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