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.views.decorators.csrf 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.experiments.admin import LaneOptions
  16 from htsworkflow.frontend.samples.changelist import HTSChangeList
  17 from htsworkflow.frontend.samples.models import Antibody, Library, Species, HTSUser
  18 from htsworkflow.frontend.samples.admin import LibraryOptions
  19 from htsworkflow.frontend.samples.results import get_flowcell_result_dict
  20 from htsworkflow.frontend.bcmagic.forms import BarcodeMagicForm
  21 from htsworkflow.pipelines.runfolder import load_pipeline_run_xml
  22 from htsworkflow.pipelines import runfolder
  23 from htsworkflow.pipelines.eland import ResultLane
  24 from htsworkflow.pipelines.samplekey import SampleKey
  25 from htsworkflow.util.conversion import unicode_or_none, parse_flowcell_id
  26 from htsworkflow.util import makebed
  27 from htsworkflow.util import opener
  28
  29
  30 from django.core.exceptions import ObjectDoesNotExist
  31 from django.http import HttpResponse, HttpResponseRedirect, Http404
  32 from django.shortcuts import render_to_response, get_object_or_404
  33 from django.template import RequestContext
  34 from django.template.loader import get_template
  35 from django.contrib.auth.decorators import login_required
  36 from django.conf import settings
  37
  38 LANE_LIST = [1,2,3,4,5,6,7,8]
  39 SAMPLES_CONTEXT_DEFAULTS = {
  40     'app_name': 'Flowcell/Library Tracker',
  41     'bcmagic': BarcodeMagicForm()
  42 }
  43
  44 LOGGER = logging.getLogger(__name__)
  45
  46 def count_lanes(lane_set):
  47     single = 0
  48     paired = 1
  49     short_read = 0
  50     medium_read = 1
  51     long_read = 2
  52     counts = [[0,0,0,],[0,0,0]]
  53
  54     for lane in lane_set.all():
  55         if lane.flowcell.paired_end:
  56             lane_type = paired
  57         else:
  58             lane_type = single
  59         if lane.flowcell.read_length < 40:
  60             read_type = short_read
  61         elif lane.flowcell.read_length < 100:
  62             read_type = medium_read
  63         else:
  64             read_type = long_read
  65         counts[lane_type][read_type] += 1
  66
  67     return counts
  68
  69 def create_library_context(cl):
  70     """
  71      Create a list of libraries that includes how many lanes were run
  72     """
  73     records = []
  74     #for lib in library_items.object_list:
  75     for lib in cl.result_list:
  76        summary = {}
  77        summary['library'] = lib
  78        summary['library_id'] = lib.id
  79        summary['library_name'] = lib.library_name
  80        summary['species_name' ] = lib.library_species.scientific_name
  81        if lib.amplified_from_sample is not None:
  82            summary['amplified_from'] = lib.amplified_from_sample.id
  83        else:
  84            summary['amplified_from'] = ''
  85        lanes_run = count_lanes(lib.lane_set)
  86        # suppress zeros
  87        for row in xrange(len(lanes_run)):
  88            for col in xrange(len(lanes_run[row])):
  89                if lanes_run[row][col] == 0:
  90                    lanes_run[row][col] = ''
  91        summary['lanes_run'] = lanes_run
  92        summary['is_archived'] = lib.is_archived()
  93        records.append(summary)
  94     cl.result_count = unicode(cl.paginator._count)
  95     return {'library_list': records }
  96
  97
  98 def library(request, todo_only=False):
  99     queryset = Library.objects.filter(hidden__exact=0)
 100     filters = {'hidden__exact': 0}
 101     if todo_only:
 102         filters[lane] = None
 103     # build changelist
 104     fcl = HTSChangeList(request, Library,
 105         list_filter=['affiliations', 'library_species'],
 106         search_fields=['id', 'library_name', 'amplified_from_sample__id'],
 107         list_per_page=200,
 108         model_admin=LibraryOptions(Library, None),
 109         extra_filters=filters
 110     )
 111
 112     context = { 'cl': fcl, 'title': 'Library Index', 'todo_only': todo_only}
 113     context.update(create_library_context(fcl))
 114     t = get_template('samples/library_index.html')
 115     c = RequestContext(request, context)
 116     return HttpResponse( t.render(c) )
 117
 118
 119 def library_not_run(request):
 120     return library(request, todo_only=True)
 121
 122
 123 def library_to_flowcells(request, lib_id):
 124     """
