor_probs, ref_seq_acceptor_probs = pangolin_predict_probs(ref_seq, models=pang_models)
+        mut_seq_donor_probs, mut_seq_acceptor_probs = pangolin_predict_probs(mut_seq, models=pang_models)
+        ref_indices, mut_indices = ref_indices[5000:-5000], mut_indices[5000:-5000]
 
     else:
         raise ValueError(f"{engine} not implemented")
 
+    visible_donors = np.intersect1d(ref_transcript.donors, ref_indices)
+    visible_acceptors = np.intersect1d(ref_transcript.acceptors, ref_indices)
+    # print(ref_indices.index(visible_donors[0]), ref_seq_donor_probs[ref_indices.index(visible_donors[0])], mut_seq_donor_probs[mut_indices.index(visible_donors[0])])
+
     assert len(ref_indices) == len(ref_seq_acceptor_probs), 'Reference pos not the same'
     assert len(mut_indices) == len(mut_seq_acceptor_probs), 'Mut pos not the same'
 
     iap, dap = find_ss_changes({p: v for p, v in list(zip(ref_indices, ref_seq_acceptor_probs))},
                                {p: v for p, v in list(zip(mut_indices, mut_seq_acceptor_probs))},
                                visible_acceptors,
-                               threshold=sai_threshold)
+                               threshold=threshold)
 
     assert len(ref_indices) == len(ref_seq_donor_probs), 'Reference pos not the same'
     assert len(mut_indices) == len(mut_seq_donor_probs), 'Mut pos not the same'
@@ -823,13 +834,15 @@ def run_spliceai_transcript(mutations, transcript_data, sai_mrg_context=5000, mi
     idp, ddp = find_ss_changes({p: v for p, v in list(zip(ref_indices, ref_seq_donor_probs))},
                                {p: v for p, v in list(zip(mut_indices, mut_seq_donor_probs))},
                                visible_donors,
-                               threshold=sai_threshold)
+                               threshold=threshold)
 
     ref_acceptors = {a: b for a, b in list(zip(ref_indices, ref_seq_acceptor_probs))}
     ref_donors = {a: b for a, b in list(zip(ref_indices, ref_seq_donor_probs))}
 
-    lost_acceptors = {int(p): {'absolute': np.float64(0), 'delta': round(float(-ref_acceptors[p]), 3)} for p in visible_acceptors if p not in mut_indices and p not in dap}
-    lost_donors = {int(p): {'absolute': np.float64(0), 'delta': round(float(-ref_donors[p]), 3)} for p in visible_donors if p not in mut_indices and p not in ddp}
+    lost_acceptors = {int(p): {'absolute': np.float64(0), 'delta': round(float(-ref_acceptors[p]), 3)} for p in
+                      visible_acceptors if p not in mut_indices and p not in dap}
+    lost_donors = {int(p): {'absolute': np.float64(0), 'delta': round(float(-ref_donors[p]), 3)} for p in visible_donors
+                   if p not in mut_indices and p not in ddp}
     dap.update(lost_acceptors)
     ddp.update(lost_donors)
 
@@ -838,6 +851,90 @@ def run_spliceai_transcript(mutations, transcript_data, sai_mrg_context=5000, mi
     return {outk: {int(k) if k.is_integer() else k: v for k, v in outv.items()} for outk, outv in missplicing.items()}
 
