1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350
|
import numpy.testing as npt
import numpy as np
cimport numpy as np
from bp._bp cimport BP, mM
fig1_B = np.array([1, 1, 1, 0, 1, 0, 1, 1 ,0, 0, 0, 1, 0, 1, 1, 1, 0, 1, 0,
0, 0, 0], dtype=np.uint8)
def get_test_obj():
return BP(fig1_B)
def test_rank():
cdef BP obj = get_test_obj()
counts_1 = fig1_B.cumsum()
counts_0 = (1 - fig1_B).cumsum()
for exp, t in zip((counts_1, counts_0), (1, 0)):
for idx, e in enumerate(exp):
npt.assert_equal(obj.rank(t, idx), e)
def test_select():
cdef BP obj = get_test_obj()
pos_1 = np.unique(fig1_B.cumsum(), return_index=True)[1] #- 1
pos_0 = np.unique((1 - fig1_B).cumsum(), return_index=True)[1]
for exp, t in zip((pos_1, pos_0), (1, 0)):
for k in range(1, len(exp)):
npt.assert_equal(obj.select(t, k), exp[k])
def test_rank_property():
cdef BP obj = get_test_obj()
for i in range(len(fig1_B)):
npt.assert_equal(obj.rank(1, i) + obj.rank(0, i), i+1)
def test_rank_select_property():
cdef BP obj = get_test_obj()
pos_1 = np.unique(fig1_B.cumsum(), return_index=True)[1] #- 1
pos_0 = np.unique((1 - fig1_B).cumsum(), return_index=True)[1]
for t, pos in zip((0, 1), (pos_0, pos_1)):
for k in range(len(pos)):
# needed +t on expectation, unclear at this time why.
npt.assert_equal(obj.rank(t, obj.select(t, k)), k + t)
def test_excess():
cdef BP obj = get_test_obj()
# from fig 2
exp = [1, 2, 3, 2, 3, 2, 3, 4, 3, 2, 1, 2, 1, 2, 3, 4, 3, 4, 3, 2, 1, 0]
for idx, e in enumerate(exp):
npt.assert_equal(obj.excess(idx), e)
def test_depth():
cdef BP obj = get_test_obj()
# from fig 2
exp = [1, 2, 3, 2, 3, 2, 3, 4, 3, 2, 1, 2, 1, 2, 3, 4, 3, 4, 3, 2, 1, 0]
for idx, e in enumerate(exp):
npt.assert_equal(obj.depth(idx), e)
def test_close():
cdef BP obj = get_test_obj()
exp = [21, 10, 3, 5, 9, 8, 12, 20, 19, 16, 18]
for i, e in zip(np.argwhere(fig1_B == 1).squeeze(), exp):
npt.assert_equal(obj.close(i), e)
npt.assert_equal(obj.excess(obj.close(i)), obj.excess(i) - 1)
def test_open():
cdef BP obj = get_test_obj()
exp = [2, 4, 7, 6, 1, 11, 15, 17, 14, 13, 0]
for i, e in zip(np.argwhere(fig1_B == 0).squeeze(), exp):
npt.assert_equal(obj.open(i), e)
npt.assert_equal(obj.excess(obj.open(i)) - 1,
obj.excess(i))
def test_enclose():
cdef BP obj = get_test_obj()
# i > 0 and i < (len(B) - 1)
exp = [0, 1, 1, 1, 1, 1, 6, 6, 1, 0, 0, 0, 0, 13, 14, 14, 14, 14, 13, 0]
for i, e in zip(range(1, len(fig1_B) - 1), exp):
npt.assert_equal(obj.enclose(i), e)
def test_parent():
cdef BP obj = get_test_obj()
exp = [-1, 0, 1, 1, 1, 1, 1, 6, 6, 1, 0, 0, 0, 0, 13, 14, 14, 14, 14, 13,
0, -1]
for i, e in zip(range(len(fig1_B)), exp):
npt.assert_equal(obj.parent(i), e)
def test_root():
cdef BP obj = get_test_obj()
npt.assert_equal(obj.root(), 0)
def test_isleaf():
cdef BP obj = get_test_obj()
exp = [0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0]
for i, e in enumerate(exp):
npt.assert_equal(obj.isleaf(i), e)
def test_fchild():
cdef BP obj = get_test_obj()
exp = [1, 2, 0, 0, 0, 0, 7, 0, 0, 7, 2, 0, 0, 14, 15, 0, 0, 0, 0, 15, 14,
1]
for i, e in enumerate(exp):
npt.assert_equal(obj.fchild(i), e)
def test_lchild():
cdef BP obj = get_test_obj()
