Change default value of min_len to extsize in tl.macs3.

.github/workflows/test_python.yml

@@ -21,7 +21,7 @@ jobs:
     - name: Install dependencies
       run: |
         sudo pip install --upgrade pip
-        pip install --user coverage pytest hypothesis wheel
+        pip install --user coverage pytest hypothesis==6.72.4 wheel
 
     - name: Build wheel files
       run: |

.github/workflows/wheels.yml

@@ -31,7 +31,7 @@ jobs:
        CIBW_ENVIRONMENT: 'PATH="$PATH:$HOME/.cargo/bin"'
        CIBW_SKIP: "pp* *-win32 *-musllinux*"
        CIBW_BUILD: ${{ matrix.python_version }}
-       CIBW_TEST_REQUIRES: pytest hypothesis
+       CIBW_TEST_REQUIRES: pytest hypothesis==6.72.4
        CIBW_TEST_COMMAND: "pytest {project}/snapatac2-python/tests"
 
     steps:

docs/changelog.md

@@ -11,6 +11,8 @@
 - Fix: #221: 'pp.knn' with 'method=pynndescent' invalid csr matrix.
 - Fix: #226: Backed AnnData does not support numpy string array. 
 - Fix: #232: `tempdir` is not respected in `tl.macs3`.
+- Fix: #242: Change default value of `min_len` to `extsize` in `tl.macs3`, in order to be
+  consistent with the `macs3` command line tool.
 - Other minor bug fixes.
 
 ## Release 2.5.3 (released Jan 16, 2024)

snapatac2-python/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "snapatac2"
-version = "2.5.4-dev3"
+version = "2.5.4-dev4"
 edition = "2021"
 authors = ["Kai Zhang <[email protected]>"]
 description = "Rust APIs"

snapatac2-python/python/snapatac2/tools/_call_peaks.py

@@ -18,6 +18,7 @@ def macs3(
     nolambda: bool = False,
     shift: int = -100,
     extsize: int = 200,
+    min_len: int | None = None,
     blacklist: Path | None = None,
     key_added: str = 'macs3',
     tempdir: Path | None = None,
@@ -60,6 +61,8 @@ def macs3(
         The shift size in MACS.
     extsize
         The extension size in MACS.
+    min_len
+        The minimum length of a called peak. If None, it is set to `extsize`.
     blacklist
         Path to the blacklist file in BED format. If provided, regions in the blacklist will be
         removed.
@@ -110,7 +113,7 @@ def macs3(
 
     options.gsize = adata.uns['reference_sequences']['reference_seq_length'].sum()    # Estimated genome size
     options.maxgap = 30    # The maximum allowed gap between two nearby regions to be merged
-    options.minlen = 50    # The minimum length of a called peak
+    options.minlen = extsize if min_len is None else min_len
     options.shift = shift
     options.nolambda = nolambda
     options.smalllocal = 1000

snapatac2-python/src/call_peaks.rs

@@ -258,7 +258,9 @@ pub fn create_fwtrack_obj<'py>(
     let replicates = files
         .into_iter()
         .map(|fl| {
-            let fwt = macs.getattr("FWTrack")?.call1((100000,))?;
+            let kwargs = pyo3::types::PyDict::new(py);
+            kwargs.set_item("buffer_size", 100000)?;
+            let fwt = macs.getattr("FWTrack")?.call((), Some(kwargs))?;
             let reader = utils::open_file_for_read(&fl);
             bed_utils::bed::io::Reader::new(reader, None)
                 .into_records::<Fragment>()