MetaPont 0.1.2__tar.gz → 0.1.4__tar.gz

{metapont-0.1.2/src/MetaPont.egg-info → metapont-0.1.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: MetaPont
-Version: 0.1.2
+Version: 0.1.4
 Summary: Integrated analysis of metagenomic assembly, readmapping, Kraken2 and eggNOG-mapper outputs.
 Author-email: Nicholas Dimonaco <nicholas@dimonaco.co.uk>
 License-Expression: GPL-3.0-or-later
@@ -20,7 +20,7 @@ Provides-Extra: test
 Requires-Dist: pytest>=7.0; extra == "test"
 Dynamic: license-file
 
-# MetaPont v0.1.2
+# MetaPont v0.1.4
 
 MetaPont integrates existing metagenomic workflow outputs into per-sample and
 cross-sample tables for contigs, taxonomy, readmapping support, and eggNOG-mapper
@@ -79,10 +79,20 @@ sample directory for the relevant files. The main inputs are:
 | Contig FASTA | Contig inventory and lengths when a readmapping contig summary is absent |
 
 Inputs are discovered independently, so incomplete samples can still be
-included. Missing or stale outputs are reported as warnings; MetaPont does not
-rerun missing upstream analyses. Kraken2 assignments shorter than the selected
-minimum contig length are omitted from taxonomy attribution by default. This
-filter does not rerun Kraken2.
+included. Missing or empty core inputs are reported as warnings; MetaPont does not
+rerun missing upstream analyses. The inventory checks the files MetaPont
+actually consumes: contigs, Pyrodigal GFF/protein FASTA, eggNOG annotation TSV,
+Kraken2 per-contig output/report, and readmapping contig/CDS summaries. Optional
+sidecars (for example BAM/index, eggNOG XLSX, MPA report, Pyrodigal NT/QC files)
+do not make an otherwise usable MetaPont input set incomplete.
+
+MetaPont deliberately does not use modification-time comparisons to decide
+whether outputs are stale. Scratch-retention commands such as `touch` can
+update timestamps without changing file contents, so timestamp ordering is not
+a reliable freshness test. The inventory reports presence/completeness, not
+whether an upstream tool was rerun against the latest source assembly. Kraken2
+assignments shorter than the selected minimum contig length are omitted from
+taxonomy attribution by default. This filter does not rerun Kraken2.
 
 ## Commands
 
@@ -103,9 +113,10 @@ Useful options:
 | Option | Default | Description |
 | --- | --- | --- |
 | `--force` | Off | Regenerate outputs even when the input signature matches a cached run. |
-| `--no-heal-flags` | Off | Do not recreate missing upstream success flags when their outputs are complete and current. |
+| `--no-heal-flags` | Off | Do not recreate missing upstream success flags when the required non-empty core outputs are present. |
 | `--min-contig-length N` | `2500` | Minimum Kraken2-assigned contig length retained for taxonomy attribution; use `0` to disable the filter. |
-| `--taxon-function-abundance` | Off | Build the expanded taxon-by-function table. This can use substantial memory and disk space. |
+| `--taxon-function-abundance` | Off | Build the expanded taxon-by-function table using a disk-backed SQLite aggregation. This adds disk I/O and temporary disk usage. |
+| `--signature-mode` | `sampled` | Cache input detection. `sampled` hashes seven evenly spaced 128 KiB blocks from each input and is insensitive to `touch`; `full` hashes every byte and is slower but provides whole-file SHA256 verification. |
 | `--plots` | Off | Write optional quick-look PNG plots. Requires the `plots` installation extra. |
 
 `metapont combine` is an alias for `metapont run`. The `metapont-combine`
@@ -117,9 +128,13 @@ command is also available for compatibility with the earlier combine command.
 metapont inventory --root-dir /path/to/workflow-output --prefix E
 ```
 
-The inventory reports discovered inputs, expected outputs, success flags, and
-warnings for each sample. Add `--heal-flags` to recreate a missing stage flag
-only when its corresponding output files are complete and current.
+The inventory reports discovered inputs, core output availability, success
+flags, and warnings for each sample. Add `--heal-flags` to recreate a missing
+stage flag only when the full expected product set for that upstream stage is
+present (including its QC/sidecar outputs where applicable). This is stricter
+than MetaPont's integration-input check, so it does not mark an upstream stage
+complete merely because MetaPont can consume a subset of its outputs. Timestamp
+freshness is intentionally not inferred.
 
 ### Summarise readmapping coverage
 
@@ -140,7 +155,7 @@ Find taxa supporting a functional ID:
 
