htsworkflow/frontend/samples/views.py

   1 # Create your views here.
   2 import StringIO
   3 import logging
   4 import os
   5 import sys
   6
   7 try:
   8     import json
   9 except ImportError, e:
  10     import simplejson as json
  11
  12 from htsworkflow.frontend.auth import require_api_key
  13 from htsworkflow.frontend.experiments.models import FlowCell
  14 from htsworkflow.frontend.samples.changelist import ChangeList
  15 from htsworkflow.frontend.samples.models import Library
  16 from htsworkflow.frontend.samples.results import get_flowcell_result_dict, parse_flowcell_id
  17 from htsworkflow.frontend.bcmagic.forms import BarcodeMagicForm
  18 from htsworkflow.pipelines.runfolder import load_pipeline_run_xml
  19 from htsworkflow.pipelines import runfolder
  20 from htsworkflow.pipelines.eland import ResultLane
  21 from htsworkflow.frontend import settings
  22 from htsworkflow.util.conversion import unicode_or_none
  23 from htsworkflow.util import makebed
  24 from htsworkflow.util import opener
  25
  26
  27 from django.core.exceptions import ObjectDoesNotExist
  28 from django.http import HttpResponse, HttpResponseRedirect, Http404
  29 from django.shortcuts import render_to_response
  30 from django.template import RequestContext
  31 from django.template.loader import get_template
  32 from django.contrib.auth.decorators import login_required
  33
  34 LANE_LIST = [1,2,3,4,5,6,7,8]
  35 SAMPLES_CONTEXT_DEFAULTS = {
  36     'app_name': 'Flowcell/Library Tracker',
  37     'bcmagic': BarcodeMagicForm()
  38 }
  39
  40 def create_library_context(cl):
  41     """
  42     Create a list of libraries that includes how many lanes were run
  43     """
  44     records = []
  45     #for lib in library_items.object_list:
  46     for lib in cl.result_list:
  47        summary = {}
  48        summary['library_id'] = lib.id
  49        summary['library_name'] = lib.library_name
  50        summary['species_name' ] = lib.library_species.scientific_name
  51        if lib.amplified_from_sample is not None:
  52            summary['amplified_from'] = lib.amplified_from_sample.id
  53        else:
  54            summary['amplified_from'] = ''
  55        lanes_run = 0
  56        #for lane_id in LANE_LIST:
  57        #    lane = getattr(lib, 'lane_%d_library' % (lane_id,))
  58        #    lanes_run += len( lane.all() )
  59        lanes_run = lib.lane_set.count()
  60        summary['lanes_run'] = lanes_run
  61        summary['is_archived'] = lib.is_archived()
  62        records.append(summary)
  63     cl.result_count = unicode(cl.paginator._count) + u" libraries"
  64     return {'library_list': records }
  65
  66 def library(request):
  67    # build changelist
  68     fcl = ChangeList(request, Library,
  69         list_filter=['affiliations', 'library_species'],
  70         search_fields=['id', 'library_name', 'amplified_from_sample__id'],
  71         list_per_page=200,
  72         queryset=Library.objects.filter(hidden__exact=0)
  73     )
  74
  75     context = { 'cl': fcl, 'title': 'Library Index'}
  76     context.update(create_library_context(fcl))
  77     t = get_template('samples/library_index.html')
  78     c = RequestContext(request, context)
  79
  80     app_context = {
  81         'page_name': 'Library Index',
  82         'east_region_config_div': 'changelist-filter',
  83         'body': t.render(c)
  84     }
  85     app_context.update(SAMPLES_CONTEXT_DEFAULTS)
  86
  87     app_t = get_template('flowcell_libraries_app.html')
  88     app_c = RequestContext(request, app_context)
  89     return HttpResponse( app_t.render(app_c) )
  90
  91 def library_to_flowcells(request, lib_id):
  92     """
  93     Display information about all the flowcells a library has been run on.
  94     """
  95
  96     try:
  97       lib = Library.objects.get(id=lib_id)
  98     except:
  99       return HttpResponse("Library %s does not exist" % (lib_id))
 100
 101     flowcell_list = []
 102     flowcell_run_results = {} # aka flowcells we're looking at
 103     for lane in lib.lane_set.all():
 104         fc = lane.flowcell
 105         flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
 106         if flowcell_id not in flowcell_run_results:
 107             flowcell_run_results[flowcell_id] = get_flowcell_result_dict(flowcell_id)
 108         flowcell_list.append((fc.flowcell_id, lane.lane_number))
 109
 110     flowcell_list.sort()
 111     lane_summary_list = []
 112     eland_results = []
 113     for fc, lane_number in flowcell_list:
 114         lane_summary, err_list = _summary_stats(fc, lane_number)
 115
 116         eland_results.extend(_make_eland_results(fc, lane_number, flowcell_run_results))
 117         lane_summary_list.extend(lane_summary)
 118
 119     context = {
 120         'page_name': 'Library Details',
 121         'lib': lib,
 122         'eland_results': eland_results,
 123         'lane_summary_list': lane_summary_list,
 124     }
 125     context.update(SAMPLES_CONTEXT_DEFAULTS)
 126
 127     return render_to_response(
 128         'samples/library_detail.html',
 129         context,
 130         context_instance = RequestContext(request))
 131
 132 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
 133     """
 134     returns a Summary.htm file if it exists.
