diff options
-rw-r--r-- | gn3/computations/slink.py | 75 | ||||
-rw-r--r-- | tests/unit/computations/test_slink.py | 174 |
2 files changed, 249 insertions, 0 deletions
diff --git a/gn3/computations/slink.py b/gn3/computations/slink.py new file mode 100644 index 0000000..7b4016a --- /dev/null +++ b/gn3/computations/slink.py @@ -0,0 +1,75 @@ +class LengthError(BaseException): + pass + +class MirrorError(BaseException): + pass + +def raise_valueerror_if_data_is_not_lists_or_tuples(lists): + """Check that `lists` is a list of lists: If not, raise an exception.""" + def is_list_or_tuple(item): + return type(item) == type([]) or type(item) == type(tuple) + + if (not is_list_or_tuple(lists)) or (not all(map(is_list_or_tuple, lists))): + raise ValueError("Expected list or tuple") + +def raise_valueerror_if_lists_empty(lists): + """Check that the list and its direct children are not empty.""" + def empty(lst): + return len(lst) == 0 + if (empty(lists)) or not all(map(lambda x: not empty(x), lists)): + raise ValueError("List/Tuple should NOT be empty!") + +def raise_lengtherror_if_child_lists_are_not_same_as_parent(lists): + def len_is_same_as_parent(lst): + return len(lst) == len(lists) + if not all(map(len_is_same_as_parent, lists)): + raise LengthError("All children lists should be same length as the parent.") + +def raise_valueerror_if_child_list_distance_from_itself_is_not_zero(lists): + def get_child_distance(child): + idx = lists.index(child) + return lists[idx][idx] + def distance_is_zero(dist): + return dist == 0 + children_distances = map(get_child_distance, lists) + if not all(map(distance_is_zero, children_distances)): + raise ValueError("Distance of each child list/tuple from itself should be zero!") + +def raise_mirrorerror_of_distances_one_way_are_not_same_other_way(lists): + """Check that the distance from A to B, is the same as the distance from B to A. +If the two distances are different, throw an exception.""" + for i in range(len(lists)): + for j in range(len(lists)): + if lists[i][j] != lists[j][i]: + raise MirrorError( + ("Distance from one child({}) to the other ({}) " + "should be the same in both directions.").format( + lists[i][j], lists[j][i])) + +def raise_valueerror_on_negative_distances(lists): + """Check that distances between 'somethings' are all positive, otherwise, +raise an exception.""" + def zero_or_positive(val): + return val >= 0; + # flatten lists + flattened = [distance for child in lists for distance in child] + if not all(map(zero_or_positive, flattened)): + raise ValueError("Distances should be positive.") + +def nearest(lists, i, j): + """Computes some form of distance. +This is 'copied' over from genenetwork1, from https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/slink.py#L42-L64. + +This description should be updated once the form/type of 'distance' identified.""" + + #### Guard Functions: Should we do this a different way? #### + raise_valueerror_if_data_is_not_lists_or_tuples(lists) + raise_valueerror_if_lists_empty(lists) + raise_lengtherror_if_child_lists_are_not_same_as_parent(lists) + raise_valueerror_if_child_list_distance_from_itself_is_not_zero(lists) + raise_mirrorerror_of_distances_one_way_are_not_same_other_way(lists) + raise_valueerror_on_negative_distances(lists) + #### END: Guard Functions #### + return None + if type(i) == int and type(j) == int: # From member i to member j + return lists[i][j] diff --git a/tests/unit/computations/test_slink.py b/tests/unit/computations/test_slink.py new file mode 100644 index 0000000..166eff9 --- /dev/null +++ b/tests/unit/computations/test_slink.py @@ -0,0 +1,174 @@ +"""Module contains tests for slink""" +import unittest +from unittest import TestCase + +from gn3.computations.slink import nearest +from gn3.computations.slink import LengthError +from gn3.computations.slink import MirrorError + +class TestSlink(TestCase): + """Class for testing slink functions""" + + def test_nearest_expects_list_of_lists(self): + # This might be better handled with type-hints and mypy + for item in [9, "some string", 5.432, + [1,2,3], ["test", 7.4]]: + with self.subTest(item=item): + with self.assertRaises(ValueError, msg="Expected list or tuple"): + nearest(item, 1, 1) + + def test_nearest_does_not_allow_empty_lists(self): + for lst in [[], + [[],[]], + [[],[],[]], + [[0, 1, 2],[],[1, 2, 0]]]: + with self.subTest(lst=lst): + with self.assertRaises(ValueError): + nearest(lst, 1, 1) + + def test_nearest_expects_exception_if_all_child_lists_are_not_of_equal_length_to_length_of_parent_list(self): + for lst in [[[0,1]], + [[0,1,2],[3,4,5]], + [[0,1,2,3],[4,5,6],[7,8,9,0]], + [[0,1,2,3,4],[5,6,7,8,9],[1,2,3,4,5],[2,3],[3,4,5,6,7]]]: + with self.subTest(lst=lst): + with self.assertRaises(LengthError): + nearest(lst, 1, 1) + + def