 ```bash
 metapont query-function \
-  --root-dir /path/to/workflow-output \
+  --root-dir /path/to/MetaPont_Results \
   --function-id GO:0016597 \
   --category GO \
   --rank Genus \
@@ -157,7 +172,7 @@ Find the most supported features associated with a taxon:
 
 ```bash
 metapont query-taxon \
-  --root-dir /path/to/workflow-output \
+  --root-dir /path/to/MetaPont_Results \
   --taxon Bacillus \
   --category GO \
   --rank Genus \
@@ -170,6 +185,20 @@ full lineage. `--category` is required for this query. Both query commands
 accept an explicit `--gene-table` path; plain TSV and gzip-compressed TSV files
 are supported.
 
+Build the expanded taxon-by-function table **after** a normal MetaPont run,
+without re-parsing the original eggNOG files or rerunning the integrated
+analysis:
+
+```bash
+metapont taxon-function-abundance \
+  --root-dir /path/to/workflow-output
+```
+
+Alternatively, provide `--gene-table` to use an existing
+`gene_annotations.tsv.gz`. This post-processing command uses a temporary
+SQLite database to keep aggregation memory bounded. The output is written to
+`MetaPont_Results/functional/taxon_function_abundance.tsv` by default.
+
 ## Results
 
 The default output tree is:
@@ -182,7 +211,7 @@ MetaPont_Results/
 ├── functional/
 │   ├── functional_abundance.tsv
 │   ├── matrices/
-│   └── taxon_function_abundance.tsv
+│   └── taxon_function_abundance.tsv  # optional; only when requested
 ├── inventory/
 │   └── sample_inventory.tsv
 ├── metadata/
@@ -212,9 +241,16 @@ Supported functional categories include `COG`, `eggNOG_OG`, `GO`, `EC`,
 is not a count of unique reads across genes. A read can support multiple genes
 or annotations, so functional support values are not mutually exclusive.
 Taxonomic relative abundance uses total mapped-read support as its denominator.
-
-Samples with missing or stale workflow outputs are retained where possible and
-listed in `summary/pipeline_warnings.tsv`. A clean run writes
+The new `Percent_Mapped_Reads_on_Classified_Contigs` field in the sample summary
+shows how much of that denominator lies on Kraken-classified contigs.
+
+Samples with missing or empty core workflow inputs are retained where
+possible and listed in `summary/pipeline_warnings.tsv`. The sample summary also
+reports genes processed versus genes with at least one functional feature, and
+readmapping support on Kraken-classified, unclassified, length-filtered and
+Kraken-unassigned contigs. `Genes_Annotated` remains for compatibility and is
+equivalent to `Genes_Processed`; use `Genes_With_Function` when you specifically
+mean genes with a parsed functional feature. A clean run writes
 `MetaPont_analysis.success`; a run with warnings writes
 `completed_with_warnings.flag`.
 

{metapont-0.1.2 → metapont-0.1.4}/README.md

@@ -1,4 +1,4 @@
-# MetaPont v0.1.2
+# MetaPont v0.1.4
 
 MetaPont integrates existing metagenomic workflow outputs into per-sample and
 cross-sample tables for contigs, taxonomy, readmapping support, and eggNOG-mapper
@@ -57,10 +57,20 @@ sample directory for the relevant files. The main inputs are:
 | Contig FASTA | Contig inventory and lengths when a readmapping contig summary is absent |
 
 Inputs are discovered independently, so incomplete samples can still be
-included. Missing or stale outputs are reported as warnings; MetaPont does not
-rerun missing upstream analyses. Kraken2 assignments shorter than the selected
-minimum contig length are omitted from taxonomy attribution by default. This
-filter does not rerun Kraken2.
+included. Missing or empty core inputs are reported as warnings; MetaPont does not
+rerun missing upstream analyses. The inventory checks the files MetaPont
+actually consumes: contigs, Pyrodigal GFF/protein FASTA, eggNOG annotation TSV,
+Kraken2 per-contig output/report, and readmapping contig/CDS summaries. Optional
+sidecars (for example BAM/index, eggNOG XLSX, MPA report, Pyrodigal NT/QC files)
+do not make an otherwise usable MetaPont input set incomplete.
+
+MetaPont deliberately does not use modification-time comparisons to decide
+whether outputs are stale. Scratch-retention commands such as `touch` can
+update timestamps without changing file contents, so timestamp ordering is not
+a reliable freshness test. The inventory reports presence/completeness, not
+whether an upstream tool was rerun against the latest source assembly. Kraken2
+assignments shorter than the selected minimum contig length are omitted from
+taxonomy attribution by default. This filter does not rerun Kraken2.
 
 ## Commands
 
@@ -81,9 +91,10 @@ Useful options:
 | Option | Default | Description |
 | --- | --- | --- |
 | `--force` | Off | Regenerate outputs even when the input signature matches a cached run. |
-| `--no-heal-flags` | Off | Do not recreate missing upstream success flags when their outputs are complete and current. |
+| `--no-heal-flags` | Off | Do not recreate missing upstream success flags when the required non-empty core outputs are present. |
 | `--min-contig-length N` | `2500` | Minimum Kraken2-assigned contig length retained for taxonomy attribution; use `0` to disable the filter. |
-| `--taxon-function-abundance` | Off | Build the expanded taxon-by-function table. This can use substantial memory and disk space. |
+| `--taxon-function-abundance` | Off | Build the expanded taxon-by-function table using a disk-backed SQLite aggregation. This adds disk I/O and temporary disk usage. |
+| `--signature-mode` | `sampled` | Cache input detection. `sampled` hashes seven evenly spaced 128 KiB blocks from each input and is insensitive to `touch`; `full` hashes every byte and is slower but provides whole-file SHA256 verification. |
 | `--plots` | Off | Write optional quick-look PNG plots. Requires the `plots` installation extra. |
 
 `metapont combine` is an alias for `metapont run`. The `metapont-combine`
@@ -95,9 +106,13 @@ command is also available for compatibility with the earlier combine command.
 metapont inventory --root-dir /path/to/workflow-output --prefix E
 ```
 