 125     Display information about all the flowcells a library has been run on.
 126     """
 127     try:
 128         lib = Library.objects.get(id=lib_id)
 129     except:
 130         raise Http404('Library %s does not exist' % (lib_id,))
 131
 132     flowcell_list = []
 133     flowcell_run_results = {} # aka flowcells we're looking at
 134     for lane in lib.lane_set.all():
 135         fc = lane.flowcell
 136         flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
 137         if flowcell_id not in flowcell_run_results:
 138             flowcell_run_results[flowcell_id] = get_flowcell_result_dict(flowcell_id)
 139         flowcell_list.append((fc.flowcell_id, lane.lane_number))
 140
 141     flowcell_list.sort()
 142     lane_summary_list = []
 143     eland_results = []
 144     for fc, lane_number in flowcell_list:
 145         lane_summary, err_list = _summary_stats(fc, lane_number, lib_id)
 146         lane_summary_list.extend(lane_summary)
 147
 148         eland_results.extend(_make_eland_results(fc, lane_number, flowcell_run_results))
 149
 150     context = {
 151         'page_name': 'Library Details',
 152         'lib': lib,
 153         'eland_results': eland_results,
 154         'lane_summary_list': lane_summary_list,
 155     }
 156     context.update(SAMPLES_CONTEXT_DEFAULTS)
 157
 158     return render_to_response(
 159         'samples/library_detail.html',
 160         context,
 161         context_instance = RequestContext(request))
 162
 163 def lanes_for(request, username=None):
 164     """
 165     Generate a report of recent activity for a user
 166     """
 167     query = {}
 168     if username is not None:
 169         user = HTSUser.objects.get(username=username)
 170         query.update({'library__affiliations__users__id':user.id})
 171     fcl = HTSChangeList(request, Lane,
 172         list_filter=[],
 173         search_fields=['flowcell__flowcell_id', 'library__id', 'library__library_name'],
 174         list_per_page=200,
 175         model_admin=LaneOptions,
 176         queryset=Lane.objects.filter(**query)
 177     )
 178
 179     context = { 'lanes': fcl, 'title': 'Lane Index'}
 180
 181     return render_to_response(
 182         'samples/lanes_for.html',
 183         context,
 184         context_instance = RequestContext(request)
 185     )
 186
 187
 188 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
 189     """
 190     returns a Summary.htm file if it exists.
 191     """
 192     fc_id, status = parse_flowcell_id(flowcell_id)
 193     d = get_flowcell_result_dict(fc_id)
 194
 195     if d is None:
 196         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 197
 198     if cnm not in d:
 199         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 200
 201     summary_filepath = d[cnm]['summary']
 202
 203     if summary_filepath is None:
 204         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 205
 206     f = open(summary_filepath, 'r')
 207
 208     return HttpResponse(f)
 209
 210
 211 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
 212     """
 213     returns an eland_file upon calling.
 214     """
 215     fc_id, status = parse_flowcell_id(flowcell_id)
 216     d = get_flowcell_result_dict(fc_id)
 217
 218     if d is None:
 219         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 220
 221     if cnm not in d:
 222         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 223
 224     erd = d[cnm]['eland_results']
 225     lane = int(lane)
 226
 227     if lane not in erd:
 228         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 229
 230     filepath = erd[lane]
 231
 232     #f = opener.autoopen(filepath, 'r')
 233     # return HttpResponse(f, mimetype="application/x-elandresult")
 234
 235     f = open(filepath, 'r')
 236     return HttpResponse(f, mimetype='application/x-bzip2')
 237
 238
 239
 240 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
 241     """
 242     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
 243     """
 244     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
 245
 246
 247 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
 248     """
 249     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
 250     """
 251     fc_id, status = parse_flowcell_id(flowcell_id)
 252     d = get_flowcell_result_dict(fc_id)
 253
 254     if d is None:
 255         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 256
 257     if cnm not in d:
 258         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 259
 260     erd = d[cnm]['eland_results']
 261     lane = int(lane)
 262
 263     if lane not in erd:
 264         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 265
 266     filepath = erd[lane]
 267
 268     # Eland result file
 269     fi = opener.autoopen(filepath, 'r')
 270     # output memory file
 271
 272     name, description = makebed.make_description( fc_id, lane )
 273
 274     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
 275
 276     if ucsc_compatible:
 277         return HttpResponse(bedgen)
 278     else:
 279         return HttpResponse(bedgen, mimetype="application/x-bedfile")
 280
 281
 282 def _summary_stats(flowcell_id, lane_id, library_id):
 283     """