 135     """
 136     fc_id, status = parse_flowcell_id(flowcell_id)
 137     d = get_flowcell_result_dict(fc_id)
 138
 139     if d is None:
 140         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 141
 142     if cnm not in d:
 143         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 144
 145     summary_filepath = d[cnm]['summary']
 146
 147     if summary_filepath is None:
 148         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 149
 150     f = open(summary_filepath, 'r')
 151
 152     return HttpResponse(f)
 153
 154
 155 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
 156     """
 157     returns an eland_file upon calling.
 158     """
 159     fc_id, status = parse_flowcell_id(flowcell_id)
 160     d = get_flowcell_result_dict(fc_id)
 161
 162     if d is None:
 163         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 164
 165     if cnm not in d:
 166         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 167
 168     erd = d[cnm]['eland_results']
 169     lane = int(lane)
 170
 171     if lane not in erd:
 172         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 173
 174     filepath = erd[lane]
 175
 176     #f = opener.autoopen(filepath, 'r')
 177     # return HttpResponse(f, mimetype="application/x-elandresult")
 178
 179     f = open(filepath, 'r')
 180     return HttpResponse(f, mimetype='application/x-bzip2')
 181
 182
 183
 184 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
 185     """
 186     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
 187     """
 188     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
 189
 190
 191 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
 192     """
 193     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
 194     """
 195     fc_id, status = parse_flowcell_id(flowcell_id)
 196     d = get_flowcell_result_dict(fc_id)
 197
 198     if d is None:
 199         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 200
 201     if cnm not in d:
 202         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 203
 204     erd = d[cnm]['eland_results']
 205     lane = int(lane)
 206
 207     if lane not in erd:
 208         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 209
 210     filepath = erd[lane]
 211
 212     # Eland result file
 213     fi = opener.autoopen(filepath, 'r')
 214     # output memory file
 215
 216     name, description = makebed.make_description( fc_id, lane )
 217
 218     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
 219
 220     if ucsc_compatible:
 221         return HttpResponse(bedgen)
 222     else:
 223         return HttpResponse(bedgen, mimetype="application/x-bedfile")
 224
 225
 226 def _summary_stats(flowcell_id, lane_id):
 227     """
 228     Return the summary statistics for a given flowcell, lane, and end.
 229     """
 230     fc_id, status = parse_flowcell_id(flowcell_id)
 231     fc_result_dict = get_flowcell_result_dict(fc_id)
 232
 233     summary_list = []
 234     err_list = []
 235
 236     if fc_result_dict is None:
 237         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 238         return (summary_list, err_list)
 239
 240     for cycle_width in fc_result_dict:
 241         xmlpath = fc_result_dict[cycle_width]['run_xml']
 242
 243         if xmlpath is None:
 244             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
 245             continue
 246
 247         run = load_pipeline_run_xml(xmlpath)
 248         gerald_summary = run.gerald.summary.lane_results
 249         for end in range(len(gerald_summary)):
 250             end_summary = run.gerald.eland_results.results[end]
 251             if end_summary.has_key(lane_id):
 252                 eland_summary = run.gerald.eland_results.results[end][lane_id]
 253             else:
 254                 eland_summary = ResultLane(lane_id=lane_id, end=end)
 255             # add information to lane_summary
 256             eland_summary.flowcell_id = flowcell_id
 257             if len(gerald_summary) > end and gerald_summary[end].has_key(lane_id):
 258                 eland_summary.clusters = gerald_summary[end][lane_id].cluster
 259             else:
 260                 eland_summary.clusters = None
 261             eland_summary.cycle_width = cycle_width
 262             if hasattr(eland_summary, 'genome_map'):
 263                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads(
 264                                                    eland_summary.genome_map,
 265                                                    eland_summary.mapped_reads)
 266
 267             # grab some more information out of the flowcell db
 268             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 269             #pm_field = 'lane_%d_pM' % (lane_id)
 270             lane_obj = flowcell.lane_set.get(lane_number=lane_id)
 271             eland_summary.successful_pm = lane_obj.pM
 272
 273             summary_list.append(eland_summary)
 274
 275         #except Exception, e:
 276         #    summary_list.append("Summary report needs to be updated.")
