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