-The inventory reports discovered inputs, expected outputs, success flags, and
-warnings for each sample. Add `--heal-flags` to recreate a missing stage flag
-only when its corresponding output files are complete and current.
+The inventory reports discovered inputs, core output availability, success
+flags, and warnings for each sample. Add `--heal-flags` to recreate a missing
+stage flag only when the full expected product set for that upstream stage is
+present (including its QC/sidecar outputs where applicable). This is stricter
+than MetaPont's integration-input check, so it does not mark an upstream stage
+complete merely because MetaPont can consume a subset of its outputs. Timestamp
+freshness is intentionally not inferred.
 
 ### Summarise readmapping coverage
 
@@ -118,7 +133,7 @@ Find taxa supporting a functional ID:
 
 ```bash
 metapont query-function \
-  --root-dir /path/to/workflow-output \
+  --root-dir /path/to/MetaPont_Results \
   --function-id GO:0016597 \
   --category GO \
   --rank Genus \
@@ -135,7 +150,7 @@ Find the most supported features associated with a taxon:
 
 ```bash
 metapont query-taxon \
-  --root-dir /path/to/workflow-output \
+  --root-dir /path/to/MetaPont_Results \
   --taxon Bacillus \
   --category GO \
   --rank Genus \
@@ -148,6 +163,20 @@ full lineage. `--category` is required for this query. Both query commands
 accept an explicit `--gene-table` path; plain TSV and gzip-compressed TSV files
 are supported.
 
+Build the expanded taxon-by-function table **after** a normal MetaPont run,
+without re-parsing the original eggNOG files or rerunning the integrated
+analysis:
+
+```bash
+metapont taxon-function-abundance \
+  --root-dir /path/to/workflow-output
+```
+
+Alternatively, provide `--gene-table` to use an existing
+`gene_annotations.tsv.gz`. This post-processing command uses a temporary
+SQLite database to keep aggregation memory bounded. The output is written to
+`MetaPont_Results/functional/taxon_function_abundance.tsv` by default.
+
 ## Results
 
