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 django.contrib.csrf.middleware import csrf_exempt
  13 from htsworkflow.frontend.auth import require_api_key
  14 from htsworkflow.frontend.experiments.models import FlowCell, Lane, LANE_STATUS_MAP
  15 from htsworkflow.frontend.samples.changelist import ChangeList
  16 from htsworkflow.frontend.samples.models import Antibody, Library, Species, HTSUser
  17 from htsworkflow.frontend.samples.results import get_flowcell_result_dict
  18 from htsworkflow.frontend.bcmagic.forms import BarcodeMagicForm
  19 from htsworkflow.pipelines.runfolder import load_pipeline_run_xml
  20 from htsworkflow.pipelines import runfolder
  21 from htsworkflow.pipelines.eland import ResultLane
  22 from htsworkflow.util.conversion import unicode_or_none, parse_flowcell_id
  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, get_object_or_404
  30 from django.template import RequestContext
  31 from django.template.loader import get_template
  32 from django.contrib.auth.decorators import login_required
  33 from django.conf import settings
  34
  35 LANE_LIST = [1,2,3,4,5,6,7,8]
  36 SAMPLES_CONTEXT_DEFAULTS = {
  37     'app_name': 'Flowcell/Library Tracker',
  38     'bcmagic': BarcodeMagicForm()
  39 }
  40
  41 def count_lanes(lane_set):
  42     single = 0
  43     paired = 1
  44     short_read = 0
  45     medium_read = 1
  46     long_read = 2
  47     counts = [[0,0,0,],[0,0,0]]
  48
  49     for lane in lane_set.all():
  50         if lane.flowcell.paired_end:
  51             lane_type = paired
  52         else:
  53             lane_type = single
  54         if lane.flowcell.read_length < 40:
  55             read_type = short_read
  56         elif lane.flowcell.read_length < 100:
  57             read_type = medium_read
  58         else:
  59             read_type = long_read
  60         counts[lane_type][read_type] += 1
  61
  62     return counts
  63
  64 def create_library_context(cl):
  65     """
  66      Create a list of libraries that includes how many lanes were run
  67     """
  68     records = []
  69     #for lib in library_items.object_list:
  70     for lib in cl.result_list:
  71        summary = {}
  72        summary['library'] = lib
  73        summary['library_id'] = lib.id
  74        summary['library_name'] = lib.library_name
  75        summary['species_name' ] = lib.library_species.scientific_name
  76        if lib.amplified_from_sample is not None:
  77            summary['amplified_from'] = lib.amplified_from_sample.id
  78        else:
  79            summary['amplified_from'] = ''
  80        lanes_run = count_lanes(lib.lane_set)
  81        # suppress zeros
  82        for row in xrange(len(lanes_run)):
  83            for col in xrange(len(lanes_run[row])):
  84                if lanes_run[row][col] == 0:
  85                    lanes_run[row][col] = ''
  86        summary['lanes_run'] = lanes_run
  87        summary['is_archived'] = lib.is_archived()
  88        records.append(summary)
  89     cl.result_count = unicode(cl.paginator._count)
  90     return {'library_list': records }
  91
  92 def library(request):
  93     # build changelist
  94     fcl = ChangeList(request, Library,
  95         list_filter=['affiliations', 'library_species'],
  96         search_fields=['id', 'library_name', 'amplified_from_sample__id'],
  97         list_per_page=200,
  98         queryset=Library.objects.filter(hidden__exact=0)
  99     )
 100
 101     context = { 'cl': fcl, 'title': 'Library Index'}
 102     context.update(create_library_context(fcl))
 103     t = get_template('samples/library_index.html')
 104     c = RequestContext(request, context)
 105     return HttpResponse( t.render(c) )
 106
 107
 108 def library_to_flowcells(request, lib_id):
 109     """