 
+# def run_spliceai_transcript(mutations, transcript_data, sai_mrg_context=5000, min_coverage=2500, sai_threshold=0.5, engine='spliceai'):
+#     positions = mutations.positions
+#     end_positions = [m.start + len(m.ref) for m in mutations.variants]
+#     positions.extend(end_positions)
+#
+#     seq_start_pos = min(positions) - sai_mrg_context - min_coverage
+#     seq_end_pos = max(positions) + sai_mrg_context + min_coverage
+#
+#     fasta_obj = Fasta_segment()
+#     ref_seq, ref_indices = fasta_obj.read_segment_endpoints(
+#         config_setup[transcript_data.organism]['CHROM_SOURCE'] / f'chr{mutations.chrom}.fasta',
+#         seq_start_pos,
+#         seq_end_pos)
+#
+#     transcript_start, transcript_end, rev = transcript_data.transcript_lower, transcript_data.transcript_upper, transcript_data.rev
+#
+#     # visible_donors = np.intersect1d(transcript_data.donors, ref_indices)
+#     # visible_acceptors = np.intersect1d(transcript_data.acceptors, ref_indices)
+#
+#     start_pad = ref_indices.index(transcript_start) if transcript_start in ref_indices else 0
+#     end_cutoff = ref_indices.index(transcript_end) if transcript_end in ref_indices else len(ref_indices)
+#     end_pad = len(ref_indices) - end_cutoff
+#     ref_seq = 'N' * start_pad + ref_seq[start_pad:end_cutoff] + 'N' * end_pad
+#     ref_indices = [-1] * start_pad + ref_indices[start_pad:end_cutoff] + [-1] * end_pad
+#     mut_seq, mut_indices = ref_seq, ref_indices
+#
+#     for mut in mutations:
+#         mut_seq, mut_indices = generate_mut_variant(seq=mut_seq, indices=mut_indices, mut=mut)
+#
+#     ref_indices = ref_indices[sai_mrg_context:-sai_mrg_context]
+#     mut_indices = mut_indices[sai_mrg_context:-sai_mrg_context]
+#     copy_mut_indices = mut_indices.copy()
+#
+#     visible_donors = np.intersect1d(transcript_data.donors, ref_indices)
+#     visible_acceptors = np.intersect1d(transcript_data.acceptors, ref_indices)
+#
+#     if rev:
+#         ref_seq = reverse_complement(ref_seq)
+#         mut_seq = reverse_complement(mut_seq)
+#         ref_indices = ref_indices[::-1]
+#         mut_indices = mut_indices[::-1]
+#
+#     if engine == 'spliceai':
+#         ref_seq_probs_temp = sai_predict_probs(ref_seq, sai_models)
+#         mut_seq_probs_temp = sai_predict_probs(mut_seq, sai_models)
+#         ref_seq_acceptor_probs, ref_seq_donor_probs = ref_seq_probs_temp[0, :], ref_seq_probs_temp[1, :]
+#         mut_seq_acceptor_probs, mut_seq_donor_probs = mut_seq_probs_temp[0, :], mut_seq_probs_temp[1, :]
+#
+#     elif engine == 'pangolin':
+#         ref_seq_donor_probs, ref_seq_acceptor_probs = pangolin_predict_probs(ref_seq, pangolin_models=pang_models)
+#         mut_seq_donor_probs, mut_seq_acceptor_probs = pangolin_predict_probs(mut_seq, pangolin_models=pang_models)
+#
+#     else:
+#         raise ValueError(f"{engine} not implemented")
+#
+#     assert len(ref_indices) == len(ref_seq_acceptor_probs), 'Reference pos not the same'
+#     assert len(mut_indices) == len(mut_seq_acceptor_probs), 'Mut pos not the same'
+#
+#     iap, dap = find_ss_changes({p: v for p, v in list(zip(ref_indices, ref_seq_acceptor_probs))},
+#                                {p: v for p, v in list(zip(mut_indices, mut_seq_acceptor_probs))},
+#                                visible_acceptors,
+#                                threshold=sai_threshold)
+#
+#     assert len(ref_indices) == len(ref_seq_donor_probs), 'Reference pos not the same'
+#     assert len(mut_indices) == len(mut_seq_donor_probs), 'Mut pos not the same'
+#
+#     idp, ddp = find_ss_changes({p: v for p, v in list(zip(ref_indices, ref_seq_donor_probs))},
+#                                {p: v for p, v in list(zip(mut_indices, mut_seq_donor_probs))},
+#                                visible_donors,
+#                                threshold=sai_threshold)
+#
+#     ref_acceptors = {a: b for a, b in list(zip(ref_indices, ref_seq_acceptor_probs))}
+#     ref_donors = {a: b for a, b in list(zip(ref_indices, ref_seq_donor_probs))}
+#
+#     lost_acceptors = {int(p): {'absolute': np.float64(0), 'delta': round(float(-ref_acceptors[p]), 3)} for p in visible_acceptors if p not in mut_indices and p not in dap}
+#     lost_donors = {int(p): {'absolute': np.float64(0), 'delta': round(float(-ref_donors[p]), 3)} for p in visible_donors if p not in mut_indices and p not in ddp}
+#     dap.update(lost_acceptors)
+#     ddp.update(lost_donors)
+#
+#     missplicing = {'missed_acceptors': dap, 'missed_donors': ddp, 'discovered_acceptors': iap, 'discovered_donors': idp}
+#     missplicing = {outk: {float(k): v for k, v in outv.items()} for outk, outv in missplicing.items()}
+#     return {outk: {int(k) if k.is_integer() else k: v for k, v in outv.items()} for outk, outv in missplicing.items()}
+
+
 # def run_spliceai(mutations, gene_data, sai_mrg_context=5000, min_coverage=2500, sai_threshold=0.5):
 #     positions = mutations.positions
 #     seq_start_pos = min(positions) - sai_mrg_context - min_coverage