test_nearest_expects_exception_if_distance_of_child_from_itself_is_not_zero(self): + for lst in [[[1]], + [[1,2],[3,4]], + [1,0,0],[0,0,5],[0,3,4], + [0,0,0,0],[0,0,3,3],[0,1,2,3],[0,3,2,0]]: + with self.subTest(lst=lst): + with self.assertRaises(ValueError): + nearest(lst, 1, 1) + + def test_nearest_expects_exception_if_distance_from_child_a_to_child_b_is_not_distance_from_child_b_to_child_a(self): + for lst in [[[0,1],[2,0]], + [[0,1,2],[1,0,3],[9,7,0]], + [[0,1,2,3],[7,0,2,3],[2,3,0,1],[8,9,5,0]]]: + with self.subTest(lst=lst): + with self.assertRaises(MirrorError): + nearest(lst, 1, 1) + + def test_nearest_expects_zero_or_positive_distances(self): + # Based on: + # https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/slink.py#L87-L89 + for lst in [[[0,-1,2,3],[-1,0,3,4],[2,3,0,5],[3,4,5,0]], + [[0,1,-2,3],[1,0,3,4],[-2,3,0,5],[3,4,5,0]], + [[0,1,2,3],[1,0,-3,4],[2,-3,0,5],[3,4,5,0]], + [[0,1,2,-3],[1,0,3,4],[2,3,0,5],[-3,4,5,0]], + [[0,1,2,3],[1,0,3,-4],[2,3,0,5],[3,-4,5,0]], + [[0,1,2,3],[1,0,3,4],[2,3,0,-5],[3,4,-5,0]]]: + with self.subTest(lst=lst): + with self.assertRaises(ValueError, msg="Distances should be positive."): + nearest(lst, 1, 1) + + def test_nearest_returns_shortest_distance_given_coordinates_to_both_group_members(self): + # This test is named wrong - at least I think it is, from the expected results + # This tests distance when both `i`, and `j` are integers + # We still need to add tests for when (either one/both) (is/are) not (an) integer(s) + # https://github.com/genenetwork/genenetwork1/blob/master/web/webqtl/heatmap/slink.py#L39-L40 + for lst, i, j, expected in [[[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 0,0,0], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 0,1,9], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 0,2,3], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 0,3,6], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 0,4,11], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 1,0,9], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 1,1,0], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 1,2,7], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 1,3,5], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 1,4,10], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 2,0,3], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 2,1,7], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 2,2,0], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 2,3,9], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 2,4,2], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 3,0,6], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 3,1,5], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 3,2,9], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 3,3,0], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 3,4,8], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 4,0,11], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 4,1,10], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 4,2,2], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 4,3,8], + [[[0,9,3,6,11],[9,0,7,5,10],[3,7,0,9,2],[6,5,9,0,8],[11,10,2,8,0]], + 4,4,0], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 0,0,0], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 0,1,9], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 0,2,5.5], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 0,3,6], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 0,4,11], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 1,0,9], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 1,1,0], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 1,2,7], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 1,3,5], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 1,4,10], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 2,0,5.5], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 2,1,7], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 2,2,0], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 2,3,9], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 2,4,2], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 3,0,6], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 3,1,5], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 3,2,9], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 3,3,0], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 3,4,3], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 4,0,11], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 4,1,10], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 4,2,2], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 4,3,3], + [[[0,9,5.5,6,11],[9,0,7,5,10],[5.5,7,0,9,2],[6,5,9,0,3],[11,10,2,3,0]], + 4,4,0]]: + with self.subTest(lst=lst): + self.assertEqual(nearest(lst, i, j), expected) |