 The default output tree is:
@@ -160,7 +189,7 @@ MetaPont_Results/
 ├── functional/
 │   ├── functional_abundance.tsv
 │   ├── matrices/
-│   └── taxon_function_abundance.tsv
+│   └── taxon_function_abundance.tsv  # optional; only when requested
 ├── inventory/
 │   └── sample_inventory.tsv
 ├── metadata/
@@ -190,9 +219,16 @@ Supported functional categories include `COG`, `eggNOG_OG`, `GO`, `EC`,
 is not a count of unique reads across genes. A read can support multiple genes
 or annotations, so functional support values are not mutually exclusive.
 Taxonomic relative abundance uses total mapped-read support as its denominator.
-
-Samples with missing or stale workflow outputs are retained where possible and
-listed in `summary/pipeline_warnings.tsv`. A clean run writes
+The new `Percent_Mapped_Reads_on_Classified_Contigs` field in the sample summary
+shows how much of that denominator lies on Kraken-classified contigs.
+
+Samples with missing or empty core workflow inputs are retained where
+possible and listed in `summary/pipeline_warnings.tsv`. The sample summary also
+reports genes processed versus genes with at least one functional feature, and
+readmapping support on Kraken-classified, unclassified, length-filtered and
+Kraken-unassigned contigs. `Genes_Annotated` remains for compatibility and is
+equivalent to `Genes_Processed`; use `Genes_With_Function` when you specifically
+mean genes with a parsed functional feature. A clean run writes
 `MetaPont_analysis.success`; a run with warnings writes
 `completed_with_warnings.flag`.
 

{metapont-0.1.2 → metapont-0.1.4}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "MetaPont"
-version = "0.1.2"
+version = "0.1.4"
 authors = [
     { name = "Nicholas Dimonaco", email = "nicholas@dimonaco.co.uk" }
 ]

{metapont-0.1.2 → metapont-0.1.4}/src/MetaPont/__init__.py

@@ -1,6 +1,6 @@
 """MetaPont: integrated analysis of metagenomic workflow outputs."""
 
-__version__ = "0.1.2"
+__version__ = "0.1.4"
 MetaPont_Version = f"v{__version__}"
 
 __all__ = ["__version__", "MetaPont_Version"]

{metapont-0.1.2 → metapont-0.1.4}/src/MetaPont/analysis.py

@@ -11,7 +11,7 @@ from typing import Any
 
 from .constants import FEATURE_COLUMNS
 from .layout import discover_samples
-from .parsers import open_text, parse_contig_summary
+from .parsers import open_text, parse_contig_summary, split_features
 from .pipeline import write_tsv
 
 
@@ -25,12 +25,15 @@ def _read_gene_table(path: str | Path):
 def coverage_summary(root_dir: str | Path, prefixes: str, output: str | Path | None = None,
                      read_length: int | None = None) -> tuple[Path, int]:
     root = Path(root_dir).expanduser().resolve()
+    if read_length is not None and read_length <= 0:
+        raise ValueError("read_length must be a positive integer.")
     out = Path(output).expanduser().resolve() if output else root / "MetaPont_Results" / "summary" / "contig_coverage_summary.tsv"
     samples = discover_samples(root, prefixes)
     rows: list[dict[str, Any]] = []
     for sample in samples:
-        if sample.contig_summary is None:
-            rows.append({"Sample": sample.name, "Status": "missing_contig_summary"})
+        if sample.contig_summary is None or not sample.contig_summary.is_file() or sample.contig_summary.stat().st_size == 0:
+            rows.append({"Sample": sample.name, "Status": "missing_or_empty_contig_summary",
+                         "Contig_Summary": str(sample.contig_summary or "")})
             continue
         n_contigs = total_mapped = total_reported = 0
         mapped_counts: list[int] = []
@@ -88,8 +91,9 @@ def coverage_summary(root_dir: str | Path, prefixes: str, output: str | Path | N
     return out, len(rows)
 
 
-def _feature_values(cell: str) -> list[str]:
-    return [part.strip() for part in cell.replace(",", "|").replace(";", "|").split("|") if part.strip() and part.strip() != "-"]
+def _feature_values(cell: str, category: str) -> list[str]:
+    """Use the same feature parsing rules as the main annotation pipeline."""
+    return split_features(cell, category)
 
