samples/views.py

   1 from __future__ import absolute_import, print_function, unicode_literals
   2
   3 # Create your views here.
   4 import logging
   5 import os
   6 import sys
   7
   8 import json
   9
  10 from django.views.decorators.csrf import csrf_exempt
  11 from django.core.exceptions import ObjectDoesNotExist
  12 from django.http import HttpResponse, HttpResponseRedirect, Http404
  13 from django.shortcuts import render_to_response, get_object_or_404
  14 from django.template import RequestContext
  15 from django.template.loader import get_template
  16 from django.contrib.auth.decorators import login_required
  17 from django.conf import settings
  18
  19 from htsworkflow.auth import require_api_key
  20 from experiments.models import FlowCell, Lane, LANE_STATUS_MAP
  21 from experiments.admin import LaneOptions
  22 from .changelist import HTSChangeList
  23 from .models import Antibody, Library, Species, HTSUser
  24 from .admin import LibraryOptions
  25 from .results import get_flowcell_result_dict
  26 from bcmagic.forms import BarcodeMagicForm
  27 from htsworkflow.pipelines import runfolder
  28 from htsworkflow.pipelines.eland import ResultLane
  29 from htsworkflow.pipelines.samplekey import SampleKey
  30 from htsworkflow.util.conversion import str_or_none, parse_flowcell_id
  31 from htsworkflow.util import makebed
  32 from htsworkflow.util import opener
  33
  34
  35 LANE_LIST = [1, 2, 3, 4, 5, 6, 7, 8]
  36 SAMPLES_CONTEXT_DEFAULTS = {
  37     'app_name': 'Flowcell/Library Tracker',
  38     'bcmagic': BarcodeMagicForm()
  39 }
  40
  41 LOGGER = logging.getLogger(__name__)
  42
  43
  44 def count_lanes(lane_set):
  45     single = 0
  46     paired = 1
  47     short_read = 0
  48     medium_read = 1
  49     long_read = 2
  50     counts = [[0, 0, 0], [0, 0, 0]]
  51
  52     for lane in lane_set.all():
  53         if lane.flowcell.paired_end:
  54             lane_type = paired
  55         else:
  56             lane_type = single
  57         if lane.flowcell.read_length < 40:
  58             read_type = short_read
  59         elif lane.flowcell.read_length < 100:
  60             read_type = medium_read
  61         else:
  62             read_type = long_read
  63         counts[lane_type][read_type] += 1
  64
  65     return counts
  66
  67
  68 def create_library_context(cl):
  69     """
  70      Create a list of libraries that includes how many lanes were run
  71     """
  72     records = []
  73     #for lib in library_items.object_list:
  74     for lib in cl.result_list:
  75         summary = {}
  76         summary['library'] = lib
  77         summary['library_id'] = lib.id
  78         summary['library_name'] = lib.library_name
  79         summary['species_name'] = lib.library_species.scientific_name
  80         if lib.amplified_from_sample is not None:
  81             summary['amplified_from'] = lib.amplified_from_sample.id
  82         else:
  83             summary['amplified_from'] = ''
  84         lanes_run = count_lanes(lib.lane_set)
  85         # suppress zeros
  86         for row_index, row in enumerate(lanes_run):
  87             for col_index, cell in enumerate(row):
  88                 if lanes_run[row_index][col_index] == 0:
  89                     lanes_run[row_index][col_index] = ''
  90         summary['lanes_run'] = lanes_run
  91         summary['is_archived'] = lib.is_archived()
  92         records.append(summary)
  93     cl.result_count = str(cl.paginator._count)
  94     return {'library_list': records}
  95
  96
  97 def library(request, todo_only=False):
  98     queryset = Library.objects.filter(hidden__exact=0)
  99     filters = {'hidden__exact': 0}
 100     if todo_only:
 101         filters['lane'] = None
 102     # build changelist
 103     fcl = HTSChangeList(request, Library,
 104                         list_filter=['affiliations', 'library_species'],
 105                         search_fields=['id', 'library_name', 'amplified_from_sample__id'],
 106                         list_per_page=200,
 107                         model_admin=LibraryOptions(Library, None),
 108                         extra_filters=filters
 109                         )
 110
 111     context = {'cl': fcl, 'title': 'Library Index', 'todo_only': todo_only}
 112     context.update(create_library_context(fcl))
 113     t = get_template('samples/library_index.html')
 114     c = RequestContext(request, context)
 115     return HttpResponse(t.render(c))
 116
 117
 118 def library_not_run(request):
 119     return library(request, todo_only=True)
 120
 121
 122 def library_to_flowcells(request, lib_id):
 123     """