 110     Display information about all the flowcells a library has been run on.
 111     """
 112
 113     try:
 114       lib = Library.objects.get(id=lib_id)
 115     except:
 116       return HttpResponse("Library %s does not exist" % (lib_id))
 117
 118     flowcell_list = []
 119     flowcell_run_results = {} # aka flowcells we're looking at
 120     for lane in lib.lane_set.all():
 121         fc = lane.flowcell
 122         flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
 123         if flowcell_id not in flowcell_run_results:
 124             flowcell_run_results[flowcell_id] = get_flowcell_result_dict(flowcell_id)
 125         flowcell_list.append((fc.flowcell_id, lane.lane_number))
 126
 127     flowcell_list.sort()
 128     lane_summary_list = []
 129     eland_results = []
 130     for fc, lane_number in flowcell_list:
 131         lane_summary, err_list = _summary_stats(fc, lane_number)
 132         lane_summary_list.extend(lane_summary)
 133
 134         eland_results.extend(_make_eland_results(fc, lane_number, flowcell_run_results))
 135
 136     context = {
 137         'page_name': 'Library Details',
 138         'lib': lib,
 139         'eland_results': eland_results,
 140         'lane_summary_list': lane_summary_list,
 141     }
 142     context.update(SAMPLES_CONTEXT_DEFAULTS)
 143
 144     return render_to_response(
 145         'samples/library_detail.html',
 146         context,
 147         context_instance = RequestContext(request))
 148
 149 def lanes_for(request, username=None):
 150     """
 151     Generate a report of recent activity for a user
 152     """
 153     query = {}
 154     if username is not None:
 155         user = HTSUser.objects.get(username=username)
 156         query.update({'library__affiliations__users__id':user.id})
 157     fcl = ChangeList(request, Lane,
 158         list_filter=[],
 159         search_fields=['flowcell__flowcell_id', 'library__id', 'library__library_name'],
 160         list_per_page=200,
 161         queryset=Lane.objects.filter(**query)
 162     )
 163
 164     context = { 'lanes': fcl, 'title': 'Lane Index'}
 165
 166     return render_to_response(
 167         'samples/lanes_for.html',
 168         context,
 169         context_instance = RequestContext(request)
 170     )
 171
 172
 173 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
 174     """
 175     returns a Summary.htm file if it exists.
 176     """
 177     fc_id, status = parse_flowcell_id(flowcell_id)
 178     d = get_flowcell_result_dict(fc_id)
 179
 180     if d is None:
 181         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 182
 183     if cnm not in d:
 184         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 185
 186     summary_filepath = d[cnm]['summary']
 187
 188     if summary_filepath is None:
 189         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 190
 191     f = open(summary_filepath, 'r')
 192
 193     return HttpResponse(f)
 194
 195
 196 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
 197     """
 198     returns an eland_file upon calling.
 199     """
 200     fc_id, status = parse_flowcell_id(flowcell_id)
 201     d = get_flowcell_result_dict(fc_id)
 202
 203     if d is None:
 204         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 205
 206     if cnm not in d:
 207         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 208
 209     erd = d[cnm]['eland_results']
 210     lane = int(lane)
 211
 212     if lane not in erd:
 213         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 214
 215     filepath = erd[lane]
 216
 217     #f = opener.autoopen(filepath, 'r')
 218     # return HttpResponse(f, mimetype="application/x-elandresult")
 219
 220     f = open(filepath, 'r')
 221     return HttpResponse(f, mimetype='application/x-bzip2')
 222
 223
 224
 225 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
 226     """
 227     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
 228     """
 229     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
 230
 231
 232 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
 233     """
 234     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
 235     """
 236     fc_id, status = parse_flowcell_id(flowcell_id)
 237     d = get_flowcell_result_dict(fc_id)
 238
 239     if d is None:
 240         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 241
 242     if cnm not in d:
 243         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 244
 245     erd = d[cnm]['eland_results']
 246     lane = int(lane)
 247
 248     if lane not in erd:
 249         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 250
 251     filepath = erd[lane]
 252
 253     # Eland result file
 254     fi = opener.autoopen(filepath, 'r')
 255     # output memory file
 256
 257     name, description = makebed.make_description( fc_id, lane )
 258
 259     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
 260
 261     if ucsc_compatible:
 262         return HttpResponse(bedgen)
 263     else:
 264         return HttpResponse(bedgen, mimetype="application/x-bedfile")
 265
 266
 267 def _summary_stats(flowcell_id, lane_id):
 268     """