 
 def query_function(gene_table: str | Path, function_id: str, output: str | Path,
@@ -98,6 +102,8 @@ def query_function(gene_table: str | Path, function_id: str, output: str | Path,
     path = Path(gene_table)
     if not path.is_file():
         raise FileNotFoundError(f"Gene annotation table not found: {path}")
+    if top_n is not None and top_n < 1:
+        raise ValueError("top_n must be a positive integer.")
     cat_norm = category.upper().replace("-", "_") if category else None
     cat_aliases = {k.upper(): k for k in FEATURE_COLUMNS}
     if cat_norm and cat_norm not in cat_aliases:
@@ -110,7 +116,7 @@ def query_function(gene_table: str | Path, function_id: str, output: str | Path,
     denominator: dict[str, int] = defaultdict(int)
     categories = [chosen_cat] if chosen_cat else list(FEATURE_COLUMNS)
     for row in _read_gene_table(path):
-        found = any(function_id in _feature_values(row.get(cat, "")) for cat in categories)
+        found = any(function_id in _feature_values(row.get(cat, ""), cat) for cat in categories)
         if not found:
             continue
         sample = row.get("Sample", "")
@@ -151,6 +157,8 @@ def query_taxon(gene_table: str | Path, taxon_query: str, output: str | Path,
     path = Path(gene_table)
     if not path.is_file():
         raise FileNotFoundError(f"Gene annotation table not found: {path}")
+    if top_n < 1:
+        raise ValueError("top_n must be a positive integer.")
     category_key = category.upper().replace("-", "_")
     aliases = {key.upper(): key for key in FEATURE_COLUMNS}
     if category_key not in aliases:
@@ -167,7 +175,7 @@ def query_taxon(gene_table: str | Path, taxon_query: str, output: str | Path,
             continue
         sample = row.get("Sample", "")
         support = int(float(row.get("Gene_Read_Support", 0) or 0))
-        for feature in _feature_values(row.get(category_name, "")):
+        for feature in _feature_values(row.get(category_name, ""), category_name):
             counts[(sample, value, lineage, feature)][0] += 1
             counts[(sample, value, lineage, feature)][1] += support
     per_sample: dict[str, list[dict[str, Any]]] = defaultdict(list)

{metapont-0.1.2 → metapont-0.1.4}/src/MetaPont/cli.py

@@ -10,7 +10,7 @@ from pathlib import Path
 from . import MetaPont_Version, __version__
 from .analysis import coverage_summary, query_function, query_taxon
 from .layout import discover_samples
-from .pipeline import OUTPUT_DIRNAME, run_analysis, write_inventory
+from .pipeline import OUTPUT_DIRNAME, build_taxon_function_abundance, run_analysis, write_inventory
 
 
 def _add_root_prefix(parser: argparse.ArgumentParser) -> None:
@@ -22,11 +22,19 @@ def _add_root_prefix(parser: argparse.ArgumentParser) -> None:
                         help=f"Results directory (default: ROOT/{OUTPUT_DIRNAME}).")
 
 
-def _gene_table(root_dir: str, output_dir: str | None, supplied: str | None) -> Path:
+def _output_root(root_dir: str | None, output_dir: str | None) -> Path:
+    if output_dir:
+        return Path(output_dir).expanduser().resolve()
+    if not root_dir:
+        raise ValueError("Provide --root-dir, --output-dir or --gene-table.")
+    root = Path(root_dir).expanduser().resolve()
+    return root if root.name == OUTPUT_DIRNAME else root / OUTPUT_DIRNAME
+
+
+def _gene_table(root_dir: str | None, output_dir: str | None, supplied: str | None) -> Path:
     if supplied:
         return Path(supplied).expanduser().resolve()
-    output_root = Path(output_dir).expanduser().resolve() if output_dir else Path(root_dir).expanduser().resolve() / OUTPUT_DIRNAME
-    return output_root / "combined" / "gene_annotations.tsv.gz"
+    return _output_root(root_dir, output_dir) / "combined" / "gene_annotations.tsv.gz"
 
