From e6bef3d102066a0acf4a9d7974ac28dd8352e249 Mon Sep 17 00:00:00 2001
From: Alex Rubinsteyn <alex.rubinsteyn@gmail.com>
Date: Fri, 29 Apr 2016 12:29:35 -0400
Subject: [PATCH] added sanity checking to imputation helpers
---
mhcflurry/amino_acid.py | 7 +++++-
.../class1_allele_specific_hyperparameters.py | 2 +-
mhcflurry/data.py | 2 +-
mhcflurry/feedforward.py | 2 +-
mhcflurry/imputation.py | 23 ++++++++++++++++++-
mhcflurry/paths.py | 7 +++++-
mhcflurry/peptide_encoding.py | 5 ++++
mhcflurry/predictor_base.py | 5 ++++
8 files changed, 47 insertions(+), 6 deletions(-)
diff --git a/mhcflurry/amino_acid.py b/mhcflurry/amino_acid.py
index f4022561..9c06fada 100644
--- a/mhcflurry/amino_acid.py
+++ b/mhcflurry/amino_acid.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2015. Mount Sinai School of Medicine
+# Copyright (c) 2016. Mount Sinai School of Medicine
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -12,6 +12,11 @@
# See the License for the specific language governing permissions and
# limitations under the License.
+from __future__ import (
+ print_function,
+ division,
+ absolute_import,
+)
import numpy as np
diff --git a/mhcflurry/class1_allele_specific_hyperparameters.py b/mhcflurry/class1_allele_specific_hyperparameters.py
index 33e15cdb..cfdf846b 100644
--- a/mhcflurry/class1_allele_specific_hyperparameters.py
+++ b/mhcflurry/class1_allele_specific_hyperparameters.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2015. Mount Sinai School of Medicine
+# Copyright (c) 2016. Mount Sinai School of Medicine
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
diff --git a/mhcflurry/data.py b/mhcflurry/data.py
index 1ea71212..59c2be51 100644
--- a/mhcflurry/data.py
+++ b/mhcflurry/data.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2015. Mount Sinai School of Medicine
+# Copyright (c) 2016. Mount Sinai School of Medicine
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
diff --git a/mhcflurry/feedforward.py b/mhcflurry/feedforward.py
index e401b4b8..e526d6df 100644
--- a/mhcflurry/feedforward.py
+++ b/mhcflurry/feedforward.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2015. Mount Sinai School of Medicine
+# Copyright (c) 2016. Mount Sinai School of Medicine
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
diff --git a/mhcflurry/imputation.py b/mhcflurry/imputation.py
index 1bcb08dc..cddeb5e3 100644
--- a/mhcflurry/imputation.py
+++ b/mhcflurry/imputation.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2015. Mount Sinai School of Medicine
+# Copyright (c) 2016. Mount Sinai School of Medicine
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -18,6 +18,8 @@ from __future__ import (
absolute_import,
)
from collections import defaultdict
+import logging
+
import numpy as np
from fancyimpute.dictionary_helpers import (
dense_matrix_from_nested_dictionary
@@ -34,6 +36,21 @@ from .data import (
create_allele_data_from_peptide_to_ic50_dict,
)
+def _check_dense_pMHC_array(X, peptide_list, allele_list):
+ if len(peptide_list) != len(set(peptide_list)):
+ raise ValueError("Duplicate peptides detected in peptide list")
+ if len(allele_list) != len(set(allele_list)):
+ raise ValueError("Duplicate alleles detected in allele list")
+ n_rows, n_cols = X.shape
+ if n_rows != len(peptide_list):
+ raise ValueError(
+ "Expected dense array with shape %s to have %d rows" % (
+ X.shape, len(peptide_list)))
+ if n_cols != len(allele_list):
+ raise ValueError(
+ "Expected dense array with shape %s to have %d columns" % (
+ X.shape, len(allele_list)))
+
def prune_dense_matrix_and_labels(
X,
peptide_list,
@@ -89,6 +106,7 @@ def prune_dense_matrix_and_labels(
X = X[:, keep_allele_indices]
observed_mask = observed_mask[:, keep_allele_indices]
allele_list = [allele_list[i] for i in keep_allele_indices]
+ _check_dense_pMHC_array(X, peptide_list, allele_list)
return X, peptide_list, allele_list
@@ -135,6 +153,8 @@ def create_incomplete_dense_pMHC_matrix(
X, peptide_list, allele_list = \
dense_matrix_from_nested_dictionary(peptide_to_allele_to_affinity_dict)
+ _check_dense_pMHC_array(X, peptide_list, allele_list)
+
return prune_dense_matrix_and_labels(
X,
peptide_list,
@@ -187,6 +207,7 @@ def create_imputed_datasets(
# if all entries in the matrix are already filled in then don't
# try using an imputation algorithm since it might raise an
# exception.
+ logging.warn("No missing values, using original data instead of imputation")
X_complete = X_incomplete
else:
X_complete = imputer.complete(X_incomplete)
diff --git a/mhcflurry/paths.py b/mhcflurry/paths.py
index ec9e8377..1ef41eb9 100644
--- a/mhcflurry/paths.py
+++ b/mhcflurry/paths.py
@@ -1,4 +1,4 @@
-# Copyright (c) 2015. Mount Sinai School of Medicine
+# Copyright (c) 2016. Mount Sinai School of Medicine
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
@@ -12,6 +12,11 @@
# See the License for the specific language governing permissions and
# limitations under the License.
+from __future__ import (
+ print_function,
+ division,
+ absolute_import,
+)
from os.path import join
from appdirs import user_data_dir
diff --git a/mhcflurry/peptide_encoding.py b/mhcflurry/peptide_encoding.py
index 45045680..cad5f53c 100644
--- a/mhcflurry/peptide_encoding.py
+++ b/mhcflurry/peptide_encoding.py
@@ -12,6 +12,11 @@
# See the License for the specific language governing permissions and
# limitations under the License.
+from __future__ import (
+ print_function,
+ division,
+ absolute_import,
+)
import itertools
import logging
diff --git a/mhcflurry/predictor_base.py b/mhcflurry/predictor_base.py
index dba570da..a335158c 100644
--- a/mhcflurry/predictor_base.py
+++ b/mhcflurry/predictor_base.py
@@ -12,6 +12,11 @@
# See the License for the specific language governing permissions and
# limitations under the License.
+from __future__ import (
+ print_function,
+ division,
+ absolute_import,
+)
from collections import defaultdict
import numpy as np
--
GitLab