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