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