Merge pull request #51 from martinghunt/remove_numpy

Remove numpy

pyfastaq/common.py

@@ -1 +1 @@
-version = '3.9.0'
+version = '3.10.0'

pyfastaq/tasks.py

@@ -2,7 +2,6 @@
 import sys
 import copy
 import random
-import numpy
 from pyfastaq import sequences, utils, caf
 
 class Error (Exception): pass
@@ -404,62 +403,6 @@ def make_random_contigs(contigs, length, outfile, name_by_letters=False, prefix=
     utils.close(fout)
 
 
-def make_long_reads(infile, outfile, method='tiling', fixed_read_length=20000, tile_step=10000, gamma_shape=1.2,  gamma_scale=6000, coverage=10, gamma_min_length=20000, seed=None, ins_skip=None, ins_window=None,):
-    assert method in ['tiling', 'gamma', 'uniform']
-    assert ins_skip == ins_window == None or None not in [ins_skip, ins_window]
-    if seed is not None:
-        random.seed(a=seed)
-    seq_reader = sequences.file_reader(infile)
-    f = utils.open_file_write(outfile)
-
-    for seq in seq_reader:
-        if method == 'tiling':
-            if len(seq) < fixed_read_length:
-                print('Skipping sequence', seq.id, 'because it is too short at', len(seq), 'bases', file=sys.stderr)
-                continue
-            for i in range(0, len(seq), tile_step):
-                end = min(len(seq), i + fixed_read_length)
-                fa = sequences.Fasta('_'.join([seq.id, str(i + 1), str(end)]), seq[i:end])
-                if ins_skip:
-                    fa.add_insertions(skip=ins_skip, window=ins_window)
-                print(fa, file=f)
-                if end >= len(seq):
-                    break
-        elif method == 'gamma':
-            if len(seq) < gamma_min_length:
-                print('Skipping sequence', seq.id, 'because it is too short at', len(seq), 'bases', file=sys.stderr)
-                continue
-            total_read_length = 0
-            while total_read_length < coverage * len(seq) - 0.5 * gamma_min_length:
-                read_length = int(numpy.random.gamma(gamma_shape, scale=gamma_scale))
-                while read_length < gamma_min_length or read_length > len(seq):
-                    read_length = int(numpy.random.gamma(gamma_shape, scale=gamma_scale))
-
-                start = random.randint(0, len(seq) - read_length)
-                end = start + read_length - 1
-                fa = sequences.Fasta('_'.join([seq.id, str(start + 1), str(end + 1)]), seq[start:end+1])
-                total_read_length += len(fa)
-                if ins_skip:
-                    fa.add_insertions(skip=ins_skip, window=ins_window)
-                print(fa, file=f)
-        elif method == 'uniform':
-            if len(seq) < fixed_read_length:
-                print('Skipping sequence', seq.id, 'because it is too short at', len(seq), 'bases', file=sys.stderr)
-                continue
-            total_read_length = 0
-            while total_read_length < coverage * len(seq) - 0.5 * fixed_read_length:
-                start = random.randint(0, len(seq) - fixed_read_length)
-                end = start + fixed_read_length - 1
-                fa = sequences.Fasta('_'.join([seq.id, str(start + 1), str(end + 1)]), seq[start:end+1])
-                total_read_length += len(fa)
-                if ins_skip:
-                    fa.add_insertions(skip=ins_skip, window=ins_window)
-                print(fa, file=f)
-
-
-    utils.close(f)
-
-
 def mean_length(infile, limit=None):
     '''Returns the mean length of the sequences in the input file. By default uses all sequences. To limit to the first N sequences, use limit=N'''
     total = 0

pyfastaq/tests/tasks_test.py

@@ -271,15 +271,6 @@ def files_are_equal(file1, file2):
         os.unlink(tmp)
 
 
-class TestMakeLongReads(unittest.TestCase):
-    def test_tiling_reads(self):
-        tmp = 'tmp.out.fa'
-        fa_in = os.path.join(data_dir, 'tasks_test_make_long_reads.input.fa')
-        tasks.make_long_reads(fa_in, tmp, method='tiling', fixed_read_length=10, tile_step=5)
-        self.assertTrue(filecmp.cmp(os.path.join(data_dir, 'tasks_test_make_long_reads.output.fa'), tmp, shallow=False))
-        os.unlink(tmp)
-
-
 class TestMeanLength(unittest.TestCase):
     def test_mean_length(self):
         '''Test mean_length'''

setup.py

@@ -4,7 +4,7 @@
 
 setup(
     name='pyfastaq',
-    version='3.9.0',
+    version='3.10.0',
     description='Script to manipulate FASTA and FASTQ files, plus API for developers',
     packages = find_packages(),
     author='Martin Hunt',
@@ -13,7 +13,6 @@
     scripts=glob.glob('scripts/*'),
     test_suite='nose.collector',
     tests_require=['nose >= 1.3'],
-    install_requires=['numpy >= 1.7.1'],
     license='GPLv3',
     classifiers=[
         'Development Status :: 4 - Beta',