 124     Display information about all the flowcells a library has been run on.
 125     """
 126     try:
 127         lib = Library.objects.get(id=lib_id)
 128     except:
 129         raise Http404('Library %s does not exist' % (lib_id,))
 130
 131     flowcell_list = []
 132     flowcell_run_results = {}  # aka flowcells we're looking at
 133     for lane in lib.lane_set.all():
 134         fc = lane.flowcell
 135         flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
 136         if flowcell_id not in flowcell_run_results:
 137             flowcell_run_results[flowcell_id] = get_flowcell_result_dict(flowcell_id)
 138         flowcell_list.append((fc.flowcell_id, lane.lane_number))
 139
 140     flowcell_list.sort()
 141     lane_summary_list = []
 142     eland_results = []
 143     for fc, lane_number in flowcell_list:
 144         lane_summary, err_list = _summary_stats(fc, lane_number, lib_id)
 145         lane_summary_list.extend(lane_summary)
 146
 147         eland_results.extend(_make_eland_results(fc, lane_number, flowcell_run_results))
 148
 149     context = {
 150         'page_name': 'Library Details',
 151         'lib': lib,
 152         'eland_results': eland_results,
 153         'lane_summary_list': lane_summary_list,
 154     }
 155     context.update(SAMPLES_CONTEXT_DEFAULTS)
 156
 157     return render_to_response(
 158         'samples/library_detail.html',
 159         context,
 160         context_instance=RequestContext(request))
 161
 162
 163 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
 164     """
 165     returns a Summary.htm file if it exists.
 166     """
 167     fc_id, status = parse_flowcell_id(flowcell_id)
 168     d = get_flowcell_result_dict(fc_id)
 169
 170     if d is None:
 171         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 172
 173     if cnm not in d:
 174         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 175
 176     summary_filepath = d[cnm]['summary']
 177
 178     if summary_filepath is None:
 179         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 180
 181     f = open(summary_filepath, 'r')
 182
 183     return HttpResponse(f)
 184
 185
 186 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
 187     """
 188     returns an eland_file upon calling.
 189     """
 190     fc_id, status = parse_flowcell_id(flowcell_id)
 191     d = get_flowcell_result_dict(fc_id)
 192
 193     if d is None:
 194         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 195
 196     if cnm not in d:
 197         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 198
 199     erd = d[cnm]['eland_results']
 200     lane = int(lane)
 201
 202     if lane not in erd:
 203         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 204
 205     filepath = erd[lane]
 206
 207     #f = opener.autoopen(filepath, 'r')
 208     # return HttpResponse(f, content_type="application/x-elandresult")
 209
 210     f = open(filepath, 'r')
 211     return HttpResponse(f, content_type='application/x-bzip2')
 212
 213
 214 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
 215     """
 216     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
 217     """
 218     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
 219
 220
 221 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
 222     """
 223     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
 224     """
 225     fc_id, status = parse_flowcell_id(flowcell_id)
 226     d = get_flowcell_result_dict(fc_id)
 227
 228     if d is None:
 229         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 230
 231     if cnm not in d:
 232         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 233
 234     erd = d[cnm]['eland_results']
 235     lane = int(lane)
 236
 237     if lane not in erd:
 238         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 239
 240     filepath = erd[lane]
 241
 242     # Eland result file
 243     fi = opener.autoopen(filepath, 'r')
 244     # output memory file
 245
 246     name, description = makebed.make_description(fc_id, lane)
 247
 248     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
 249
 250     if ucsc_compatible:
 251         return HttpResponse(bedgen)
 252     else:
 253         return HttpResponse(bedgen, content_type="application/x-bedfile")
 254
 255
 256 def _summary_stats(flowcell_id, lane_id, library_id):
 257     """
 258     Return the summary statistics for a given flowcell, lane, and end.
 259     """
 260     fc_id, status = parse_flowcell_id(flowcell_id)
 261     fc_result_dict = get_flowcell_result_dict(fc_id)
 262
 263     summary_list = []
 264     err_list = []
 265
 266     if fc_result_dict is None:
 267         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 268         return (summary_list, err_list)
 269
 270     for cycle_width in fc_result_dict:
 271         xmlpath = fc_result_dict[cycle_width]['run_xml']
 272
 273         if xmlpath is None:
 274             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
 275             continue
 276
 277         run = runfolder.load_pipeline_run_xml(xmlpath)
 278         # skip if we don't have available metadata.