 277         #    logging.error("Exception: " + str(e))
 278
 279     return (summary_list, err_list)
 280
 281 def _summary_stats_old(flowcell_id, lane):
 282     """
 283     return a dictionary of summary stats for a given flowcell_id & lane.
 284     """
 285     fc_id, status = parse_flowcell_id(flowcell_id)
 286     fc_result_dict = get_flowcell_result_dict(fc_id)
 287
 288     dict_list = []
 289     err_list = []
 290     summary_list = []
 291
 292     if fc_result_dict is None:
 293         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 294         return (dict_list, err_list, summary_list)
 295
 296     for cnm in fc_result_dict:
 297
 298         xmlpath = fc_result_dict[cnm]['run_xml']
 299
 300         if xmlpath is None:
 301             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cnm))
 302             continue
 303
 304         tree = ElementTree.parse(xmlpath).getroot()
 305         results = runfolder.PipelineRun(pathname='', xml=tree)
 306         try:
 307             lane_report = runfolder.summarize_lane(results.gerald, lane)
 308             summary_list.append(os.linesep.join(lane_report))
 309         except Exception, e:
 310             summary_list.append("Summary report needs to be updated.")
 311             logging.error("Exception: " + str(e))
 312
 313         print >>sys.stderr, "----------------------------------"
 314         print >>sys.stderr, "-- DOES NOT SUPPORT PAIRED END ---"
 315         print >>sys.stderr, "----------------------------------"
 316         lane_results = results.gerald.summary[0][lane]
 317         lrs = lane_results
 318
 319         d = {}
 320
 321         d['average_alignment_score'] = lrs.average_alignment_score
 322         d['average_first_cycle_intensity'] = lrs.average_first_cycle_intensity
 323         d['cluster'] = lrs.cluster
 324         d['lane'] = lrs.lane
 325         d['flowcell'] = flowcell_id
 326         d['cnm'] = cnm
 327         d['percent_error_rate'] = lrs.percent_error_rate
 328         d['percent_intensity_after_20_cycles'] = lrs.percent_intensity_after_20_cycles
 329         d['percent_pass_filter_align'] = lrs.percent_pass_filter_align
 330         d['percent_pass_filter_clusters'] = lrs.percent_pass_filter_clusters
 331
 332         #FIXME: function finished, but need to take advantage of
 333         #   may need to take in a list of lanes so we only have to
 334         #   load the xml file once per flowcell rather than once
 335         #   per lane.
 336         dict_list.append(d)
 337
 338     return (dict_list, err_list, summary_list)
 339
 340
 341 def get_eland_result_type(pathname):
 342     """
 343     Guess the eland result file type from the filename
 344     """
 345     path, filename = os.path.split(pathname)
 346     if 'extended' in filename:
 347         return 'extended'
 348     elif 'multi' in filename:
 349         return 'multi'
 350     elif 'result' in filename:
 351         return 'result'
 352     else:
 353         return 'unknown'
 354
 355 def _make_eland_results(flowcell_id, lane, interesting_flowcells):
 356     fc_id, status = parse_flowcell_id(flowcell_id)
 357     cur_fc = interesting_flowcells.get(fc_id, None)
 358     if cur_fc is None:
 359       return []
 360
 361     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 362     # Loop throw storage devices if a result has been archived
 363     storage_id_list = []
 364     if cur_fc is not None:
 365         for lts in flowcell.longtermstorage_set.all():
 366             for sd in lts.storage_devices.all():
 367                 # Use barcode_id if it exists
 368                 if sd.barcode_id is not None and sd.barcode_id != '':
 369                     storage_id_list.append(sd.barcode_id)
 370                 # Otherwise use UUID
 371                 else:
 372                     storage_id_list.append(sd.uuid)
 373
 374     # Formatting for template use
 375     if len(storage_id_list) == 0:
 376         storage_ids = None
 377     else:
 378         storage_ids = ', '.join([ '<a href="/inventory/%s/">%s</a>' % (s,s) for s in storage_id_list ])
 379
 380     results = []
 381     for cycle in cur_fc.keys():
 382         result_path = cur_fc[cycle]['eland_results'].get(lane, None)
 383         result_link = make_result_link(fc_id, cycle, lane, result_path)
 384         results.append({'flowcell_id': fc_id,
 385                         'run_date': flowcell.run_date,
 386                         'cycle': cycle,
 387                         'lane': lane,
 388                         'summary_url': make_summary_url(flowcell_id, cycle),
 389                         'result_url': result_link[0],
 390                         'result_label': result_link[1],
 391                         'bed_url': result_link[2],
 392                         'storage_ids': storage_ids
 393         })
 394     return results
 395
 396 def make_summary_url(flowcell_id, cycle_name):
 397     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
 398     return url
 399
 400 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
