htsworkflow/frontend/samples/views.py

   1 # Create your views here.
   2 from htsworkflow.frontend.experiments.models import FlowCell
   3 from htsworkflow.frontend.samples.changelist import ChangeList
   4 from htsworkflow.frontend.samples.models import Library
   5 from htsworkflow.frontend.samples.results import get_flowcell_result_dict, parse_flowcell_id
   6 from htsworkflow.pipelines.runfolder import load_pipeline_run_xml
   7 from htsworkflow.pipelines import runfolder
   8 from htsworkflow.frontend import settings
   9 from htsworkflow.util import makebed
  10 from htsworkflow.util import opener
  11
  12 from django.core.exceptions import ObjectDoesNotExist
  13 from django.http import HttpResponse, HttpResponseRedirect
  14 from django.shortcuts import render_to_response
  15 from django.template import RequestContext
  16 from django.template.loader import get_template
  17 from django.contrib.auth.decorators import login_required
  18
  19 import StringIO
  20 import logging
  21 import os
  22
  23 LANE_LIST = [1,2,3,4,5,6,7,8]
  24 SAMPLES_CONTEXT_DEFAULTS = {
  25     'app_name': 'Flowcell/Library Tracker'
  26 }
  27
  28 def create_library_context(cl):
  29     """
  30     Create a list of libraries that includes how many lanes were run
  31     """
  32     records = []
  33     #for lib in library_items.object_list:
  34     for lib in cl.result_list:
  35        summary = {}
  36        summary['library_id'] = lib.library_id
  37        summary['library_name'] = lib.library_name
  38        summary['species_name' ] = lib.library_species.scientific_name
  39        if lib.amplified_from_sample is not None:
  40            summary['amplified_from'] = lib.amplified_from_sample.library_id
  41        else:
  42            summary['amplified_from'] = ''
  43        lanes_run = 0
  44        for lane_id in LANE_LIST:
  45            lane = getattr(lib, 'lane_%d_library' % (lane_id,))
  46            lanes_run += len( lane.all() )
  47        summary['lanes_run'] = lanes_run
  48        summary['is_archived'] = lib.is_archived()
  49        records.append(summary)
  50     cl.result_count = unicode(cl.paginator._count) + u" libraries"
  51     return {'library_list': records }
  52
  53 def library(request):
  54    # build changelist
  55     fcl = ChangeList(request, Library,
  56         list_filter=['affiliations', 'library_species'],
  57         search_fields=['library_id', 'library_name', 'amplified_from_sample__library_id'],
  58         list_per_page=200,
  59         queryset=Library.objects.filter(hidden__exact=0)
  60     )
  61
  62     context = { 'cl': fcl, 'title': 'Library Index'}
  63     context.update(create_library_context(fcl))
  64     t = get_template('samples/library_index.html')
  65     c = RequestContext(request, context)
  66
  67     app_context = {
  68         'page_name': 'Library Index',
  69         'body': t.render(c)
  70     }
  71     app_context.update(SAMPLES_CONTEXT_DEFAULTS)
  72
  73     app_t = get_template('flowcell_libraries_app.html')
  74     app_c = RequestContext(request, app_context)
  75     return HttpResponse( app_t.render(app_c) )
  76
  77 def library_to_flowcells(request, lib_id):
  78     """
  79     Display information about all the flowcells a library has been run on.
  80     """
  81
  82     try:
  83       lib = Library.objects.get(library_id=lib_id)
  84     except:
  85       return HttpResponse("Library %s does not exist" % (lib_id))
  86
  87     flowcell_list = []
  88     interesting_flowcells = {} # aka flowcells we're looking at
  89     for lane in LANE_LIST:
  90         lane_library = getattr(lib, 'lane_%d_library' % (lane,))
  91         for fc in lane_library.all():
  92             flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
  93             if flowcell_id not in interesting_flowcells:
  94                 interesting_flowcells[flowcell_id] = get_flowcell_result_dict(flowcell_id)
  95             flowcell_list.append((fc.flowcell_id, lane))
  96
  97     flowcell_list.sort()
  98
  99     lane_summary_list = []
 100     eland_results = []
 101     for fc, lane in flowcell_list:
 102         lane_summary, err_list = _summary_stats(fc, lane)
 103
 104         eland_results.extend(_make_eland_results(fc, lane, interesting_flowcells))
 105         lane_summary_list.extend(lane_summary)
 106
 107     return render_to_response(
 108         'samples/library_detail.html',
 109         {'lib': lib,
 110          'eland_results': eland_results,
 111          'lane_summary_list': lane_summary_list,
 112         },
 113         context_instance = RequestContext(request))
 114
 115 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
 116     """
 117     returns a Summary.htm file if it exists.
 118     """
 119     fc_id, status = parse_flowcell_id(flowcell_id)
 120     d = get_flowcell_result_dict(fc_id)
 121
 122     if d is None:
 123         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 124
 125     if cnm not in d:
 126         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 127
 128     summary_filepath = d[cnm]['summary']
 129
 130     if summary_filepath is None:
 131         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 132
 133     f = open(summary_filepath, 'r')
 134
 135     return HttpResponse(f)
 136
 137
 138 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
 139     """
 140     returns an eland_file upon calling.