 
 def _create_plots(output_root: Path) -> list[str]:
@@ -91,20 +99,24 @@ def main(argv: list[str] | None = None) -> int:
     run_p = sub.add_parser("run", aliases=["combine"], help="Build the integrated sample inventory, contig table and abundance summaries.")
     _add_root_prefix(run_p)
     run_p.add_argument("--force", action="store_true", help="Regenerate outputs even when the recorded input signature is unchanged.")
-    run_p.add_argument("--no-heal-flags", action="store_true", help="Do not recreate missing pipeline flags when all corresponding outputs exist.")
+    run_p.add_argument("--no-heal-flags", action="store_true", help="Do not recreate missing upstream flags, even when the full expected stage output set is present.")
     run_p.add_argument("--min-contig-length", type=int, default=2500, help="Minimum contig length to retain an existing Kraken2 assignment; classifications are not rerun.")
     run_p.add_argument("--plots", action="store_true", help="Generate optional quick-look PNG plots (requires matplotlib).")
+    run_p.add_argument(
+        "--signature-mode", choices=("sampled", "full"), default="sampled",
+        help="Input change detection: sampled hashes seven evenly spaced blocks (touch-safe, default), or full SHA256 of every input byte (slower).",
+    )
     run_p.add_argument(
         "--taxon-function-abundance",
         action="store_true",
         help=(
-            "Generate the expanded taxon-function abundance table. "
-            "Can require substantial memory and disk space."
+            "Generate the expanded taxon-function abundance table using a disk-backed aggregation. "
+            "This requires additional disk I/O and temporary space."
         ),
     )
     inv_p = sub.add_parser("inventory", help="Audit per-sample inputs, pipeline outputs and success flags.")
     _add_root_prefix(inv_p)
-    inv_p.add_argument("--heal-flags", action="store_true", help="Recreate missing flags only when the relevant output set is complete and current.")
+    inv_p.add_argument("--heal-flags", action="store_true", help="Recreate missing upstream flags only when the full expected stage output set is present; timestamps are not used.")
 
     cov_p = sub.add_parser("coverage", help="Summarise per-contig readmapping statistics by sample.")
     _add_root_prefix(cov_p)
@@ -114,8 +126,9 @@ def main(argv: list[str] | None = None) -> int:
     qf_p = sub.add_parser("query-function", help="Find taxa supporting a particular eggNOG functional ID.")
     qf_p.add_argument("--root-dir", "-d", default=None,
                       help="Workflow root used to find the default gene table; not needed with --gene-table or --output-dir.")
+    qf_p.add_argument("--gene-table", default=None,
+                      help="Gene annotation TSV/TSV.GZ; defaults to the integrated output table.")
     qf_p.add_argument("--output-dir", default=None)
-    qf_p.add_argument("--gene-table", default=None, help="Gene annotation TSV/TSV.GZ; defaults to the integrated output table.")
     qf_p.add_argument("--function-id", "-f", required=True, help="Exact function ID, e.g. GO:0016597 or EC:2.7.11.1.")
     qf_p.add_argument("--category", "-c", default=None, help="Optional annotation category, e.g. GO, EC, KEGG_KO, PFAM.")
     qf_p.add_argument("--rank", default="Genus", help="Taxonomic rank for attribution (default: Genus).")
@@ -123,15 +136,25 @@ def main(argv: list[str] | None = None) -> int:
     qf_p.add_argument("--output", "-o", required=True)
 
     qt_p = sub.add_parser("query-taxon", help="Report the most supported functions for a taxon.")
-    qt_p.add_argument("--root-dir", "-d", required=True)
+    qt_p.add_argument("--root-dir", "-d", default=None)
+    qt_p.add_argument("--gene-table", default=None,
+                      help="Gene annotation TSV/TSV.GZ; defaults to the integrated output table.")
     qt_p.add_argument("--output-dir", default=None)
-    qt_p.add_argument("--gene-table", default=None)
     qt_p.add_argument("--taxon", "-t", required=True, help="Taxon name or lineage token, e.g. Bacillus or g__Bacillus.")
     qt_p.add_argument("--category", "-c", required=True, help="Annotation category, e.g. GO, EC, KEGG_KO.")
     qt_p.add_argument("--rank", default="Genus")
     qt_p.add_argument("--top", type=int, default=20)
     qt_p.add_argument("--output", "-o", required=True)
 