 401     if eland_result_path is None:
 402         return ("", "", "")
 403
 404     result_type = get_eland_result_type(eland_result_path)
 405     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
 406     result_label = 'eland %s' % (result_type,)
 407     bed_url = None
 408     if result_type == 'result':
 409        bed_url_pattern = '/results/%s/%s/bedfile/%s'
 410        bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
 411
 412     return (result_url, result_label, bed_url)
 413
 414 def _files(flowcell_id, lane):
 415     """
 416     Sets up available files for download
 417     """
 418     lane = int(lane)
 419
 420     flowcell_id, id = parse_flowcell_id(flowcell_id)
 421     d = get_flowcell_result_dict(flowcell_id)
 422
 423     if d is None:
 424         return ''
 425
 426     output = []
 427
 428     # c_name == 'CN-M' (i.e. C1-33)
 429     for c_name in d:
 430
 431         if d[c_name]['summary'] is not None:
 432             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' \
 433                           % (flowcell_id, c_name, c_name))
 434
 435         erd = d[c_name]['eland_results']
 436         if lane in erd:
 437             result_type = get_eland_result_type(erd[lane])
 438             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
 439             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
 440             if result_type == 'result':
 441                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
 442                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
 443
 444     if len(output) == 0:
 445         return ''
 446
 447     return '(' + '|'.join(output) + ')'
 448
 449 def library_id_to_admin_url(request, lib_id):
 450     lib = Library.objects.get(id=lib_id)
 451     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
 452
 453 def library_dict(library_id):
 454     """
 455     Given a library id construct a dictionary containing important information
 456     return None if nothing was found
 457     """
 458     try:
 459         lib = Library.objects.get(id = library_id)
 460     except Library.DoesNotExist, e:
 461         return None
 462
 463     #lane_info = lane_information(lib.lane_set)
 464     lane_info = []
 465     for lane in lib.lane_set.all():
 466         lane_info.append( {'flowcell':lane.flowcell.flowcell_id,
 467                            'lane_number': lane.lane_number} )
 468
 469     info = {
 470         # 'affiliations'?
 471         # 'aligned_reads': lib.aligned_reads,
 472         #'amplified_into_sample': lib.amplified_into_sample, # into is a colleciton...
 473         #'amplified_from_sample_id': lib.amplified_from_sample,
 474         #'antibody_name': lib.antibody_name(), # we have no antibodies.
 475         'antibody_id': lib.antibody_id,
 476         'avg_lib_size': lib.avg_lib_size,
 477         'cell_line_id': lib.cell_line_id,
 478         'cell_line': unicode_or_none(lib.cell_line),
 479         'experiment_type': lib.experiment_type.name,
 480         'experiment_type_id': lib.experiment_type_id,
 481         'id': lib.id,
 482         'lane_set': lane_info,
 483         'library_id': lib.id,
 484         'library_name': lib.library_name,
 485         'library_species': lib.library_species.scientific_name,
 486         'library_species_id': lib.library_species_id,
 487         #'library_type': lib.library_type.name,
 488         'library_type_id': lib.library_type_id,
 489         'made_for': lib.made_for,
 490         'made_by': lib.made_by,
 491         'notes': lib.notes,
 492         'replicate': lib.replicate,
 493         'stopping_point': lib.stopping_point,
 494         'successful_pM': unicode_or_none(lib.successful_pM),
 495         'undiluted_concentration': unicode_or_none(lib.undiluted_concentration)
 496         }
 497     if lib.library_type_id is None:
 498         info['library_type'] = None
 499     else:
 500         info['library_type'] = lib.library_type.name
 501     return info
 502
 503 def library_json(request, library_id):
 504     """
 505     Return a json formatted library dictionary
 506     """
 507     require_api_key(request)
 508     # what validation should we do on library_id?
 509
 510     lib = library_dict(library_id)
 511     if lib is None:
 512         raise Http404
 513
 514     lib_json = json.dumps(lib)
 515     return HttpResponse(lib_json, mimetype='application/json')
 516
 517 def species_json(request, species_id):
 518     """
 519     Return information about a species.
 520     """
 521     raise Http404
 522
 523 @login_required
 524 def user_profile(request):
 525     """
 526     Information about the user
 527     """
 528     context = {
 529                 'page_name': 'User Profile',
 530                 'media': '',
 531                 #'bcmagic': BarcodeMagicForm(),
 532                 #'select': 'settings',
 533             }
 534     context.update(SAMPLES_CONTEXT_DEFAULTS)
 535     return render_to_response('registration/profile.html', context,
 536                               context_instance=RequestContext(request))
 537