 269     Return the summary statistics for a given flowcell, lane, and end.
 270     """
 271     fc_id, status = parse_flowcell_id(flowcell_id)
 272     fc_result_dict = get_flowcell_result_dict(fc_id)
 273
 274     summary_list = []
 275     err_list = []
 276
 277     if fc_result_dict is None:
 278         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 279         return (summary_list, err_list)
 280
 281     for cycle_width in fc_result_dict:
 282         xmlpath = fc_result_dict[cycle_width]['run_xml']
 283
 284         if xmlpath is None:
 285             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
 286             continue
 287
 288         run = load_pipeline_run_xml(xmlpath)
 289         gerald_summary = run.gerald.summary.lane_results
 290         for end in range(len(gerald_summary)):
 291             end_summary = run.gerald.eland_results.results[end]
 292             if end_summary.has_key(lane_id):
 293                 eland_summary = run.gerald.eland_results.results[end][lane_id]
 294             else:
 295                 eland_summary = ResultLane(lane_id=lane_id, end=end)
 296             # add information to lane_summary
 297             eland_summary.flowcell_id = flowcell_id
 298             if len(gerald_summary) > end and gerald_summary[end].has_key(lane_id):
 299                 eland_summary.clusters = gerald_summary[end][lane_id].cluster
 300             else:
 301                 eland_summary.clusters = None
 302             eland_summary.cycle_width = cycle_width
 303             if hasattr(eland_summary, 'genome_map'):
 304                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads(
 305                                                    eland_summary.genome_map,
 306                                                    eland_summary.mapped_reads)
 307
 308             # grab some more information out of the flowcell db
 309             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 310             #pm_field = 'lane_%d_pM' % (lane_id)
 311             lane_obj = flowcell.lane_set.get(lane_number=lane_id)
 312             eland_summary.flowcell = flowcell
 313             eland_summary.lane = lane_obj
 314
 315             summary_list.append(eland_summary)
 316
 317         #except Exception, e:
 318         #    summary_list.append("Summary report needs to be updated.")
 319         #    logging.error("Exception: " + str(e))
 320
 321     return (summary_list, err_list)
 322
 323
 324 def get_eland_result_type(pathname):
 325     """
 326     Guess the eland result file type from the filename
 327     """
 328     path, filename = os.path.split(pathname)
 329     if 'extended' in filename:
 330         return 'extended'
 331     elif 'multi' in filename:
 332         return 'multi'
 333     elif 'result' in filename:
 334         return 'result'
 335     else:
 336         return 'unknown'
 337
 338 def _make_eland_results(flowcell_id, lane_number, interesting_flowcells):
 339     fc_id, status = parse_flowcell_id(flowcell_id)
 340     cur_fc = interesting_flowcells.get(fc_id, None)
 341     if cur_fc is None:
 342       return []
 343
 344     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 345     lane = flowcell.lane_set.get(lane_number=lane_number)
 346     # Loop throw storage devices if a result has been archived
 347     storage_id_list = []
 348     if cur_fc is not None:
 349         for lts in flowcell.longtermstorage_set.all():
 350             for sd in lts.storage_devices.all():
 351                 # Use barcode_id if it exists
 352                 if sd.barcode_id is not None and sd.barcode_id != '':
 353                     storage_id_list.append(sd.barcode_id)
 354                 # Otherwise use UUID
 355                 else:
 356                     storage_id_list.append(sd.uuid)
 357
 358     # Formatting for template use
 359     if len(storage_id_list) == 0:
 360         storage_ids = None
 361     else:
 362         storage_ids = ', '.join([ '<a href="/inventory/%s/">%s</a>' % (s,s) for s in storage_id_list ])
 363
 364     results = []
 365     for cycle in cur_fc.keys():
 366         result_path = cur_fc[cycle]['eland_results'].get(lane, None)
 367         result_link = make_result_link(fc_id, cycle, lane, result_path)
 368         results.append({'flowcell_id': fc_id,
 369                         'flowcell': flowcell,
 370                         'run_date': flowcell.run_date,
 371                         'cycle': cycle,
 372                         'lane': lane,
 373                         'summary_url': make_summary_url(flowcell_id, cycle),
 374                         'result_url': result_link[0],
 375                         'result_label': result_link[1],
 376                         'bed_url': result_link[2],
 377                         'storage_ids': storage_ids
 378         })
 379     return results
 380
 381 def make_summary_url(flowcell_id, cycle_name):
 382     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
 383     return url
 384
 385 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
 386     if eland_result_path is None:
 387         return ("", "", "")
 388
 389     result_type = get_eland_result_type(eland_result_path)
 390     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
 391     result_label = 'eland %s' % (result_type,)
