speconsense 0.7.4__py3-none-any.whl → 0.7.5__py3-none-any.whl

speconsense/__init__.py

@@ -5,7 +5,7 @@ A Python tool for experimental clustering and consensus generation as an alterna
 in the fungal DNA barcoding pipeline.
 """
 
-__version__ = "0.7.4"
+__version__ = "0.7.5"
 __author__ = "Josh Walker"
 __email__ = "joshowalker@yahoo.com"
 

speconsense/summarize/cli.py

@@ -392,19 +392,19 @@ def process_single_specimen(file_consensuses: List[ConsensusInfo],
             final_consensus.append(renamed_variant)
             group_naming.append((variant.sample_name, new_name))
 
-        # Generate full consensus from PRE-MERGE variants
+        # Generate full consensus from PRE-MERGE variants that contributed
+        # to surviving post-merge variants (after select-min-size-ratio)
         if getattr(args, 'enable_full_consensus', False):
-            pre_merge_variants = variant_groups[group_id]
-
-            # Apply size-ratio filter (same as merge pipeline)
-            if args.merge_min_size_ratio > 0 and len(pre_merge_variants) > 1:
-                largest_size = max(v.size for v in pre_merge_variants)
-                filtered = [v for v in pre_merge_variants
-                            if (v.size / largest_size) >= args.merge_min_size_ratio]
-                if len(filtered) < len(pre_merge_variants):
-                    filtered_count = len(pre_merge_variants) - len(filtered)
-                    logging.debug(f"Full consensus: filtered out {filtered_count} variants with size ratio < {args.merge_min_size_ratio} relative to largest (size={largest_size})")
-                    pre_merge_variants = filtered
+            # Collect original cluster names from surviving post-merge variants
+            surviving_originals = set()
+            for v in group_members:
+                if v.sample_name in all_merge_traceability:
+                    surviving_originals.update(all_merge_traceability[v.sample_name])
+                else:
+                    surviving_originals.add(v.sample_name)
+
+            pre_merge_variants = [v for v in variant_groups[group_id]
+                                  if v.sample_name in surviving_originals]
 
             specimen_base = selected_variants[0].sample_name.rsplit('-c', 1)[0]
             full_name = f"{specimen_base}-{group_idx + 1}.full"

{speconsense-0.7.4.dist-info → speconsense-0.7.5.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: speconsense
-Version: 0.7.4
+Version: 0.7.5
 Summary: High-quality clustering and consensus generation for Oxford Nanopore amplicon reads
 Author-email: Josh Walker <joshowalker@yahoo.com>
 License: BSD-3-Clause
@@ -295,14 +295,14 @@ When using `speconsense-summarize` for post-processing, creates `__Summary__/` d
 |---------------|-------------|------------|-------------|
 | **Original** | Source `cluster_debug/` | `-c1`, `-c2`, `-c3` | Preserves speconsense clustering results |
 | **Summarization** | `__Summary__/`, `FASTQ Files/`, `variants/` | `-1.v1`, `-1.v2`, `-2.v1`, `.raw1` | Post-processing groups and variants |
-| **Full consensus** | `__Summary__/` | `-1.full` | IUPAC consensus from all pre-merge variants in a group |
+| **Full consensus** | `__Summary__/` | `-1.full` | IUPAC consensus from pre-merge components of surviving variants |
 
