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, Lane, LANE_STATUS_MAP
  14 from htsworkflow.frontend.samples.changelist import ChangeList
  15 from htsworkflow.frontend.samples.models import Library, Species, HTSUser
  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.util.conversion import unicode_or_none
  22 from htsworkflow.util import makebed
  23 from htsworkflow.util import opener
  24
  25
  26 from django.core.exceptions import ObjectDoesNotExist
  27 from django.http import HttpResponse, HttpResponseRedirect, Http404
  28 from django.shortcuts import render_to_response, get_object_or_404
  29 from django.template import RequestContext
  30 from django.template.loader import get_template
  31 from django.contrib.auth.decorators import login_required
  32 from django.conf import settings
  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 count_lanes(lane_set):
  41     single = 0
  42     paired = 1
  43     short_read = 0
  44     medium_read = 1
  45     long_read = 2
  46     counts = [[0,0,0,],[0,0,0]]
  47
  48     for lane in lane_set.all():
  49         if lane.flowcell.paired_end:
  50             lane_type = paired
  51         else:
  52             lane_type = single
  53         if lane.flowcell.read_length < 40:
  54             read_type = short_read
  55         elif lane.flowcell.read_length < 100:
  56             read_type = medium_read
  57         else:
  58             read_type = long_read
  59         counts[lane_type][read_type] += 1
  60
  61     return counts
  62
  63 def create_library_context(cl):
  64     """
  65      Create a list of libraries that includes how many lanes were run
  66     """
  67     records = []
  68     #for lib in library_items.object_list:
  69     for lib in cl.result_list:
  70        summary = {}
  71        summary['library'] = lib
  72        summary['library_id'] = lib.id
  73        summary['library_name'] = lib.library_name
  74        summary['species_name' ] = lib.library_species.scientific_name
  75        if lib.amplified_from_sample is not None:
  76            summary['amplified_from'] = lib.amplified_from_sample.id
  77        else:
  78            summary['amplified_from'] = ''
  79        lanes_run = count_lanes(lib.lane_set)
  80        # suppress zeros
  81        for row in xrange(len(lanes_run)):
  82            for col in xrange(len(lanes_run[row])):
  83                if lanes_run[row][col] == 0:
  84                    lanes_run[row][col] = ''
  85        summary['lanes_run'] = lanes_run
  86        summary['is_archived'] = lib.is_archived()
  87        records.append(summary)
  88     cl.result_count = unicode(cl.paginator._count)
  89     return {'library_list': records }
  90
  91 def library(request):
  92     # build changelist
  93     fcl = ChangeList(request, Library,
  94         list_filter=['affiliations', 'library_species'],
  95         search_fields=['id', 'library_name', 'amplified_from_sample__id'],
  96         list_per_page=200,
  97         queryset=Library.objects.filter(hidden__exact=0)
  98     )
  99
 100     context = { 'cl': fcl, 'title': 'Library Index'}
 101     context.update(create_library_context(fcl))
 102     t = get_template('samples/library_index.html')
 103     c = RequestContext(request, context)
 104     return HttpResponse( t.render(c) )
 105
 106
 107 def library_to_flowcells(request, lib_id):
 108     """