exp = [obj.preorderselect(7),
obj.preorderselect(4),
0,
0,
0,
0,
obj.preorderselect(5),
0,
0,
obj.preorderselect(5),
obj.preorderselect(4),
0,
0,
obj.preorderselect(8),
obj.preorderselect(10),
0,
0,
0,
0,
obj.preorderselect(10),
obj.preorderselect(8),
obj.preorderselect(7)]
for i, e in enumerate(exp):
npt.assert_equal(obj.lchild(i), e)
def test_nsibling():
cdef BP obj = get_test_obj()
exp = [0, 11, 4, 4, 6, 6, 0, 0, 0, 0, 11, 13, 13, 0, 0, 17, 17, 0, 0, 0, 0,
0]
for i, e in enumerate(exp):
npt.assert_equal(obj.nsibling(i), e)
def test_psibling():
cdef BP obj = get_test_obj()
exp = [0, 0, 0, 0, 2, 2, 4, 0, 0, 4, 0, 1, 1, 11, 0, 0, 0, 15, 15, 0, 11,
0]
for i, e in enumerate(exp):
npt.assert_equal(obj.psibling(i), e)
def test_fwdsearch():
cdef BP obj = get_test_obj()
exp = {(0, 0): 10, # close of first child
(3, -2): 21, # close of root
(11, 2): 15} # from one tip to the next
for (i, d), e in exp.items():
npt.assert_equal(obj.fwdsearch(i, d), e)
def test_bwdsearch():
cdef BP obj = get_test_obj()
exp = {(3, 0): 1, # open of parent
(21, 4): 17, # nested tip
(9, 2): 7} # open of the node
for (i, d), e in exp.items():
npt.assert_equal(obj.bwdsearch(i, d), e)
def test_fwdsearch_more():
cdef BP bp
from bp import parse_newick
bp = parse_newick('((a,b,(c)),d,((e,f)));')
# simulating close so only testing open parentheses. A "close" on a closed
# parenthesis does not make sense, so the result is not useful.
# In practice, an "close" method should ensure it is operating on a closed
# parenthesis.
# [(open_idx, close_idx), ...]
exp = [(1, 10), (0, 21), (2, 3), (4, 5), (6, 9), (7, 8), (11, 12),
(13, 20), (14, 19), (15, 16), (17, 18)]
for open_, exp_close in exp:
obs_close = bp.fwdsearch(open_, -1)
assert obs_close == exp_close
# slightly modified version of fig2 with an extra child forcing a test
# of the direct sibling check with negative partial excess
# this translates into:
# 012345678901234567890123
# ((()()(()))()((()()())))
bp = parse_newick('((a,b,(c)),d,((e,f,g)));')
#enmM = rmm(bp.B, bp.B.size)
# simulating close so only testing open parentheses. A "close" on a closed
# parenthesis does not make sense, so the result is not useful.
# In practice, an "close" method should ensure it is operating on a closed
# parenthesis.
# [(open_idx, close_idx), ...]
exp = [(0, 23), (1, 10), (2, 3), (4, 5), (6, 9), (7, 8), (11, 12),
(13, 22), (14, 21), (15, 16), (17, 18), (19, 20)]
for open_, exp_close in exp:
obs_close = bp.fwdsearch(open_, -1)
assert obs_close == exp_close
def test_bwdsearch_more():
cdef BP bp
from bp import parse_newick
bp = parse_newick('((a,b,(c)),d,((e,f)));')
# simulating open so only testing closed parentheses.
# [(close_idx, open_idx), ...]
exp = [(21, 0), (8, 7), (9, 6), (10, 1), (3, 2), (5, 4), (12, 11),
(16, 15), (20, 13), (19, 14), (18, 17)]
for close_, exp_open in exp:
obs_open = bp.bwdsearch(close_, 0) + 1
assert obs_open == exp_open
# slightly modified version of fig2 with an extra child forcing a test
# of the direct sibling check with negative partial excess
# this translates into:
# 012345678901234567890123
# ((()()(()))()((()()())))
bp = parse_newick('((a,b,(c)),d,((e,f,g)));')
