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