 141     """
 142     fc_id, status = parse_flowcell_id(flowcell_id)
 143     d = get_flowcell_result_dict(fc_id)
 144
 145     if d is None:
 146         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 147
 148     if cnm not in d:
 149         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 150
 151     erd = d[cnm]['eland_results']
 152     lane = int(lane)
 153
 154     if lane not in erd:
 155         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 156
 157     filepath = erd[lane]
 158
 159     #f = opener.autoopen(filepath, 'r')
 160     # return HttpResponse(f, mimetype="application/x-elandresult")
 161
 162     f = open(filepath, 'r')
 163     return HttpResponse(f, mimetype='application/x-bzip2')
 164
 165
 166
 167 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
 168     """
 169     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
 170     """
 171     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
 172
 173
 174 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
 175     """
 176     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
 177     """
 178     fc_id, status = parse_flowcell_id(flowcell_id)
 179     d = get_flowcell_result_dict(fc_id)
 180
 181     if d is None:
 182         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
 183
 184     if cnm not in d:
 185         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
 186
 187     erd = d[cnm]['eland_results']
 188     lane = int(lane)
 189
 190     if lane not in erd:
 191         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
 192
 193     filepath = erd[lane]
 194
 195     # Eland result file
 196     fi = opener.autoopen(filepath, 'r')
 197     # output memory file
 198
 199     name, description = makebed.make_description( fc_id, lane )
 200
 201     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
 202
 203     if ucsc_compatible:
 204         return HttpResponse(bedgen)
 205     else:
 206         return HttpResponse(bedgen, mimetype="application/x-bedfile")
 207
 208
 209 def _summary_stats(flowcell_id, lane_id):
 210     """
 211     Return the summary statistics for a given flowcell, lane, and end.
 212     """
 213     fc_id, status = parse_flowcell_id(flowcell_id)
 214     fc_result_dict = get_flowcell_result_dict(fc_id)
 215
 216     summary_list = []
 217     err_list = []
 218
 219     if fc_result_dict is None:
 220         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 221         return (summary_list, err_list)
 222
 223     for cycle_width in fc_result_dict:
 224         xmlpath = fc_result_dict[cycle_width]['run_xml']
 225
 226         if xmlpath is None:
 227             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
 228             continue
 229
 230         run = load_pipeline_run_xml(xmlpath)
 231         gerald_summary = run.gerald.summary.lane_results
 232         for end in range(len(gerald_summary)):
 233             eland_summary = run.gerald.eland_results.results[end][lane_id]
 234             # add information to lane_summary
 235             eland_summary.flowcell_id = flowcell_id
 236             eland_summary.clusters = gerald_summary[end][lane_id].cluster
 237             eland_summary.cycle_width = cycle_width
 238             if hasattr(eland_summary, 'genome_map'):
 239                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads(
 240                                                    eland_summary.genome_map,
 241                                                    eland_summary.mapped_reads)
 242
 243             # grab some more information out of the flowcell db
 244             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 245             pm_field = 'lane_%d_pM' % (lane_id)
 246             eland_summary.successful_pm = getattr(flowcell, pm_field)
 247
 248             summary_list.append(eland_summary)
 249
 250         #except Exception, e:
 251         #    summary_list.append("Summary report needs to be updated.")
 252         #    logging.error("Exception: " + str(e))
 253
 254     return (summary_list, err_list)
 255
 256 def _summary_stats_old(flowcell_id, lane):
 257     """
 258     return a dictionary of summary stats for a given flowcell_id & lane.
 259     """
 260     fc_id, status = parse_flowcell_id(flowcell_id)
 261     fc_result_dict = get_flowcell_result_dict(fc_id)
 262
 263     dict_list = []
 264     err_list = []
 265     summary_list = []
 266
 267     if fc_result_dict is None:
 268         err_list.append('Results for Flowcell %s not found.' % (fc_id))
 269         return (dict_list, err_list, summary_list)
 270
 271     for cnm in fc_result_dict:
 272
 273         xmlpath = fc_result_dict[cnm]['run_xml']
 274
 275         if xmlpath is None:
 276             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cnm))
 277             continue
 278
 279         tree = ElementTree.parse(xmlpath).getroot()
 280         results = runfolder.PipelineRun(pathname='', xml=tree)
 281         try:
 282             lane_report = runfolder.summarize_lane(results.gerald, lane)
 283             summary_list.append(os.linesep.join(lane_report))
 284         except Exception, e:
 285             summary_list.append("Summary report needs to be updated.")
