feat(CARMA): adding time limited CARMA (#475)

* feat(CARMA): adding time limited CARMA

* feat: correct the function

* feat(CARMA): added tests

* chore: small fix of varaible types

* chore: fix of test descriptions

src/gentropy/method/carma.py

@@ -1,6 +1,7 @@
 """CARMA outlier detection method."""
 from __future__ import annotations
 
+import concurrent.futures
 from itertools import combinations
 from math import floor, lgamma
 from typing import Any
@@ -14,6 +15,34 @@
 class CARMA:
     """Implementation of CARMA outlier detection method."""
 
+    @staticmethod
+    def time_limited_CARMA_spike_slab_noEM(
+        z: np.ndarray, ld: np.ndarray, sec_threshold: float = 600
+    ) -> dict[str, Any]:
+        """The wrapper for the CARMA_spike_slab_noEM function that runs the function in a separate thread and terminates it if it takes too long.
+
+        Args:
+            z (np.ndarray): Numeric vector representing z-scores.
+            ld (np.ndarray): Numeric matrix representing the linkage disequilibrium (LD) matrix.
+            sec_threshold (float): The time threshold in seconds.
+
+        Returns:
+            dict[str, Any]: A dictionary containing the following results:
+                - PIPs: A numeric vector of posterior inclusion probabilities (PIPs) for all SNPs or None.
+                - B_list: A dataframe containing the marginal likelihoods and the corresponding model space or None.
+                - Outliers: A list of outlier SNPs or None.
+        """
+        try:
+            # Execute CARMA.CARMA_spike_slab_noEM with a timeout
+            with concurrent.futures.ThreadPoolExecutor(max_workers=1) as executor:
+                future = executor.submit(CARMA.CARMA_spike_slab_noEM, z=z, ld=ld)
+                result = future.result(timeout=sec_threshold)
+        except concurrent.futures.TimeoutError:
+            # If execution exceeds the timeout, return None
+            result = {"PIPs": None, "B_list": None, "Outliers": None}
+
+        return result
+
     @staticmethod
     def CARMA_spike_slab_noEM(
         z: np.ndarray,
@@ -105,7 +134,6 @@ def CARMA_spike_slab_noEM(
             "Outliers": all_C_list["conditional_S_list"],
         }
 
-
         return results_list
 
     @staticmethod

tests/gentropy/method/test_carma.py

@@ -54,3 +54,22 @@ def test_MCS_modified(
             num_causal=10,
         )
         assert np.allclose(np.round(np.corrcoef(l1_pips, pips)[0, 1], decimals=2), 1)
+
+    def test_time_limited_CARMA_spike_slab_noEM_pips_no_restriction(
+        self: TestCARMA, sample_data_for_carma: list[np.ndarray]
+    ) -> None:
+        """Test of CARMA PIPs with liberal (no) time restriction."""
+        ld = sample_data_for_carma[0]
+        z = sample_data_for_carma[1]
+        pips = sample_data_for_carma[2]
+        _l = CARMA.time_limited_CARMA_spike_slab_noEM(z=z, ld=ld, sec_threshold=600)
+        assert np.allclose(np.round(np.corrcoef(_l["PIPs"], pips)[0, 1], decimals=2), 1)
+
+    def test_time_limited_CARMA_spike_slab_noEM_pips_restriction(
+        self: TestCARMA, sample_data_for_carma: list[np.ndarray]
+    ) -> None:
+        """Test of CARMA PIPs with time restriction."""
+        ld = sample_data_for_carma[0]
+        z = sample_data_for_carma[1]
+        _l = CARMA.time_limited_CARMA_spike_slab_noEM(z=z, ld=ld, sec_threshold=0.001)
+        assert _l["Outliers"] is None and _l["PIPs"] is None