 109     Display information about all the flowcells a library has been run on.
 110     """
 111
 112     try:
 113       lib = Library.objects.get(id=lib_id)
 114     except:
 115       return HttpResponse("Library %s does not exist" % (lib_id))
 116
 117     flowcell_list = []
 118     flowcell_run_results = {} # aka flowcells we're looking at
 119     for lane in lib.lane_set.all():
 120         fc = lane.flowcell
 121         flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
 122         if flowcell_id not in flowcell_run_results:
 123             flowcell_run_results[flowcell_id] = get_flowcell_result_dict(flowcell_id)
 124         flowcell_list.append((fc.flowcell_id, lane.lane_number))
 125
 126     flowcell_list.sort()
 127     lane_summary_list = []
 128     eland_results = []
 129     for fc, lane_number in flowcell_list:
 130         lane_summary, err_list = _summary_stats(fc, lane_number)
 131
 132         eland_results.extend(_make_eland_results(fc, lane_number, flowcell_run_results))
 133         lane_summary_list.extend(lane_summary)
 134
 135     context = {
 136         'page_name': 'Library Details',
 137         'lib': lib,
 138         'eland_results': eland_results,
 139         'lane_summary_list': lane_summary_list,
 140     }
 141     context.update(SAMPLES_CONTEXT_DEFAULTS)
 142
 143     return render_to_response(
 144         'samples/library_detail.html',
 145         context,
 146         context_instance = RequestContext(request))
 147
 148 def lanes_for(request, username=None):
 149     """
 150     Generate a report of recent activity for a user
 151     """
 152     query = {}
 153     if username is not None:
 154         user = HTSUser.objects.get(username=username)
 155         query.update({'library__affiliations__users__id':user.id})
 156     fcl = ChangeList(request, Lane,
 157         list_filter=[],
 158         search_fields=['flowcell__flowcell_id', 'library__id', 'library__library_name'],
 159         list_per_page=200,
 160         queryset=Lane.objects.filter(**query)
 161     )
 162
 163     context = { 'lanes': fcl, 'title': 'Lane Index'}
 164
 165     return render_to_response(
 166         'samples/lanes_for.html',
 167         context,
 168         context_instance = RequestContext(request)
 169     )
 170
 171
 172 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
 173     """
 174     returns a Summary.htm file if it exists.
 175     """
 176     fc_id, status = parse_flowcell_id(flowcell_id)
 177     d = get_flowcell_result_dict(fc_id)
 178
 179     if d is None:
 180         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 181
 182     if cnm not in d:
 183         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 184
 185     summary_filepath = d[cnm]['summary']
 186
 187     if summary_filepath is None:
 188         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 189
 190     f = open(summary_filepath, 'r')
 191
 192     return HttpResponse(f)
 193
 194
 195 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
 196     """
 197     returns an eland_file upon calling.
 198     """
 199     fc_id, status = parse_flowcell_id(flowcell_id)
 200     d = get_flowcell_result_dict(fc_id)
 201
 202     if d is None:
 203         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 204
 205     if cnm not in d:
 206         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 207
 208     erd = d[cnm]['eland_results']
 209     lane = int(lane)
 210
 211     if lane not in erd:
 212         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 213
 214     filepath = erd[lane]
 215
 216     #f = opener.autoopen(filepath, 'r')
 217     # return HttpResponse(f, mimetype="application/x-elandresult")
 218
 219     f = open(filepath, 'r')
 220     return HttpResponse(f, mimetype='application/x-bzip2')
 221
 222
 223
 224 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
 225     """
 226     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
 227     """
 228     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
 229
 230
 231 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
 232     """
 233     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
 234     """
 235     fc_id, status = parse_flowcell_id(flowcell_id)
 236     d = get_flowcell_result_dict(fc_id)
 237
 238     if d is None:
 239         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 240
 241     if cnm not in d:
 242         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 243
 244     erd = d[cnm]['eland_results']
 245     lane = int(lane)
 246
 247     if lane not in erd:
 248         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 249
 250     filepath = erd[lane]
 251
 252     # Eland result file
 253     fi = opener.autoopen(filepath, 'r')
 254     # output memory file
 255
 256     name, description = makebed.make_description( fc_id, lane )
 257
 258     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
 259
 260     if ucsc_compatible:
 261         return HttpResponse(bedgen)
 262     else:
 263         return HttpResponse(bedgen, mimetype="application/x-bedfile")
 264
 265
 266 def _summary_stats(flowcell_id, lane_id):
 267     """