{geney-1.2.5.dist-info → geney-1.2.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: geney
-Version: 1.2.5
+Version: 1.2.6
 Summary: A Python package for gene expression modeling.
 Home-page: https://github.com/nicolaslynn/geney
 Author: Nicolas Lynn

{geney-1.2.5.dist-info → geney-1.2.6.dist-info}/RECORD

@@ -9,7 +9,7 @@ geney/gtex.py,sha256=asL2lHyU5KsbWpV096vkf1Ka7hSo_RRfZqw7p5nERmE,1919
 geney/gtex_utils.py,sha256=asL2lHyU5KsbWpV096vkf1Ka7hSo_RRfZqw7p5nERmE,1919
 geney/immune_utils.py,sha256=ZRni5ttrhpYBnmNr0d0ZatIbNPYs4nmQuoUO00SpsS4,5271
 geney/netchop.py,sha256=AMiy9YsdTmX4B3k3Y5Yh7EmoGAojM1O3AzhPKOiB--g,3050
-geney/oncosplice.py,sha256=9oZs9W_bI6O5h3284WvatkerhSCaxMZWfs1xVc1lJO0,71524
+geney/oncosplice.py,sha256=aaOxri0bXLPfB3Mu8tkzk4KEVlzzocImzj3MI-0uU_0,76949
 geney/oncosplice_mouse.py,sha256=LYLOukI9qI1IBkyl1qVRFR5d1NAw7Orlj8Zth-4xCW8,12962
 geney/oncosplice_pipeline.py,sha256=hpGqFHOdn8i8tvvs1-t3-G9Ko18zInwoDXBJbbrfbC4,68036
 geney/performance_utils.py,sha256=FQt7rA4r-Wuq3kceCxsSuMfj3wU1tMG8QnbL59aBohs,4700
@@ -45,7 +45,7 @@ geney/translation_initiation/resources/kozak_pssm.json,sha256=pcd0Olziutq-6H3mFW
 geney/translation_initiation/resources/tis_regressor_model.joblib,sha256=IXb4DUDhJ5rBDKcqMk9zE3ECTZZcdj7Jixz3KpoZ7OA,2592025
 geney/translation_termination/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 geney/translation_termination/tts_utils.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-geney-1.2.5.dist-info/METADATA,sha256=9UptuZVJWZvVN6Y9KgPUxrC4gnijFVW4CtkkESxrY9E,1198
-geney-1.2.5.dist-info/WHEEL,sha256=iYlv5fX357PQyRT2o6tw1bN-YcKFFHKqB_LwHO5wP-g,110
-geney-1.2.5.dist-info/top_level.txt,sha256=O-FuNUMb5fn9dhZ-dYCgF0aZtfi1EslMstnzhc5IIVo,6
-geney-1.2.5.dist-info/RECORD,,
+geney-1.2.6.dist-info/METADATA,sha256=hcHNz2oNxRLzNpimwrHeM2yKSnd1z5KHJ_Rl86LD3QE,1198
+geney-1.2.6.dist-info/WHEEL,sha256=iYlv5fX357PQyRT2o6tw1bN-YcKFFHKqB_LwHO5wP-g,110
+geney-1.2.6.dist-info/top_level.txt,sha256=O-FuNUMb5fn9dhZ-dYCgF0aZtfi1EslMstnzhc5IIVo,6
+geney-1.2.6.dist-info/RECORD,,