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