 392     bed_url = None
 393     if result_type == 'result':
 394        bed_url_pattern = '/results/%s/%s/bedfile/%s'
 395        bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
 396
 397     return (result_url, result_label, bed_url)
 398
 399 def _files(flowcell_id, lane):
 400     """
 401     Sets up available files for download
 402     """
 403     lane = int(lane)
 404
 405     flowcell_id, id = parse_flowcell_id(flowcell_id)
 406     d = get_flowcell_result_dict(flowcell_id)
 407
 408     if d is None:
 409         return ''
 410
 411     output = []
 412
 413     # c_name == 'CN-M' (i.e. C1-33)
 414     for c_name in d:
 415
 416         if d[c_name]['summary'] is not None:
 417             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' \
 418                           % (flowcell_id, c_name, c_name))
 419
 420         erd = d[c_name]['eland_results']
 421         if lane in erd:
 422             result_type = get_eland_result_type(erd[lane])
 423             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
 424             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
 425             if result_type == 'result':
 426                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
 427                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
 428
 429     if len(output) == 0:
 430         return ''
 431
 432     return '(' + '|'.join(output) + ')'
 433
 434 def library_id_to_admin_url(request, lib_id):
 435     lib = Library.objects.get(id=lib_id)
 436     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
 437
 438 def library_dict(library_id):
 439     """
 440     Given a library id construct a dictionary containing important information
 441     return None if nothing was found
 442     """
 443     try:
 444         lib = Library.objects.get(id = library_id)
 445     except Library.DoesNotExist, e:
 446         return None
 447
 448     #lane_info = lane_information(lib.lane_set)
 449     lane_info = []
 450     for lane in lib.lane_set.all():
 451         lane_info.append( {'flowcell':lane.flowcell.flowcell_id,
 452                            'lane_number': lane.lane_number,
 453                            'paired_end': lane.flowcell.paired_end,
 454                            'read_length': lane.flowcell.read_length,
 455                            'status_code': lane.status,
 456                            'status': LANE_STATUS_MAP[lane.status]} )
 457
 458     info = {
 459         # 'affiliations'?
 460         # 'aligned_reads': lib.aligned_reads,
 461         #'amplified_into_sample': lib.amplified_into_sample, # into is a colleciton...
 462         #'amplified_from_sample_id': lib.amplified_from_sample,
 463         #'antibody_name': lib.antibody_name(), # we have no antibodies.
 464         'antibody_id': lib.antibody_id,
 465         'cell_line_id': lib.cell_line_id,
 466         'cell_line': unicode_or_none(lib.cell_line),
 467         'experiment_type': lib.experiment_type.name,
 468         'experiment_type_id': lib.experiment_type_id,
 469         'gel_cut_size': lib.gel_cut_size,
 470         'hidden': lib.hidden,
 471         'id': lib.id,
 472         'insert_size': lib.insert_size,
 473         'lane_set': lane_info,
 474         'library_id': lib.id,
 475         'library_name': lib.library_name,
 476         'library_species': lib.library_species.scientific_name,
 477         'library_species_id': lib.library_species_id,
 478         #'library_type': lib.library_type.name,
 479         'library_type_id': lib.library_type_id,
 480         'made_for': lib.made_for,
 481         'made_by': lib.made_by,
 482         'notes': lib.notes,
 483         'replicate': lib.replicate,
 484         'stopping_point': lib.stopping_point,
 485         'successful_pM': unicode_or_none(lib.successful_pM),
 486         'undiluted_concentration': unicode_or_none(lib.undiluted_concentration)
 487         }
 488     if lib.library_type_id is None:
 489         info['library_type'] = None
 490     else:
 491         info['library_type'] = lib.library_type.name
 492     return info
 493
 494 @csrf_exempt
 495 def library_json(request, library_id):
 496     """
 497     Return a json formatted library dictionary
 498     """
 499     require_api_key(request)
 500     # what validation should we do on library_id?
 501
 502     lib = library_dict(library_id)
 503     if lib is None:
 504         raise Http404
 505
 506     lib_json = json.dumps(lib)
 507     return HttpResponse(lib_json, mimetype='application/json')
 508
 509 @csrf_exempt
 510 def species_json(request, species_id):
 511     """
 512     Return information about a species.
 513     """
 514     raise Http404
 515
 516 def species(request, species_id):
 517     species = get_object_or_404(Species, id=species_id)
 518
 519     context = RequestContext(request,
 520                              { 'species': species })
 521
 522     return render_to_response("samples/species_detail.html", context)
 523
 524 def antibodies(request):
 525     context = RequestContext(request,
 526                              {'antibodies': Antibody.objects.order_by('antigene')})
 527     return render_to_response("samples/antibody_index.html", context)
 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))
 543
 544