 279         if run.gerald is None or run.gerald.summary is None:
 280             continue
 281
 282         gerald_summary = run.gerald.summary.lane_results
 283         key = SampleKey(lane=lane_id, sample='s')
 284         eland_results = list(run.gerald.eland_results.find_keys(key))
 285         key = SampleKey(lane=lane_id, sample=library_id)
 286         eland_results.extend(run.gerald.eland_results.find_keys(key))
 287         for key in eland_results:
 288             eland_summary = run.gerald.eland_results.results[key]
 289             # add information to lane_summary
 290             eland_summary.flowcell_id = flowcell_id
 291
 292             read = key.read-1 if key.read is not None else 0
 293             try:
 294                 eland_summary.clusters = gerald_summary[read][key.lane].cluster
 295             except (IndexError, KeyError) as e:
 296                 eland_summary.clustes = None
 297             eland_summary.cycle_width = cycle_width
 298             if hasattr(eland_summary, 'genome_map'):
 299                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads(
 300                     eland_summary.genome_map,
 301                     eland_summary.mapped_reads)
 302
 303             # grab some more information out of the flowcell db
 304             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 305             #pm_field = 'lane_%d_pM' % (lane_id)
 306             lanes = flowcell.lane_set.filter(lane_number=lane_id)
 307             eland_summary.flowcell = flowcell
 308             eland_summary.lanes = lanes
 309
 310             summary_list.append(eland_summary)
 311
 312         #except Exception as e:
 313         #    summary_list.append("Summary report needs to be updated.")
 314         #    LOGGER.error("Exception: " + str(e))
 315
 316     return (summary_list, err_list)
 317
 318
 319 def get_eland_result_type(pathname):
 320     """
 321     Guess the eland result file type from the filename
 322     """
 323     path, filename = os.path.split(pathname)
 324     if 'extended' in filename:
 325         return 'extended'
 326     elif 'multi' in filename:
 327         return 'multi'
 328     elif 'result' in filename:
 329         return 'result'
 330     else:
 331         return 'unknown'
 332
 333
 334 def _make_eland_results(flowcell_id, lane_number, interesting_flowcells):
 335     fc_id, status = parse_flowcell_id(flowcell_id)
 336     cur_fc = interesting_flowcells.get(fc_id, None)
 337     if cur_fc is None:
 338         return []
 339
 340     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 341     lanes = flowcell.lane_set.filter(lane_number=lane_number)
 342     # Loop throw storage devices if a result has been archived
 343     storage_id_list = []
 344     if cur_fc is not None:
 345         for lts in flowcell.longtermstorage_set.all():
 346             for sd in lts.storage_devices.all():
 347                 # Use barcode_id if it exists
 348                 if sd.barcode_id is not None and sd.barcode_id != '':
 349                     storage_id_list.append(sd.barcode_id)
 350                 # Otherwise use UUID
 351                 else:
 352                     storage_id_list.append(sd.uuid)
 353
 354     # Formatting for template use
 355     if len(storage_id_list) == 0:
 356         storage_ids = None
 357     else:
 358         storage_ids = ', '.join(['<a href="/inventory/%s/">%s</a>' % (s, s) for s in storage_id_list])
 359
 360     results = []
 361     for cycle in cur_fc.keys():
 362         result_path = cur_fc[cycle]['eland_results'].get(lanes[0], None)
 363         result_link = make_result_link(fc_id, cycle, lanes[0], result_path)
 364         results.append({'flowcell_id': fc_id,
 365                         'flowcell': flowcell,
 366                         'run_date': flowcell.run_date,
 367                         'cycle': cycle,
 368                         'lane': lanes[0],
 369                         'summary_url': make_summary_url(flowcell_id, cycle),
 370                         'result_url': result_link[0],
 371                         'result_label': result_link[1],
 372                         'bed_url': result_link[2],
 373                         'storage_ids': storage_ids})
 374     return results
 375
 376
 377 def make_summary_url(flowcell_id, cycle_name):
 378     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
 379     return url
 380
 381
 382 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
 383     if eland_result_path is None:
 384         return ("", "", "")
 385
 386     result_type = get_eland_result_type(eland_result_path)
 387     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
 388     result_label = 'eland %s' % (result_type,)
 389     bed_url = None
 390     if result_type == 'result':
 391         bed_url_pattern = '/results/%s/%s/bedfile/%s'
 392         bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
 393
 394     return (result_url, result_label, bed_url)
 395
 396
 397 def _files(flowcell_id, lane):
 398     """
 399     Sets up available files for download
 400     """
 401     lane = int(lane)
 402
 403     flowcell_id, id = parse_flowcell_id(flowcell_id)
 404     d = get_flowcell_result_dict(flowcell_id)
 405
 406     if d is None:
 407         return ''
 408
 409     output = []
 410
 411     # c_name == 'CN-M' (i.e. C1-33)
 412     for c_name in d:
 413
 414         if d[c_name]['summary'] is not None:
 415             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' %
 416                           (flowcell_id, c_name, c_name))
 417
 418         erd = d[c_name]['eland_results']
 419         if lane in erd:
 420             result_type = get_eland_result_type(erd[lane])
 421             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
 422             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
 423             if result_type == 'result':
 424                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
 425                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
 426
 427     if len(output) == 0:
 428         return ''
 429
 430     return '(' + '|'.join(output) + ')'
 431
 432
 433 def library_id_to_admin_url(request, lib_id):
 434     lib = Library.objects.get(id=lib_id)
 435     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
 436
 437
 438 def library_dict(library_id):
 439     """
 440     Given a library id construct a dictionary containing important information
 441     return None if nothing was found
 442     """
 443     try:
 444         lib = Library.objects.get(id=library_id)
 445     except Library.DoesNotExist as e:
 446         return None
 447
 448     #lane_info = lane_information(lib.lane_set)
 449     lane_info = []
 450     for lane in lib.lane_set.all():
 451         lane_info.append({'flowcell': lane.flowcell.flowcell_id,
 452                           'lane_number': lane.lane_number,
 453                           'lane_id': lane.id,
 454                           'paired_end': lane.flowcell.paired_end,
 455                           'read_length': lane.flowcell.read_length,
 456                           'status_code': lane.status,
 457                           'status': LANE_STATUS_MAP[lane.status]})
 458
 459     info = {
 460         # 'affiliations'?
 461         # 'aligned_reads': lib.aligned_reads,
 462         #'amplified_into_sample': lib.amplified_into_sample, # into is a colleciton...
 463         #'amplified_from_sample_id': lib.amplified_from_sample,
 464         #'antibody_name': lib.antibody_name(), # we have no antibodies.
 465         'antibody_id': lib.antibody_id,
 466         'cell_line_id': lib.cell_line_id,
 467         'cell_line': str_or_none(lib.cell_line),
 468         'experiment_type': lib.experiment_type.name,
 469         'experiment_type_id': lib.experiment_type_id,
 470         'gel_cut_size': lib.gel_cut_size,
 471         'hidden': lib.hidden,
 472         'id': lib.id,
 473         'insert_size': lib.insert_size,
 474         'lane_set': lane_info,
 475         'library_id': lib.id,
 476         'library_name': lib.library_name,
 477         'library_species': lib.library_species.scientific_name,
 478         'library_species_id': lib.library_species_id,
 479         #'library_type': lib.library_type.name,
 480         'library_type_id': lib.library_type_id,
 481         'made_for': lib.made_for,
 482         'made_by': lib.made_by,
 483         'notes': lib.notes,
 484         'replicate': lib.replicate,
 485         'stopping_point': lib.stopping_point,
 486         'successful_pM': str_or_none(lib.successful_pM),
 487         'undiluted_concentration': str_or_none(lib.undiluted_concentration)
 488         }
 489     if lib.library_type_id is None:
 490         info['library_type'] = None
 491     else:
 492         info['library_type'] = lib.library_type.name
 493     return info
 494
 495
 496 @csrf_exempt
 497 def library_json(request, library_id):
 498     """
 499     Return a json formatted library dictionary
 500     """
 501     require_api_key(request)
 502     # what validation should we do on library_id?
 503
 504     lib = library_dict(library_id)
 505     if lib is None:
 506         raise Http404
 507
 508     lib_json = json.dumps({'result': lib})
 509     return HttpResponse(lib_json, content_type='application/json')
 510
 511
 512 @csrf_exempt
 513 def species_json(request, species_id):
 514     """
 515     Return information about a species.
 516     """
 517     raise Http404
 518
 519
 520 def species(request, species_id):
 521     species = get_object_or_404(Species, id=species_id)
 522
 523     context = RequestContext(request,
 524                              {'species': species})
 525
 526     return render_to_response("samples/species_detail.html", context)
 527
 528
 529 def antibodies(request):
 530     context = RequestContext(request,
 531                              {'antibodies': Antibody.objects.order_by('antigene')})
 532     return render_to_response("samples/antibody_index.html", context)
 533
 534
 535 @login_required
 536 def user_profile(request):
 537     """
 538     Information about the user
 539     """
 540     context = {
 541         'page_name': 'User Profile',
 542         'media': '',
 543         #'bcmagic': BarcodeMagicForm(),
 544         #'select': 'settings',
 545     }
 546     context.update(SAMPLES_CONTEXT_DEFAULTS)
 547     return render_to_response('registration/profile.html', context,
 548                               context_instance=RequestContext(request))