def sam():
"""Render the landing page for SAM file analysis."""
if 'id' in session and len(list_dir('sam', session['id'], ('.bam'))) > 0:
# At least one SAM file is uploaded, so show extended menu
return render_template('sam.html', showHidden=True)
return render_template('sam.html', showHidden=False)
def sam_upload():
"""Render the file manager for handling SAM and FASTA files.
Methods:
GET: Render the File Management Page with form.
POST: Process the form upload and redirect user.
"""
# Quick shortcut method so I don't have to rewrite this code
def render_form(session_id, samfiles=None, fastafile=None):
"""Render the form associated with this route."""
return render_template('sam/upload.html',
session_id=escape(session_id),
samfiles=basename(samfiles, session_id),
fastafile=basename(fastafile, session_id))
# Generate a new session ID if it doesn't already exist
if 'id' not in session:
session['id'] = random_id()
# Get all files associated with this session id
oldbams = list_dir('sam', session['id'], ('.bam'))
oldfasta = get_firstfile('sam', session['id'], ('.fa', '.fasta'))
# Render an empty form for a GET request
if request.method == 'GET':
return render_form(session['id'], oldbams, oldfasta)
# Otherwise validate the form on a POST request
if request.method == 'POST':
sams = request.files.getlist('sam')
fasta = request.files['fasta']
# Throw error if entire form is empty
if not fasta and not sams[0]:
flash('No files were selected for upload.', 'alert-warning')
return render_form(session['id'], oldbams, oldfasta)
# Throw error if any file does not have the appropriate extension
abort = False
for sam in sams:
if len(sam.filename) > 0:
ext = os.path.splitext(secure_filename(sam.filename))[1]
if ext.lower() not in ['.bam', '.sam']:
abort = True
if fasta:
ext = os.path.splitext(secure_filename(fasta.filename))[1]
if ext.lower() not in ['.fa', '.fasta']:
abort = True
if abort:
flash('Invalid file types were uploaded', 'alert-warning')
return render_form(session['id'], oldbams, oldfasta)
# Upload SAM files
if sams:
error_count = 0
for sam in sams:
if len(sam.filename) > 0:
try:
upload_sam(sam, sam.filename, escape(session['id']))
except IOError, message:
flash(message, 'alert-warning')
error_count = error_count + 1
if len(sams) == error_count:
return render_form(session['id'], oldbams, oldfasta)
# Upload Fasta file; first delete old fasta files if they exist
if fasta:
delete_files('sam', session['id'], None, ('.fa', '.fasta'))
fa = upload(fasta, fasta.filename, 'sam', escape(session['id']))
try:
Fasta(fa)
except:
flash(message, 'alert-warning')
if not sams:
return render_form(session['id'], oldbams, oldfasta)
flash('Files were successfully uploaded.', 'alert-success')
return render_template('sam.html', showHidden=True)
def sam_browse():
"""Render the sam browser.
Methods:
GET: Render the SAM browser page with form.
POST: Process the form upload display browser results.
"""
# Redirect user if there are no bam files uploaded
if not session['id']:
return redirect(url_for('sam_upload'), code=302)
bampaths = list_dir('sam', session['id'], ('.bam'))
fastapath = get_firstfile('sam', session['id'], ('.fa', '.fasta'))
if len(bampaths) == 0:
return redirect(url_for('sam_upload'), code=302)
# Load samfiles and get list of genes in samfile
bamfiles = load_sams(bampaths, 'rb')
gene_list = get_all_genes(bamfiles)
# Load Fasta file and validate it
badFasta = True
if fastapath:
try:
fasta = Fasta(fastapath)
if fasta.validate(gene_list):
badFasta = False
gene_list = sorted(fasta.genes.keys())
except:
badFasta = True
# Render an empty form for a GET request
if request.method == 'GET':
# Construct table where (rows, columns) = (genes, samples)
gene_rows = []
for gene in gene_list:
row = []
for bam in bamfiles:
if gene in bam.references:
row.append(True)
else:
row.append(False)
gene_rows.append(row)
# Render an empty form for a GET request
return render_template('sam/browse.html',
session_id=escape(session['id']),
samfiles=basename(bampaths, session['id']),
fastafile=basename(fastapath, session['id']),
badFasta=badFasta,
genes=gene_list,
rows=gene_rows)
# Process form POST
if request.method == 'POST':
# Load common variables
form = {
'mode': request.form.get('mode'),
'gene': request.form.get('gene'),
'start': request.form.get('start'),
'end': request.form.get('end'),
'custom': request.form.get('custom'),
'output': request.form.get('output'),
}
# Validate and parse form
try:
parsed_form = _validate_and_parse(form)
except ((KeyError, ValueError), message):
flash(str(message), 'alert-warning')
return redirect(url_for('sam_browse'), code=302)
# Begin main loop
i = 0
error_count = 0
results = []
for bam in bamfiles:
# Get coordinate positions of the selected gene as a list
try:
positions = _get_positions(bam, parsed_form)
except (ValueError, message):
error_count = error_count + 1
continue