+    tf_p = sub.add_parser(
+        "taxon-function-abundance", aliases=["taxon-function"],
+        help="Build the expanded taxon-by-function table from an existing gene annotation table without rerunning the main analysis.",
+    )
+    tf_p.add_argument("--root-dir", "-d", default=None, help="Workflow root or MetaPont_Results directory.")
+    tf_p.add_argument("--gene-table", default=None, help="Existing gene_annotations.tsv or .tsv.gz file.")
+    tf_p.add_argument("--output-dir", default=None, help="MetaPont results directory (default derived from --root-dir).")
+    tf_p.add_argument("--output", "-o", default=None, help="Output TSV path (default: OUTPUT_DIR/functional/taxon_function_abundance.tsv).")
+
     args = parser.parse_args(argv)
     try:
         if args.command in {"run", "combine"}:
@@ -141,6 +164,7 @@ def main(argv: list[str] | None = None) -> int:
                 heal_flags=not args.no_heal_flags,
                 force=args.force,
                 include_taxon_function_abundance=args.taxon_function_abundance,
+                signature_mode=args.signature_mode,
             )
             if args.plots:
                 output_root = Path(result["output_dir"])
@@ -160,19 +184,58 @@ def main(argv: list[str] | None = None) -> int:
                 print(f"Healed {len(events)} flags.")
             return 0
         if args.command == "coverage":
-            out, count = coverage_summary(args.root_dir, args.prefix, args.output, args.read_length)
+            supplied_root = Path(args.root_dir).expanduser().resolve()
+            samples_root = supplied_root.parent if supplied_root.name == OUTPUT_DIRNAME else supplied_root
+            default_output_root = (
+                Path(args.output_dir).expanduser().resolve() if args.output_dir
+                else supplied_root if supplied_root.name == OUTPUT_DIRNAME
+                else supplied_root / OUTPUT_DIRNAME
+            )
+            default_output = default_output_root / "summary" / "contig_coverage_summary.tsv"
+            out, count = coverage_summary(
+                samples_root, args.prefix, args.output or default_output, args.read_length,
+            )
             print(f"Coverage summary written for {count} samples: {out}")
             return 0
         if args.command == "query-function":
+            if args.root_dir is None and args.gene_table is None and args.output_dir is None:
+                raise ValueError("Provide --root-dir, --output-dir or --gene-table.")
+
+            # A MetaPont_Results directory can be supplied directly as root-dir.
+            if args.root_dir and Path(args.root_dir).expanduser().resolve().name == OUTPUT_DIRNAME and not args.output_dir:
+                args.output_dir = args.root_dir
+
             table = _gene_table(args.root_dir, args.output_dir, args.gene_table)
             count = query_function(table, args.function_id, args.output, args.category, args.rank, args.top)
             print(f"Wrote {count} contributing taxon rows to {args.output}")
             return 0
+
         if args.command == "query-taxon":
+            if args.root_dir is None and args.gene_table is None and args.output_dir is None:
+                raise ValueError("Provide --root-dir, --output-dir or --gene-table.")
+
+            # A MetaPont_Results directory can be supplied directly as root-dir.
+            if args.root_dir and Path(args.root_dir).expanduser().resolve().name == OUTPUT_DIRNAME and not args.output_dir:
+                args.output_dir = args.root_dir
+
             table = _gene_table(args.root_dir, args.output_dir, args.gene_table)
             count = query_taxon(table, args.taxon, args.output, args.category, args.top, args.rank)
+
             print(f"Wrote {count} taxon-associated function rows to {args.output}")
             return 0
+        if args.command in {"taxon-function-abundance", "taxon-function"}:
+            if args.gene_table is None and args.root_dir is None and args.output_dir is None:
+                raise ValueError("Provide --gene-table, --root-dir or --output-dir.")
+            table = _gene_table(args.root_dir, args.output_dir, args.gene_table)
+            if args.output:
+                output = Path(args.output).expanduser().resolve()
+            elif args.output_dir or args.root_dir:
+                output = _output_root(args.root_dir, args.output_dir) / "functional" / "taxon_function_abundance.tsv"
+            else:
+                output = table.parent.parent / "functional" / "taxon_function_abundance.tsv"
+            count = build_taxon_function_abundance(table, output)
+            print(f"Wrote {count} taxon/function rows to {output}")
+            return 0
     except (FileNotFoundError, RuntimeError, ValueError, OSError) as error:
         print(f"ERROR: {error}", file=sys.stderr)
         return 2

{metapont-0.1.2 → metapont-0.1.4}/src/MetaPont/compat.py

@@ -186,7 +186,7 @@ def function_by_taxa_main() -> int:
     root = args.root_dir or args.directory
     if not root:
         parser.error("provide the sample root as a positional argument or with -d")
-    result = run_analysis(root, args.prefix, args.output_dir, heal_flags=True, force=False)
+    result = run_analysis(root, args.prefix, args.output_dir, heal_flags=True, force=False, include_taxon_function_abundance=True)
     print(f"Function-by-taxon results written to {result['output_dir']}/functional/taxon_function_abundance.tsv")
     return 0