 286             logging.error("Exception: " + str(e))
 287
 288         print "----------------------------------"
 289         print "-- DOES NOT SUPPORT PAIRED END ---"
 290         print "----------------------------------"
 291         lane_results = results.gerald.summary[0][lane]
 292         lrs = lane_results
 293
 294         d = {}
 295
 296         d['average_alignment_score'] = lrs.average_alignment_score
 297         d['average_first_cycle_intensity'] = lrs.average_first_cycle_intensity
 298         d['cluster'] = lrs.cluster
 299         d['lane'] = lrs.lane
 300         d['flowcell'] = flowcell_id
 301         d['cnm'] = cnm
 302         d['percent_error_rate'] = lrs.percent_error_rate
 303         d['percent_intensity_after_20_cycles'] = lrs.percent_intensity_after_20_cycles
 304         d['percent_pass_filter_align'] = lrs.percent_pass_filter_align
 305         d['percent_pass_filter_clusters'] = lrs.percent_pass_filter_clusters
 306
 307         #FIXME: function finished, but need to take advantage of
 308         #   may need to take in a list of lanes so we only have to
 309         #   load the xml file once per flowcell rather than once
 310         #   per lane.
 311         dict_list.append(d)
 312
 313     return (dict_list, err_list, summary_list)
 314
 315
 316 def get_eland_result_type(pathname):
 317     """
 318     Guess the eland result file type from the filename
 319     """
 320     path, filename = os.path.split(pathname)
 321     if 'extended' in filename:
 322         return 'extended'
 323     elif 'multi' in filename:
 324         return 'multi'
 325     elif 'result' in filename:
 326         return 'result'
 327     else:
 328         return 'unknown'
 329
 330 def _make_eland_results(flowcell_id, lane, interesting_flowcells):
 331     fc_id, status = parse_flowcell_id(flowcell_id)
 332     cur_fc = interesting_flowcells.get(fc_id, None)
 333     if cur_fc is None:
 334       return []
 335
 336     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
 337     # Loop throw storage devices if a result has been archived
 338     storage_id_list = []
 339     if cur_fc is not None:
 340         for lts in flowcell.longtermstorage_set.all():
 341             for sd in lts.storage_devices.all():
 342                 # Use barcode_id if it exists
 343                 if sd.barcode_id is not None and sd.barcode_id != '':
 344                     storage_id_list.append(sd.barcode_id)
 345                 # Otherwise use UUID
 346                 else:
 347                     storage_id_list.append(sd.uuid)
 348
 349     # Formatting for template use
 350     if len(storage_id_list) == 0:
 351         storage_ids = None
 352     else:
 353         storage_ids = ', '.join(storage_id_list)
 354
 355     results = []
 356     for cycle in cur_fc.keys():
 357         result_path = cur_fc[cycle]['eland_results'].get(lane, None)
 358         result_link = make_result_link(fc_id, cycle, lane, result_path)
 359         results.append({'flowcell_id': fc_id,
 360                         'cycle': cycle,
 361                         'lane': lane,
 362                         'summary_url': make_summary_url(flowcell_id, cycle),
 363                         'result_url': result_link[0],
 364                         'result_label': result_link[1],
 365                         'bed_url': result_link[2],
 366                         'storage_ids': storage_ids
 367         })
 368     return results
 369
 370 def make_summary_url(flowcell_id, cycle_name):
 371     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
 372     return url
 373
 374 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
 375     if eland_result_path is None:
 376         return ("", "", "")
 377
 378     result_type = get_eland_result_type(eland_result_path)
 379     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
 380     result_label = 'eland %s' % (result_type,)
 381     bed_url = None
 382     if result_type == 'result':
 383        bed_url_pattern = '/results/%s/%s/bedfile/%s'
 384        bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
 385
 386     return (result_url, result_label, bed_url)
 387
 388 def _files(flowcell_id, lane):
 389     """
 390     Sets up available files for download
 391     """
 392     lane = int(lane)
 393
 394     flowcell_id, id = parse_flowcell_id(flowcell_id)
 395     d = get_flowcell_result_dict(flowcell_id)
 396
 397     if d is None:
 398         return ''
 399
 400     output = []
 401
 402     # c_name == 'CN-M' (i.e. C1-33)
 403     for c_name in d:
 404
 405         if d[c_name]['summary'] is not None:
 406             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' \
 407                           % (flowcell_id, c_name, c_name))
 408
 409         erd = d[c_name]['eland_results']
 410         if lane in erd:
 411             result_type = get_eland_result_type(erd[lane])
 412             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
 413             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
 414             if result_type == 'result':
 415                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
 416                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
 417
 418     if len(output) == 0:
 419         return ''
 420
 421     return '(' + '|'.join(output) + ')'
 422
 423 def library_id_to_admin_url(request, lib_id):
 424     lib = Library.objects.get(library_id=lib_id)
 425     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
 426
 427 @login_required
 428 def user_profile(request):
 429     """
 430     Information about the user
 431     """
 432     context = {
 433                 'page_name': 'User Profile',
 434                 'media': '',
 435                 #'bcmagic': BarcodeMagicForm(),
 436                 #'select': 'settings',
 437             }
 438     context.update(SAMPLES_CONTEXT_DEFAULTS)
 439     return render_to_response('registration/profile.html', context,
 440                               context_instance=RequestContext(request))