# Make the universal position container
positionContainer = OrderedDict({parsed_form['gene']: positions})
# Handle the fasta file if present
if badFasta:
fasta_seq = None
cpg_positionContainer = None
else:
# Get reference sequence from fasta file
fasta_seq = _fasta_fetch_positions(fasta, positionContainer)
# Get coordinates of CpG sites from fasta file
cpg_positionContainer = _fasta_fetch_cpg(
fasta, positionContainer)
# Get the read bases at the requested coordinated positions
readDict = _bam_fetch_positions(bam, positionContainer,
cpg_positionContainer)
results.append({
'sample':
basename(bampaths, session['id'])[i],
'positions':
_flatten_position_container(positionContainer),
'cpg_positions':
_flatten_position_container(cpg_positionContainer),
'reads':
readDict,
'fasta':
fasta_seq
})
i = i + 1
# Abort if all the bamfiles errored
if len(bamfiles) == error_count:
flash('Coordinates not found in sample(s)', 'alert-warning')
return redirect(url_for('sam_browse'), code=302)
# Determine output method - HTML
if parsed_form['output'] == 'html':
genes, positions = _split_and_add1(results[0]['positions'])
return render_template('sam/browse_results.html',
sample=results[0]['sample'],
gene=genes[0],
positions=positions,
reads=results[0]['reads'],
fasta=results[0]['fasta'])
# Output method - CSV
if parsed_form['output'] == 'csv':
if len(bamfiles) == 1: # Single CSV file, send back a .csv file
data = _format_csv_file(results[0])
filename = os.path.splitext(results[0]['sample'])[0] + '.csv'
return csv_response('result_' + filename, data)
else: # Multiple CSV files, send back a .zip file
data = {}
for result in results:
csvdata = _format_csv_file(result)
filename = os.path.splitext(result['sample'])[0] + '.csv'
data[filename] = csvdata
return zip_csv_response('results.zip', data)
# Output method - Excel
if len(bamfiles) == 1: # Single Excel file, send back a .xlsx file
workbook = _format_exl_file(results[0])
filename = os.path.splitext(results[0]['sample'])[0] + '.xlsx'
return exl_response('result_' + filename, workbook)
else: # Multiple Excel files, send back a .zip file
data = {}
for result in results:
exl_data = _format_exl_file(result)
filename = os.path.splitext(result['sample'])[0] + '.xlsx'
data[filename] = exl_data
return zip_exl_response('results.zip', data)
def sam_browse():
"""Render the sam browser.
Methods:
GET: Render the SAM browser page with form.
POST: Process the form upload display browser results.
"""
# Redirect user if there are no bam files uploaded
if not session['id']:
return redirect(url_for('sam_upload'), code=302)
bampaths = list_dir('sam', session['id'], ('.bam'))
fastapath = get_firstfile('sam', session['id'], ('.fa', '.fasta'))
if len(bampaths) == 0:
return redirect(url_for('sam_upload'), code=302)
# Load samfiles and get list of genes in samfile
bamfiles = load_sams(bampaths, 'rb')
gene_list = get_all_genes(bamfiles)
# Load Fasta file and validate it
badFasta = True
if fastapath:
try:
fasta = Fasta(fastapath)
if fasta.validate(gene_list):
badFasta = False
gene_list = sorted(fasta.genes.keys())
except:
badFasta = True
# Render an empty form for a GET request
if request.method == 'GET':
# Construct table where (rows, columns) = (genes, samples)
gene_rows = []
for gene in gene_list:
row = []
for bam in bamfiles:
if gene in bam.references:
row.append(True)
else:
row.append(False)
gene_rows.append(row)
# Render an empty form for a GET request
return render_template('sam/browse.html',
session_id=escape(session['id']),
samfiles=basename(bampaths, session['id']),
fastafile=basename(fastapath, session['id']),
badFasta=badFasta,
genes=gene_list,
rows=gene_rows)
# Process form POST
if request.method == 'POST':
# Load common variables
form = {
'mode': request.form.get('mode'),
'gene': request.form.get('gene'),
'start': request.form.get('start'),
'end': request.form.get('end'),
'custom': request.form.get('custom'),
'output': request.form.get('output'),
}
# Validate and parse form
try:
parsed_form = _validate_and_parse(form)
except (KeyError, ValueError), message:
flash(str(message), 'alert-warning')
return redirect(url_for('sam_browse'), code=302)
# Begin main loop
i = 0
error_count = 0
results = []
for bam in bamfiles:
# Get coordinate positions of the selected gene as a list
try:
positions = _get_positions(bam, parsed_form)
except ValueError, message:
error_count = error_count + 1
continue
# Make the universal position container
positionContainer = OrderedDict({parsed_form['gene']: positions})
# Get the read bases at the requested coordinated positions
readDict = _bam_fetch_positions(bam, positionContainer)
# Handle the fasta file if present
if badFasta:
fasta_seq = None
else:
fasta_seq = _fasta_fetch_positions(fasta, positionContainer)
results.append({
'sample':
basename(bampaths, session['id'])[i],
'positions':
_flatten_position_container(positionContainer),
'reads':
readDict,
'fasta':
fasta_seq
})
i = i + 1