def test_construct_pileup_list(self):
"""
test_construct_pileup_list - Checks that the pileup length and the
first few indices of the pileup are correct.
INPUT:
[None]
RETURN:
[None]
POST:
[None]
"""
bamPileup = Pileup_List.construct_pileup_list(self.test_cp_files,
self.references)
pileup_as_array = bamPileup.get_pileups_as_array()
pileup_as_numerical_array = bamPileup.get_pileups_as_numerical_array()
assert len(pileup_as_array) == 2
assert len(pileup_as_array[0]) == 2844
assert len(pileup_as_array[1]) == 2844
assert len(pileup_as_numerical_array[0]) == (2844 * 4)
assert len(pileup_as_numerical_array[1]) == (2844 * 4)
assert pileup_as_array[0][0:10] == [{
'C': 12
}, {
'C': 12
}, {
'T': 12
}, {
'C': 12
}, {
'G': 2,
'C': 3,
'T': 1,
'A': 6
}, {
'G': 12
}, {
'G': 12
}, {
'T': 12
}, {
'C': 12
}, {
'G': 2,
'C': 3,
'T': 1,
'A': 6
}]
assert pileup_as_array[1][0:10] == [{
'C': 12
}, {
'C': 12
}, {
'T': 12
}, {
'C': 12
}, {
'A': 6,
'C': 5,
'G': 1
}, {
'G': 12
}, {
'A': 4,
'G': 8
}, {
'T': 12
}, {
'C': 12
}, {
'A': 7,
'T': 1,
'C': 3,
'G': 1
}]
def dist(ctx, reference, bam, normalize, output_distance, startpos, endpos,
output, no_coverage):
"""
dist - Performs the main part of the program
INPUT:
[CONTEXT] [ctx]
[FASTA FILE LOCATION] [reference]
[BAM FILE LOCATION] [bam]
[BOOL] [normalize/dont_normalize]
[BOOL] [output_distance/output_similarity]
[INT] [startpos]
[INT] [endpos]
[STRING] [output]
Output the CSV-formatted matrix output in a file
instead of in the terminal.
[STRING] [truncate/remove_no_coverage/keep_no_coverage]
Options to truncate low-coverage regions on the ends of the pileup,
ignore all low coverage regions, or keep all low coverage regions
RETURN:
None.
POST:
The distance matrix is printed out unless an error message was raised.
"""
if len(bam) < 2:
raise click.UsageError("At least two bam file locations are required" +
" to perform quasispecies distance comparison")
# indicate if the start or end position is < 0 or a priori invalid
if type(startpos) == int and int(startpos) < 1:
raise click.UsageError("Start position must be >= 1.")
if type(endpos) == int and int(endpos) < 1:
raise click.UsageError("End position must be >= 1.")
if (type(startpos) == int and type(endpos) == int and (startpos > endpos)):
raise click.UsageError("Start position must be <= end position")
# Build the reference object.
references = parse_references_from_fasta(reference)
pileups = Pileup_List.construct_pileup_list(bam, references)
if startpos is None:
startpos = 1
if endpos is None:
endpos = pileups.get_pileup_length()
if pileups.get_pileup_length() == 0:
raise click.UsageError("Empty pileup was produced from BAM files." +
"Halting program")
click.echo("The start position is %d." % startpos)
click.echo("The end position is %d." % endpos)
click.echo("Constructed pileup from reference.")
# click.echo the number of positions in pileup
click.echo("The pileup covers %d positions before modifications." %
pileups.get_pileup_length())
# indicate whether the user-specified start and end position is out
# of bounds (comparing to actual number of positions in pileup)
if startpos > pileups.get_pileup_length():
raise click.UsageError("Start position must be less than or" +
" equal to the number of nucleotide base " +
"positions in pileup (%s)." %
pileups.get_pileup_length())
if endpos > pileups.get_pileup_length():
raise click.UsageError("End position must be less than or equal to " +
"the number of nucleotide base positions in " +
"pileup (%s)." % pileups.get_pileup_length())
# we convert the start and end positions from one-based indexing to
# zero-based indexing which is expected by distance.py and pileup.py
startpos -= 1
endpos -= 1
# if there is no errors so far, proceed with running program
modified = modify_pileups(ctx, normalize, startpos, endpos, no_coverage,
pileups)
if (no_coverage is not 'keep_no_coverage') and (len(modified) == 0):
raise click.UsageError("Entire pileup was truncated due to " +
"lack of coverage. Halting program")
dist = DistanceMatrix(modified, bam)
if output_distance:
click.echo("Outputting an angular cosine distance matrix.")
if output:
output.write(dist.get_distance_matrix_as_csv())
else:
click.echo(dist.get_distance_matrix_as_csv())
else:
click.echo("Outputting a cosine similarity matrix.")
if output:
output.write(dist.get_similarity_matrix_as_csv())
else:
click.echo(dist.get_similarity_matrix_as_csv())