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