fix: keep plasmid copy-number column float-typed (fixes --no_chromosome)

The earlier collate_depths hardening set the copy-number column to the string
"NA" when chromosome depth was missing or zero. In --no_chromosome mode a fake
chromosome of A's (zero depth) is created, so the column became "NA" and
combine_depth_mash_tsvs's `.astype(float)` raised
`ValueError: could not convert string to float: 'NA'` -- the two CI failures
(test_plassembler_run_no_chromosome / _long_no_chromosome).

Keep the column float-typed: dividing by a zero chromosome depth yields inf
(which the depth filter treats as "keep", matching prior behaviour), and NaN is
emitted only when the chromosome depth is genuinely unusable. Regression tests
updated, including one that reproduces the .astype(float) failure.

Validated: both --no_chromosome end-to-end tests pass with the real toolchain.

Author: gbouras13

src/plassembler/utils/depth.py

@@ -148,15 +148,20 @@ def collate_depths(depths, shortFlag, contig_lengths):
         )
         mean_col, copy_col = "mean_depth_long", "plasmid_copy_number_long"
 
-    # plasmid copy number = mean depth / chromosome mean depth. Guard against a
-    # missing (no "chromosome" contig), non-numeric or zero chromosome depth,
-    # which previously raised NameError/TypeError or produced inf copy numbers.
-    if not isinstance(chromosome_depth, (int, float)) or chromosome_depth == 0:
-        summary_df[copy_col] = "NA"
+    # plasmid copy number = mean depth / chromosome mean depth.
+    if not isinstance(chromosome_depth, (int, float)):
+        # no usable chromosome depth: either no "chromosome" contig at all
+        # (chromosome_depth is None) or its stats could not be computed
+        # ("NA"). Copy number is undefined; keep the column float-typed.
+        summary_df[copy_col] = float("nan")
     else:
-        summary_df[copy_col] = (
-            pd.to_numeric(summary_df[mean_col], errors="coerce") / chromosome_depth
-        ).round(2)
+        # chromosome_depth may be 0 (e.g. the fake --no_chromosome chromosome of
+        # A's, which no reads map to); dividing then yields inf, which the
+        # downstream depth filter treats as "keep" - preserving prior behaviour.
+        # A non-numeric ("NA") mean is coerced to NaN so the column stays float
+        # (a string here broke .astype(float) in combine_depth_mash_tsvs).
+        mean_numeric = pd.to_numeric(summary_df[mean_col], errors="coerce")
+        summary_df[copy_col] = (mean_numeric / chromosome_depth).round(2)
     # return df
     return summary_df
 

tests/test_deterministic.py

@@ -7,6 +7,8 @@
 
 from pathlib import Path
 
+import numpy as np
+
 from src.plassembler.utils.depth import (
     collate_depths,
     get_contig_circularity,
@@ -54,17 +56,23 @@ def test_collate_depths_copy_number_math():
     assert df.loc["plasmid00001", "plasmid_copy_number_short"] == 2.0
 
 
-def test_collate_depths_no_chromosome_returns_na():
-    """No contig named 'chromosome' -> copy number is 'NA' (previously NameError)."""
+def test_collate_depths_no_chromosome_is_nan_not_crash():
+    """No contig named 'chromosome' -> copy number NaN, not a NameError/crash."""
     depths = {"plasmid00001": [20] * 60}
     contig_lengths = {"plasmid00001": 60}
     df = collate_depths(depths, "short", contig_lengths)
-    assert df["plasmid_copy_number_short"].tolist() == ["NA"]
+    assert df["plasmid_copy_number_short"].isna().all()
+    # must stay float-convertible (downstream does .astype(float))
+    df["plasmid_copy_number_short"].astype(float)
 
 
-def test_collate_depths_zero_chromosome_depth_returns_na():
-    """Zero chromosome depth -> copy number is 'NA' (previously inf)."""
+def test_collate_depths_zero_chromosome_depth_stays_float():
+    """Zero chromosome depth (e.g. the fake --no_chromosome chromosome) -> inf,
+    which the depth filter treats as 'keep'. Regression: a string 'NA' here broke
+    the downstream .astype(float) in combine_depth_mash_tsvs (CI --no_chromosome)."""
     depths = {"chromosome": [0] * 100, "plasmid00001": [5] * 60}
     contig_lengths = {"chromosome": 100, "plasmid00001": 60}
-    df = collate_depths(depths, "short", contig_lengths)
-    assert (df["plasmid_copy_number_short"] == "NA").all()
+    df = collate_depths(depths, "short", contig_lengths).set_index("contig")
+    assert np.isinf(df.loc["plasmid00001", "plasmid_copy_number_short"])
+    # the actual CI failure: this must not raise
+    df["plasmid_copy_number_short"].astype(float)