 ### Example Directory Structure
 ```
 __Summary__/
 ├── sample-1.v1-RiC45.fasta                  # Primary variant (group 1, merged)
 ├── sample-1.v2-RiC23.fasta                  # Additional variant (not merged)
-├── sample-1.full-RiC68.fasta                # Full IUPAC consensus for group 1 (all pre-merge variants)
+├── sample-1.full-RiC68.fasta                # Full IUPAC consensus for group 1 (surviving variants' components)
 ├── sample-2.v1-RiC30.fasta                  # Second organism group, primary variant
 ├── summary.fasta                            # All final consensus sequences (excludes .raw)
 ├── summary.txt                              # Statistics
@@ -829,7 +829,7 @@ For high-throughput workflows (e.g., 100K sequences/year), this prioritization e
 ```bash
 speconsense-summarize --enable-full-consensus
 ```
-- Generates a full IUPAC consensus sequence per variant group from all pre-merge variants
+- Generates a full IUPAC consensus sequence per variant group from pre-merge variants that contributed to surviving post-merge variants
 - Output named `{specimen}-{group}.full-RiC{reads}.fasta` in the `__Summary__/` directory
 - Uses majority voting across all variants in the group; **gaps never win** — at each alignment column, the most common non-gap base is chosen, with IUPAC codes for ties among bases
 - Useful when you want a single representative sequence that captures all variation within a group as IUPAC ambiguity codes
@@ -1073,7 +1073,7 @@ The complete speconsense-summarize workflow operates in this order:
 4. **Homopolymer-aware MSA-based variant merging** within each group, including **overlap merging** for different-length sequences (`--disable-merging`, `--merge-effort`, `--merge-position-count`, `--merge-indel-length`, `--min-merge-overlap`, `--merge-snp`, `--merge-min-size-ratio`, `--disable-homopolymer-equivalence`)
 5. **Selection size ratio filtering** to remove tiny post-merge variants (`--select-min-size-ratio`)
 6. **Variant selection** within each group (`--select-max-variants`, `--select-strategy`)
-7. **Full consensus generation** (optional) — IUPAC consensus from all pre-merge variants per group (`--enable-full-consensus`)
+7. **Full consensus generation** (optional) — IUPAC consensus from pre-merge components of surviving post-merge variants (`--enable-full-consensus`)
 8. **Output generation** with customizable header fields (`--fasta-fields`) and full traceability
 
 **Key architectural features**:

{speconsense-0.7.4.dist-info → speconsense-0.7.5.dist-info}/RECORD

@@ -1,4 +1,4 @@
-speconsense/__init__.py,sha256=uLSZG2n0xobwuNT2PwZbytUg1DcyOr2aJlsbc52iKs0,537
+speconsense/__init__.py,sha256=qxithGom21C3MjbApvgOzVSIRFqw4jReMKZqipfY-Kk,537
 speconsense/cli.py,sha256=Kqb2da0IuazocAz72iqTnw71jI7UaQgxsHfb9CwiolU,85
 speconsense/msa.py,sha256=t1uDb-Tj5tDnB17QnNZPslpAiLXgAMIlnmMKBbwBKzs,31661
 speconsense/quality_report.py,sha256=Byrc115T03ybi7mpA0Bw8-gc83nhKPzDY0tyH1IIAMQ,19803
@@ -23,15 +23,15 @@ speconsense/scalability/vsearch.py,sha256=I1IzTeRzEFn9bi8mNbBRvtcHvUBzBFdE7D5yf-
 speconsense/summarize/__init__.py,sha256=PE6W9hytDxhkw7W6Fz8X3jd92N2VdhuxiQ72Nqm1xC0,3181
 speconsense/summarize/__main__.py,sha256=_hzLNqNtv4PirL1oMic37GW2QmjWquoznzNtld_3FiQ,117
 speconsense/summarize/analysis.py,sha256=1MXtKMpX1bgKEtI-JN6BwTQj99qyt1eQLqNg51EgPiE,31560
-speconsense/summarize/cli.py,sha256=uSeY7__KpdQVXqJcQ0Zpn6ePeyJDVGdml7rZgHFr3W8,27124
+speconsense/summarize/cli.py,sha256=JDgLQl8zFlbkH-Oa_n68lbUTB-cLBWzl3dphn041ph0,26961
 speconsense/summarize/clustering.py,sha256=kk-FdFCea8KRocowN_4dt_aoqZNVJMmEu7CVKPfYgK8,28346
 speconsense/summarize/fields.py,sha256=a6aK9hkPJ-sDRRSqM_7IkyqCki99KSMnsQMV-U7r2zY,8687
 speconsense/summarize/io.py,sha256=FdHLbcj0NOL3WE1e5OL85DRdJaHpyXPMcmlNg9mG3tM,32732
 speconsense/summarize/iupac.py,sha256=Y6KqELmnGy4Eya4C_4ldXY8uek0ReuSUgITLI3NW0-A,11042
 speconsense/summarize/merging.py,sha256=FakBey3qpu7ULPIsc2GDo9WG8jNU1L6q2pgQ2HrOKXk,28454
-speconsense-0.7.4.dist-info/licenses/LICENSE,sha256=T_VYPNbu9NSWjdQunfk4jqUGND_kYWe_An18s6N492o,1498
-speconsense-0.7.4.dist-info/METADATA,sha256=2vFyM5rqFEwMPIcsSAH32Dwh_bDA-bOk15D7El6MO7Y,79957
-speconsense-0.7.4.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-speconsense-0.7.4.dist-info/entry_points.txt,sha256=C0zFp5EYA8_KCb04uOyb4JNkxNH7bli1eU-XYrSX3BU,147
-speconsense-0.7.4.dist-info/top_level.txt,sha256=nYUJOHrqeX-OOxOYQKvpp7Iv8-Bed18wN1DBwWfJKnQ,12
-speconsense-0.7.4.dist-info/RECORD,,
+speconsense-0.7.5.dist-info/licenses/LICENSE,sha256=T_VYPNbu9NSWjdQunfk4jqUGND_kYWe_An18s6N492o,1498
+speconsense-0.7.5.dist-info/METADATA,sha256=JiEg26k5JEJUFSR5hw4FHHysL4ei9PPyvVXsGBRBZVc,80041
+speconsense-0.7.5.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+speconsense-0.7.5.dist-info/entry_points.txt,sha256=C0zFp5EYA8_KCb04uOyb4JNkxNH7bli1eU-XYrSX3BU,147
+speconsense-0.7.5.dist-info/top_level.txt,sha256=nYUJOHrqeX-OOxOYQKvpp7Iv8-Bed18wN1DBwWfJKnQ,12
+speconsense-0.7.5.dist-info/RECORD,,