 268     Return the summary statistics for a given flowcell, lane, and end.
 269     """
 270     fc_id, status = parse_flowcell_id(flowcell_id)
 271     fc_result_dict = get_flowcell_result_dict(fc_id)
 272
 273     summary_list = []
 274     err_list = []
 275
 276     if fc_result_dict is None:
 277         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 278         return (summary_list, err_list)
 279
 280     for cycle_width in fc_result_dict:
 281         xmlpath = fc_result_dict[cycle_width]['run_xml']
 282
 283         if xmlpath is None:
 284             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
 285             continue
 286
 287         run = load_pipeline_run_xml(xmlpath)
 288         gerald_summary = run.gerald.summary.lane_results
 289         for end in range(len(gerald_summary)):
 290             end_summary = run.gerald.eland_results.results[end]
 291             if end_summary.has_key(lane_id):
 292                 eland_summary = run.gerald.eland_results.results[end][lane_id]
 293             else:
 294                 eland_summary = ResultLane(lane_id=lane_id, end=end)
 295             # add information to lane_summary
 296             eland_summary.flowcell_id = flowcell_id
 297             if len(gerald_summary) > end and gerald_summary[end].has_key(lane_id):
 298                 eland_summary.clusters = gerald_summary[end][lane_id].cluster
 299             else:
 300                 eland_summary.clusters = None
 301             eland_summary.cycle_width = cycle_width
 302             if hasattr(eland_summary, 'genome_map'):
 303                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads(
 304                                                    eland_summary.genome_map,
 305                                                    eland_summary.mapped_reads)
 306
 307             # grab some more information out of the flowcell db
 308             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 309             #pm_field = 'lane_%d_pM' % (lane_id)
 310             lane_obj = flowcell.lane_set.get(lane_number=lane_id)
 311             eland_summary.successful_pm = lane_obj.pM
 312
 313             summary_list.append(eland_summary)
 314
 315         #except Exception, e:
 316         #    summary_list.append("Summary report needs to be updated.")
 317         #    logging.error("Exception: " + str(e))
 318
 319     return (summary_list, err_list)
 320
 321 def _summary_stats_old(flowcell_id, lane):
 322     """
 323     return a dictionary of summary stats for a given flowcell_id & lane.
 324     """
 325     fc_id, status = parse_flowcell_id(flowcell_id)
 326     fc_result_dict = get_flowcell_result_dict(fc_id)
 327
 328     dict_list = []
 329     err_list = []
 330     summary_list = []
 331
 332     if fc_result_dict is None:
 333         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 334         return (dict_list, err_list, summary_list)
 335
 336     for cnm in fc_result_dict:
 337
 338         xmlpath = fc_result_dict[cnm]['run_xml']
 339
 340         if xmlpath is None:
 341             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cnm))
 342             continue
 343
 344         tree = ElementTree.parse(xmlpath).getroot()
 345         results = runfolder.PipelineRun(pathname='', xml=tree)
 346         try:
 347             lane_report = runfolder.summarize_lane(results.gerald, lane)
 348             summary_list.append(os.linesep.join(lane_report))
 349         except Exception, e:
 350             summary_list.append("Summary report needs to be updated.")
 351             logging.error("Exception: " + str(e))
 352
 353         print >>sys.stderr, "----------------------------------"
 354         print >>sys.stderr, "-- DOES NOT SUPPORT PAIRED END ---"
 355         print >>sys.stderr, "----------------------------------"
 356         lane_results = results.gerald.summary[0][lane]
 357         lrs = lane_results
 358
 359         d = {}
 360
 361         d['average_alignment_score'] = lrs.average_alignment_score
 362         d['average_first_cycle_intensity'] = lrs.average_first_cycle_intensity
 363         d['cluster'] = lrs.cluster
 364         d['lane'] = lrs.lane
 365         d['flowcell'] = flowcell_id
 366         d['cnm'] = cnm
 367         d['percent_error_rate'] = lrs.percent_error_rate
 368         d['percent_intensity_after_20_cycles'] = lrs.percent_intensity_after_20_cycles
 369         d['percent_pass_filter_align'] = lrs.percent_pass_filter_align
 370         d['percent_pass_filter_clusters'] = lrs.percent_pass_filter_clusters
 371
 372         #FIXME: function finished, but need to take advantage of
 373         #   may need to take in a list of lanes so we only have to
 374         #   load the xml file once per flowcell rather than once
 375         #   per lane.
