PCA implementation #95

Author: eric-czech

.pre-commit-config.yaml

@@ -26,4 +26,4 @@ repos:
     hooks:
       - id: mypy
         args: ["--strict", "--show-error-codes"]
-        additional_dependencies: ["numpy", "xarray", "dask[array]", "scipy", "typing-extensions", "zarr", "numba"]
+        additional_dependencies: ["numpy", "xarray", "dask[array]", "scipy", "typing-extensions", "zarr", "numba", "dask-ml"]

requirements-dev.txt

@@ -6,6 +6,7 @@ pytest-cov
 pytest-datadir
 pytest-mock
 hypothesis
+scikit-allel
 statsmodels
 zarr
 msprime
@@ -18,3 +19,4 @@ git+https://github.com/pangeo-data/rechunker.git
 cbgen
 cyvcf2; platform_system != "Windows"
 yarl
+matplotlib

requirements.txt

@@ -1,6 +1,7 @@
 numpy
 xarray
 dask[array]
+dask-ml
 scipy
 typing-extensions
 numba

setup.cfg

@@ -28,6 +28,7 @@ install_requires =
     numpy
     xarray
     dask[array]
+    dask-ml
     scipy
     zarr
     numba
@@ -87,15 +88,14 @@ ignore =
 profile = black
 default_section = THIRDPARTY
 known_first_party = sgkit
-known_third_party = dask,fire,glow,hail,hypothesis,invoke,msprime,numba,numpy,pandas,pkg_resources,pyspark,pytest,setuptools,sgkit_plink,sklearn,sphinx,typing_extensions,xarray,yaml,zarr
+known_third_party = allel,dask,fire,glow,hail,hypothesis,invoke,msprime,numba,numpy,pandas,pkg_resources,pyspark,pytest,setuptools,sgkit_plink,sklearn,sphinx,typing_extensions,xarray,yaml,zarr
 multi_line_output = 3
 include_trailing_comma = True
 force_grid_wrap = 0
 use_parentheses = True
 line_length = 88
 
-[mypy-allel.*]
-ignore_missing_imports = True
+
 [mypy-callee.*]
 ignore_missing_imports = True
 [mypy-cyvcf2.*]
@@ -104,6 +104,8 @@ ignore_missing_imports = True
 ignore_missing_imports = True
 [mypy-fsspec.*]
 ignore_missing_imports = True
+[mypy-dask_ml.*]
+ignore_missing_imports = True
 [mypy-numpy.*]
 ignore_missing_imports = True
 [mypy-pandas.*]
@@ -132,6 +134,8 @@ ignore_missing_imports = True
 ignore_missing_imports = True
 [mypy-yarl.*]
 ignore_missing_imports = True
+[mypy-allel.*]
+ignore_missing_imports = True
 [mypy-sgkit.*]
 allow_redefinition = True
 [mypy-sgkit.*.tests.*]

sgkit/__init__.py

@@ -12,6 +12,7 @@
 from .stats.association import gwas_linear_regression
 from .stats.hwe import hardy_weinberg_test
 from .stats.pc_relate import pc_relate
+from .stats.pca import pca
 from .stats.popgen import Fst, Tajimas_D, divergence, diversity
 from .stats.regenie import regenie
 from .testing import simulate_genotype_call_dataset
@@ -38,4 +39,5 @@
     "pc_relate",
     "simulate_genotype_call_dataset",
     "variables",
+    "pca",
 ]