 284     Return the summary statistics for a given flowcell, lane, and end.
 285     """
 286     fc_id, status = parse_flowcell_id(flowcell_id)
 287     fc_result_dict = get_flowcell_result_dict(fc_id)
 288
 289     summary_list = []
 290     err_list = []
 291
 292     if fc_result_dict is None:
 293         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 294         return (summary_list, err_list)
 295
 296     for cycle_width in fc_result_dict:
 297         xmlpath = fc_result_dict[cycle_width]['run_xml']
 298
 299         if xmlpath is None:
 300             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
 301             continue
 302
 303         run = load_pipeline_run_xml(xmlpath)
 304         gerald_summary = run.gerald.summary.lane_results
 305         key = SampleKey(lane=lane_id, sample='s')
 306         eland_results = list(run.gerald.eland_results.find_keys(key))
 307         key = SampleKey(lane=lane_id, sample=library_id)
 308         eland_results.extend(run.gerald.eland_results.find_keys(key))
 309         for key in eland_results:
 310             eland_summary = run.gerald.eland_results.results[key]
 311             # add information to lane_summary
 312             eland_summary.flowcell_id = flowcell_id
 313
 314             read = key.read-1 if key.read is not None else 0
 315             try:
 316                 eland_summary.clusters = gerald_summary[read][key.lane].cluster
 317             except (IndexError, KeyError) as e:
 318                 eland_summary.clustes = None
 319             eland_summary.cycle_width = cycle_width
 320             if hasattr(eland_summary, 'genome_map'):
 321                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads(
 322                                                    eland_summary.genome_map,
 323                                                    eland_summary.mapped_reads)
 324
 325             # grab some more information out of the flowcell db
 326             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 327             #pm_field = 'lane_%d_pM' % (lane_id)
 328             lanes = flowcell.lane_set.filter(lane_number=lane_id)
 329             eland_summary.flowcell = flowcell
 330             eland_summary.lanes = lanes
 331
 332             summary_list.append(eland_summary)
 333
 334         #except Exception, e:
 335         #    summary_list.append("Summary report needs to be updated.")
 336         #    LOGGER.error("Exception: " + str(e))
 337
 338     return (summary_list, err_list)
 339
 340
 341 def get_eland_result_type(pathname):
 342     """
 343     Guess the eland result file type from the filename
 344     """
 345     path, filename = os.path.split(pathname)
 346     if 'extended' in filename:
 347         return 'extended'
 348     elif 'multi' in filename:
 349         return 'multi'
 350     elif 'result' in filename:
 351         return 'result'
 352     else:
 353         return 'unknown'
 354
 355 def _make_eland_results(flowcell_id, lane_number, interesting_flowcells):
 356     fc_id, status = parse_flowcell_id(flowcell_id)
 357     cur_fc = interesting_flowcells.get(fc_id, None)
 358     if cur_fc is None:
 359       return []
 360
 361     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 362     lanes = flowcell.lane_set.filter(lane_number=lane_number)
 363     # Loop throw storage devices if a result has been archived
 364     storage_id_list = []
 365     if cur_fc is not None:
 366         for lts in flowcell.longtermstorage_set.all():
 367             for sd in lts.storage_devices.all():
 368                 # Use barcode_id if it exists
 369                 if sd.barcode_id is not None and sd.barcode_id != '':
 370                     storage_id_list.append(sd.barcode_id)
 371                 # Otherwise use UUID
 372                 else:
 373                     storage_id_list.append(sd.uuid)
 374
 375     # Formatting for template use
 376     if len(storage_id_list) == 0:
 377         storage_ids = None
 378     else:
 379         storage_ids = ', '.join([ '<a href="/inventory/%s/">%s</a>' % (s,s) for s in storage_id_list ])
 380
 381     results = []
 382     for cycle in cur_fc.keys():
 383         result_path = cur_fc[cycle]['eland_results'].get(lanes[0], None)
 384         result_link = make_result_link(fc_id, cycle, lanes[0], result_path)
 385         results.append({'flowcell_id': fc_id,
 386                         'flowcell': flowcell,
 387                         'run_date': flowcell.run_date,
 388                         'cycle': cycle,
 389                         'lane': lanes[0],
 390                         'summary_url': make_summary_url(flowcell_id, cycle),
 391                         'result_url': result_link[0],
 392                         'result_label': result_link[1],
 393                         'bed_url': result_link[2],
 394                         'storage_ids': storage_ids
 395         })
 396     return results
 397
 398 def make_summary_url(flowcell_id, cycle_name):
 399     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
 400     return url
 401
 402 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
 403     if eland_result_path is None:
 404         return ("", "", "")
 405
 406     result_type = get_eland_result_type(eland_result_path)
 407     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
 408     result_label = 'eland %s' % (result_type,)
 409     bed_url = None
 410     if result_type == 'result':
 411        bed_url_pattern = '/results/%s/%s/bedfile/%s'
 412        bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
 413