 376         dict_list.append(d)
 377
 378     return (dict_list, err_list, summary_list)
 379
 380
 381 def get_eland_result_type(pathname):
 382     """
 383     Guess the eland result file type from the filename
 384     """
 385     path, filename = os.path.split(pathname)
 386     if 'extended' in filename:
 387         return 'extended'
 388     elif 'multi' in filename:
 389         return 'multi'
 390     elif 'result' in filename:
 391         return 'result'
 392     else:
 393         return 'unknown'
 394
 395 def _make_eland_results(flowcell_id, lane_number, interesting_flowcells):
 396     fc_id, status = parse_flowcell_id(flowcell_id)
 397     cur_fc = interesting_flowcells.get(fc_id, None)
 398     if cur_fc is None:
 399       return []
 400
 401     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 402     lane = flowcell.lane_set.get(lane_number=lane_number)
 403     # Loop throw storage devices if a result has been archived
 404     storage_id_list = []
 405     if cur_fc is not None:
 406         for lts in flowcell.longtermstorage_set.all():
 407             for sd in lts.storage_devices.all():
 408                 # Use barcode_id if it exists
 409                 if sd.barcode_id is not None and sd.barcode_id != '':
 410                     storage_id_list.append(sd.barcode_id)
 411                 # Otherwise use UUID
 412                 else:
 413                     storage_id_list.append(sd.uuid)
 414
 415     # Formatting for template use
 416     if len(storage_id_list) == 0:
 417         storage_ids = None
 418     else:
 419         storage_ids = ', '.join([ '<a href="/inventory/%s/">%s</a>' % (s,s) for s in storage_id_list ])
 420
 421     results = []
 422     for cycle in cur_fc.keys():
 423         result_path = cur_fc[cycle]['eland_results'].get(lane, None)
 424         result_link = make_result_link(fc_id, cycle, lane, result_path)
 425         results.append({'flowcell_id': fc_id,
 426                         'flowcell': flowcell,
 427                         'run_date': flowcell.run_date,
 428                         'cycle': cycle,
 429                         'lane': lane,
 430                         'summary_url': make_summary_url(flowcell_id, cycle),
 431                         'result_url': result_link[0],
 432                         'result_label': result_link[1],
 433                         'bed_url': result_link[2],
 434                         'storage_ids': storage_ids
 435         })
 436     return results
 437
 438 def make_summary_url(flowcell_id, cycle_name):
 439     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
 440     return url
 441
 442 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
 443     if eland_result_path is None:
 444         return ("", "", "")
 445
 446     result_type = get_eland_result_type(eland_result_path)
 447     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
 448     result_label = 'eland %s' % (result_type,)
 449     bed_url = None
 450     if result_type == 'result':
 451        bed_url_pattern = '/results/%s/%s/bedfile/%s'
 452        bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
 453
 454     return (result_url, result_label, bed_url)
 455
 456 def _files(flowcell_id, lane):
 457     """
 458     Sets up available files for download
 459     """
 460     lane = int(lane)
 461
 462     flowcell_id, id = parse_flowcell_id(flowcell_id)
 463     d = get_flowcell_result_dict(flowcell_id)
 464
 465     if d is None:
 466         return ''
 467
 468     output = []
 469
 470     # c_name == 'CN-M' (i.e. C1-33)
 471     for c_name in d:
 472
 473         if d[c_name]['summary'] is not None:
 474             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' \
 475                           % (flowcell_id, c_name, c_name))
 476
 477         erd = d[c_name]['eland_results']
 478         if lane in erd:
 479             result_type = get_eland_result_type(erd[lane])
 480             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
 481             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
 482             if result_type == 'result':
 483                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
 484                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
 485
 486     if len(output) == 0:
 487         return ''
 488
 489     return '(' + '|'.join(output) + ')'
 490
 491 def library_id_to_admin_url(request, lib_id):
 492     lib = Library.objects.get(id=lib_id)
 493     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
 494
 495 def library_dict(library_id):
 496     """
 497     Given a library id construct a dictionary containing important information
 498     return None if nothing was found
 499     """
 500     try:
 501         lib = Library.objects.get(id = library_id)
 502     except Library.DoesNotExist, e:
 503         return None
 504
 505     #lane_info = lane_information(lib.lane_set)
 506     lane_info = []
 507     for lane in lib.lane_set.all():
 508         lane_info.append( {'flowcell':lane.flowcell.flowcell_id,
 509                            'lane_number': lane.lane_number,
 510                            'paired_end': lane.flowcell.paired_end,
 511                            'read_length': lane.flowcell.read_length,
 512                            'status_code': lane.status,
 513                            'status': LANE_STATUS_MAP[lane.status]} )
 514
 515     info = {
 516         # 'affiliations'?