# simulating open so only testing closed parentheses.
# [(close_idx, open_idx), ...]
exp = [(23, 0), (10, 1), (3, 2), (5, 4), (9, 6), (8, 7), (12, 11),
(22, 13), (21, 14), (16, 15), (18, 17), (20, 19)]
for close_, exp_open in exp:
obs_open = bp.bwdsearch(close_, 0) + 1
assert obs_open == exp_open
def test_scan_block_forward():
cdef BP bp
from bp import parse_newick
bp = parse_newick('((a,b,(c)),d,((e,f)));')
# [(open, close), ...]
b = 4
d = -1
exp_b_4 = [(0, ((0, -1), (1, -1), (2, 3), (3, -1))),
(1, ((4, 5), (5, -1), (6, -1), (7, -1))),
# 8 and 9 are nonsensical from finding a "close" perspective
(2, ((8, 9), (9, 10), (10, -1), (11, -1))),
(3, ((12, -1), (13, -1), (14, -1), (15, -1))),
# 16 and 18 are nonsensical from a "close" perspective
(4, ((16, 19), (17, 18), (18, 19), (19, -1))),
# 20 is nonsensical from finding a "close" perspective
(5, ((20, 21), (21, -1)))]
for k, exp_results in exp_b_4:
for idx, exp_result in exp_results:
obs_result = bp.scan_block_forward(idx, k, b, bp.excess(idx) + d)
assert obs_result == exp_result
b = 8
exp_b_8 = [(0, ((0, -1), (1, -1), (2, 3), (3, -1),
(4, 5), (5, -1), (6, -1), (7, -1))),
(1, ((8, 9), (9, 10), (10, -1), (11, 12),
(12, -1), (13, -1), (14, -1), (15, -1))),
(2, ((16, 19), (17, 18), (18, 19), (19, 20),
(20, 21), (21, -1)))]
for k, exp_results in exp_b_8:
for idx, exp_result in exp_results:
obs_result = bp.scan_block_forward(idx, k, b, bp.excess(idx) + d)
assert obs_result == exp_result
def test_scan_block_backward():
cdef BP bp
from bp import parse_newick
bp = parse_newick('((a,b,(c)),d,((e,f)));')
# adding +1 to simluate "open" so calls on open parentheses are weird
# [(open, close), ...]
b = 4
d = 0
exp_b_4 = [(0, ((0, 0), (1, 0), (2, 0), (3, 2))),
(1, ((4, 0), (5, 4), (6, 5), (7, 0))),
(2, ((8, 0), (9, 0), (10, 0), (11, 10))),
(3, ((12, 0), (13, 12), (14, 0), (15, 0))),
(4, ((16, 0), (17, 16), (18, 17), (19, 0))),
(5, ((20, 0), (21, 0)))]
for k, exp_results in exp_b_4:
for idx, exp_result in exp_results:
obs_result = bp.scan_block_backward(idx, k, b, bp.excess(idx) + d)
obs_result += 1 # simulating open
assert obs_result == exp_result
b = 8
exp_b_8 = [(0, ((0, 0), (1, 0), (2, 0), (3, 2),
(4, 3), (5, 4), (6, 5), (7, 0))),
(1, ((8, 0), (9, 0), (10, 0), (11, 10),
(12, 11), (13, 12), (14, 9), (15, 8))),
(2, ((16, 0), (17, 16), (18, 17), (19, 0),
(20, 0), (21, 0)))]
for k, exp_results in exp_b_8:
for idx, exp_result in exp_results:
obs_result = bp.scan_block_backward(idx, k, b, bp.excess(idx) + d)
obs_result += 1 # simulating open
assert obs_result == exp_result
def test_rmm():
cdef BP bp
from bp import parse_newick
# test tree is ((a,b,(c)),d,((e,f)));
# this is from fig 2 of Cordova and Navarro:
# http://www.dcc.uchile.cl/~gnavarro/ps/tcs16.2.pdf
bp = parse_newick('((a,b,(c)),d,((e,f)));')
exp = np.array([[0, 1, 0, 1, 1, 0, 0, 1, 2, 1, 1, 2, 0], # m
[4, 4, 4, 4, 4, 4, 0, 3, 4, 3, 4, 4, 1]], # M
dtype=np.intp).T
obs = mM(bp.B, bp.B.size)
# original r / k0 values, preserving for posterity
# [0, 0, 10, 0, 6, 10, 0, 0, 3, 6, 7, 10, 11], # r
# [11, 6, 11, 2, 6, 11, 0, 1, 2, 5, 6, 9, 11]], # k0
assert exp.shape[0] == obs.mM.shape[0]
assert exp.shape[1] == obs.mM.shape[1]
for i in range(exp.shape[0]):
for j in range(exp.shape[1]):
assert obs.mM[i, j] == exp[i, j]
|