Skip to content

Commit 61f8a90

Browse files
authored
Merge branch 'master' into GH947-fst-zscore-zero-se
2 parents c69e565 + 06e52c5 commit 61f8a90

2 files changed

Lines changed: 10 additions & 0 deletions

File tree

malariagen_data/anoph/snp_data.py

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,4 @@
1+
import warnings
12
from functools import lru_cache
23
from typing import Any, Dict, List, Optional, Tuple, Union
34

@@ -1253,6 +1254,12 @@ def _snp_calls(
12531254
if max_cohort_size is not None:
12541255
n_samples = ds.sizes["samples"]
12551256
if n_samples > max_cohort_size:
1257+
warnings.warn(
1258+
f"Cohort downsampled from {n_samples} to {max_cohort_size} "
1259+
"samples. Set max_cohort_size=None to disable downsampling.",
1260+
UserWarning,
1261+
stacklevel=2,
1262+
)
12561263
rng = np.random.default_rng(seed=random_seed)
12571264
loc_downsample = rng.choice(
12581265
n_samples, size=max_cohort_size, replace=False

tests/anoph/test_dipclust.py

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -11,6 +11,9 @@ def random_transcripts_contig(*, api, contig, n):
1111
df_gff = api.genome_features(attributes=["ID", "Parent"])
1212
df_transcripts = df_gff.query(f"type == 'mRNA' and contig == '{contig}'")
1313
transcript_ids = df_transcripts["ID"].dropna().to_list()
14+
n = min(n, len(transcript_ids))
15+
if n == 0:
16+
pytest.skip(f"No mRNA transcripts found for contig '{contig}'")
1417
transcripts = random.sample(transcript_ids, n)
1518
return transcripts
1619

0 commit comments

Comments
 (0)