 517         # 'aligned_reads': lib.aligned_reads,
 518         #'amplified_into_sample': lib.amplified_into_sample, # into is a colleciton...
 519         #'amplified_from_sample_id': lib.amplified_from_sample,
 520         #'antibody_name': lib.antibody_name(), # we have no antibodies.
 521         'antibody_id': lib.antibody_id,
 522         'cell_line_id': lib.cell_line_id,
 523         'cell_line': unicode_or_none(lib.cell_line),
 524         'experiment_type': lib.experiment_type.name,
 525         'experiment_type_id': lib.experiment_type_id,
 526         'gel_cut_size': lib.gel_cut_size,
 527         'hidden': lib.hidden,
 528         'id': lib.id,
 529         'insert_size': lib.insert_size,
 530         'lane_set': lane_info,
 531         'library_id': lib.id,
 532         'library_name': lib.library_name,
 533         'library_species': lib.library_species.scientific_name,
 534         'library_species_id': lib.library_species_id,
 535         #'library_type': lib.library_type.name,
 536         'library_type_id': lib.library_type_id,
 537         'made_for': lib.made_for,
 538         'made_by': lib.made_by,
 539         'notes': lib.notes,
 540         'replicate': lib.replicate,
 541         'stopping_point': lib.stopping_point,
 542         'successful_pM': unicode_or_none(lib.successful_pM),
 543         'undiluted_concentration': unicode_or_none(lib.undiluted_concentration)
 544         }
 545     if lib.library_type_id is None:
 546         info['library_type'] = None
 547     else:
 548         info['library_type'] = lib.library_type.name
 549     return info
 550
 551 def library_json(request, library_id):
 552     """
 553     Return a json formatted library dictionary
 554     """
 555     require_api_key(request)
 556     # what validation should we do on library_id?
 557
 558     lib = library_dict(library_id)
 559     if lib is None:
 560         raise Http404
 561
 562     lib_json = json.dumps(lib)
 563     return HttpResponse(lib_json, mimetype='application/json')
 564
 565 def species_json(request, species_id):
 566     """
 567     Return information about a species.
 568     """
 569     raise Http404
 570
 571 def species(request, species_id):
 572     species = get_object_or_404(Species, id=species_id)
 573
 574     context = RequestContext(request,
 575                              { 'species': species })
 576
 577     return render_to_response("samples/species_detail.html", context)
 578
 579 @login_required
 580 def user_profile(request):
 581     """
 582     Information about the user
 583     """
 584     context = {
 585                 'page_name': 'User Profile',
 586                 'media': '',
 587                 #'bcmagic': BarcodeMagicForm(),
 588                 #'select': 'settings',
 589             }
 590     context.update(SAMPLES_CONTEXT_DEFAULTS)
 591     return render_to_response('registration/profile.html', context,
 592                               context_instance=RequestContext(request))
 593