DOC: causal_sites

Add documentation for causal_sites input

docs/effect-size.md

@@ -296,3 +296,22 @@ plt.show()
 
 We see that rarer variants have increased effect sizes, as expected from
 the mathematical expression given in [](eq:freq-dep).
+
+(trait_causal_sites)=
+
+## Specifying Causal Sites
+
+Instead of specifying the number of causal sites, users can directly provide the
+causal sites as an input of {py:func}`sim_phenotype` and {py:func}`sim_trait`.
+When `causal_sites` argument of these functions are provided, tstrait will use
+the inputted site IDs as causal sites, instead of randomly selecting causal
+site IDs.
+
+Example:
+
+```{code-cell}
+
+trait_df = tstrait.sim_trait(ts, causal_sites=[0, 3, 4], model=model, 
+                             alpha=-1/2, random_seed=1)
+trait_df
+```

tstrait/base.py

@@ -61,12 +61,3 @@ def _check_non_decreasing(array, array_name):
         diff = np.diff(array)
         if np.min(diff) < 0:
             raise ValueError(f"{array_name} must be non-decreasing")
-
-
-def _check_no_repeat(array, array_name):
-    """Check that there are no repeated values in the array."""
-    array = np.sort(array)
-    if len(array) > 1:
-        diff = np.diff(array)
-        if np.min(diff) == 0:
-            raise ValueError(f"There must not be repeated values in {array_name}")

tstrait/simulate_effect_size.py

@@ -1,4 +1,3 @@
-import numbers
 from dataclasses import dataclass
 
 import numpy as np