 414     return (result_url, result_label, bed_url)
 415
 416 def _files(flowcell_id, lane):
 417     """
 418     Sets up available files for download
 419     """
 420     lane = int(lane)
 421
 422     flowcell_id, id = parse_flowcell_id(flowcell_id)
 423     d = get_flowcell_result_dict(flowcell_id)
 424
 425     if d is None:
 426         return ''
 427
 428     output = []
 429
 430     # c_name == 'CN-M' (i.e. C1-33)
 431     for c_name in d:
 432
 433         if d[c_name]['summary'] is not None:
 434             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' \
 435                           % (flowcell_id, c_name, c_name))
 436
 437         erd = d[c_name]['eland_results']
 438         if lane in erd:
 439             result_type = get_eland_result_type(erd[lane])
 440             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
 441             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
 442             if result_type == 'result':
 443                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
 444                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
 445
 446     if len(output) == 0:
 447         return ''
 448
 449     return '(' + '|'.join(output) + ')'
 450
 451 def library_id_to_admin_url(request, lib_id):
 452     lib = Library.objects.get(id=lib_id)
 453     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
 454
 455 def library_dict(library_id):
 456     """
 457     Given a library id construct a dictionary containing important information
 458     return None if nothing was found
 459     """
 460     try:
 461         lib = Library.objects.get(id = library_id)
 462     except Library.DoesNotExist, e:
 463         return None
 464
 465     #lane_info = lane_information(lib.lane_set)
 466     lane_info = []
 467     for lane in lib.lane_set.all():
 468         lane_info.append( {'flowcell':lane.flowcell.flowcell_id,
 469                            'lane_number': lane.lane_number,
 470                            'lane_id': lane.id,
 471                            'paired_end': lane.flowcell.paired_end,
 472                            'read_length': lane.flowcell.read_length,
 473                            'status_code': lane.status,
 474                            'status': LANE_STATUS_MAP[lane.status]} )
 475
 476     info = {
 477         # 'affiliations'?
 478         # 'aligned_reads': lib.aligned_reads,
 479         #'amplified_into_sample': lib.amplified_into_sample, # into is a colleciton...
 480         #'amplified_from_sample_id': lib.amplified_from_sample,
 481         #'antibody_name': lib.antibody_name(), # we have no antibodies.
 482         'antibody_id': lib.antibody_id,
 483         'cell_line_id': lib.cell_line_id,
 484         'cell_line': unicode_or_none(lib.cell_line),
 485         'experiment_type': lib.experiment_type.name,
 486         'experiment_type_id': lib.experiment_type_id,
 487         'gel_cut_size': lib.gel_cut_size,
 488         'hidden': lib.hidden,
 489         'id': lib.id,
 490         'insert_size': lib.insert_size,
 491         'lane_set': lane_info,
 492         'library_id': lib.id,
 493         'library_name': lib.library_name,
 494         'library_species': lib.library_species.scientific_name,
 495         'library_species_id': lib.library_species_id,
 496         #'library_type': lib.library_type.name,
 497         'library_type_id': lib.library_type_id,
 498         'made_for': lib.made_for,
 499         'made_by': lib.made_by,
 500         'notes': lib.notes,
 501         'replicate': lib.replicate,
 502         'stopping_point': lib.stopping_point,
 503         'successful_pM': unicode_or_none(lib.successful_pM),
 504         'undiluted_concentration': unicode_or_none(lib.undiluted_concentration)
 505         }
 506     if lib.library_type_id is None:
 507         info['library_type'] = None
 508     else:
 509         info['library_type'] = lib.library_type.name
 510     return info
 511
 512 @csrf_exempt
 513 def library_json(request, library_id):
 514     """
 515     Return a json formatted library dictionary
 516     """
 517     require_api_key(request)
 518     # what validation should we do on library_id?
 519
 520     lib = library_dict(library_id)
 521     if lib is None:
 522         raise Http404
 523
 524     lib_json = json.dumps(lib)
 525     return HttpResponse(lib_json, mimetype='application/json')
 526
 527 @csrf_exempt
 528 def species_json(request, species_id):
 529     """
 530     Return information about a species.
 531     """
 532     raise Http404
 533
 534 def species(request, species_id):
 535     species = get_object_or_404(Species, id=species_id)
 536
 537     context = RequestContext(request,
 538                              { 'species': species })
 539
 540     return render_to_response("samples/species_detail.html", context)
 541
 542 def antibodies(request):
 543     context = RequestContext(request,
 544                              {'antibodies': Antibody.objects.order_by('antigene')})
 545     return render_to_response("samples/antibody_index.html", context)
 546
 547 @login_required
 548 def user_profile(request):
 549     """
 550     Information about the user
 551     """
 552     context = {
 553                 'page_name': 'User Profile',
 554                 'media': '',
 555                 #'bcmagic': BarcodeMagicForm(),
 556                 #'select': 'settings',
 557             }
 558     context.update(SAMPLES_CONTEXT_DEFAULTS)
 559     return render_to_response('registration/profile.html', context,
 560                               context_instance=RequestContext(request))