f278470734dd44f397bd51f7f850c3c2263c0fd5
[htsworkflow.git] / 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 htsworkflow.frontend.auth import require_api_key
13 from htsworkflow.frontend.experiments.models import FlowCell, Lane
14 from htsworkflow.frontend.samples.changelist import ChangeList
15 from htsworkflow.frontend.samples.models import Library, HTSUser
16 from htsworkflow.frontend.samples.results import get_flowcell_result_dict, parse_flowcell_id
17 from htsworkflow.frontend.bcmagic.forms import BarcodeMagicForm
18 from htsworkflow.pipelines.runfolder import load_pipeline_run_xml
19 from htsworkflow.pipelines import runfolder
20 from htsworkflow.pipelines.eland import ResultLane
21 from htsworkflow.frontend import settings
22 from htsworkflow.util.conversion import unicode_or_none
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
30 from django.template import RequestContext
31 from django.template.loader import get_template
32 from django.contrib.auth.decorators import login_required
33
34 LANE_LIST = [1,2,3,4,5,6,7,8]
35 SAMPLES_CONTEXT_DEFAULTS = {
36     'app_name': 'Flowcell/Library Tracker',
37     'bcmagic': BarcodeMagicForm()
38 }
39
40 def count_lanes(lane_set):
41     single = 0
42     paired = 1
43     short_read = 0
44     medium_read = 1
45     long_read = 2
46     counts = [[0,0,0,],[0,0,0]]
47     
48     for lane in lane_set.all():
49         if lane.flowcell.paired_end:
50             lane_type = paired
51         else:
52             lane_type = single
53         if lane.flowcell.read_length < 40:
54             read_type = short_read
55         elif lane.flowcell.read_length < 100:
56             read_type = medium_read
57         else:
58             read_type = long_read
59         counts[lane_type][read_type] += 1
60         
61     return counts
62
63 def create_library_context(cl):
64     """
65     Create a list of libraries that includes how many lanes were run
66     """
67     records = []
68     #for lib in library_items.object_list:
69     for lib in cl.result_list:
70        summary = {}
71        summary['library_id'] = lib.id
72        summary['library_name'] = lib.library_name
73        summary['species_name' ] = lib.library_species.scientific_name
74        if lib.amplified_from_sample is not None:
75            summary['amplified_from'] = lib.amplified_from_sample.id
76        else:
77            summary['amplified_from'] = ''
78        lanes_run = count_lanes(lib.lane_set)
79        # suppress zeros
80        for row in xrange(len(lanes_run)):
81            for col in xrange(len(lanes_run[row])):
82                if lanes_run[row][col] == 0:
83                    lanes_run[row][col] = ''
84        summary['lanes_run'] = lanes_run
85        summary['is_archived'] = lib.is_archived()
86        records.append(summary)
87     cl.result_count = unicode(cl.paginator._count) + u" libraries"
88     return {'library_list': records }
89
90 def library(request):
91    # build changelist
92     fcl = ChangeList(request, Library,
93         list_filter=['affiliations', 'library_species'],
94         search_fields=['id', 'library_name', 'amplified_from_sample__id'],
95         list_per_page=200,
96         queryset=Library.objects.filter(hidden__exact=0)
97     )
98
99     context = { 'cl': fcl, 'title': 'Library Index'}
100     context.update(create_library_context(fcl))
101     t = get_template('samples/library_index.html')
102     c = RequestContext(request, context)
103     
104     app_context = {
105         'page_name': 'Library Index',
106         'east_region_config_div': 'changelist-filter',
107         'body': t.render(c)
108     }
109     app_context.update(SAMPLES_CONTEXT_DEFAULTS)
110     
111     app_t = get_template('flowcell_libraries_app.html')
112     app_c = RequestContext(request, app_context)
113     return HttpResponse( app_t.render(app_c) )
114
115 def library_to_flowcells(request, lib_id):
116     """
117     Display information about all the flowcells a library has been run on.
118     """
119     
120     try:
121       lib = Library.objects.get(id=lib_id)
122     except:
123       return HttpResponse("Library %s does not exist" % (lib_id))
124    
125     flowcell_list = []
126     flowcell_run_results = {} # aka flowcells we're looking at
127     for lane in lib.lane_set.all():
128         fc = lane.flowcell
129         flowcell_id, id = parse_flowcell_id(fc.flowcell_id)
130         if flowcell_id not in flowcell_run_results:
131             flowcell_run_results[flowcell_id] = get_flowcell_result_dict(flowcell_id)
132         flowcell_list.append((fc.flowcell_id, lane.lane_number))
133
134     flowcell_list.sort()
135     lane_summary_list = []
136     eland_results = []
137     for fc, lane_number in flowcell_list:
138         lane_summary, err_list = _summary_stats(fc, lane_number)
139
140         eland_results.extend(_make_eland_results(fc, lane_number, flowcell_run_results))
141         lane_summary_list.extend(lane_summary)
142
143     context = {
144         'page_name': 'Library Details',
145         'lib': lib,
146         'eland_results': eland_results,
147         'lane_summary_list': lane_summary_list,
148     }
149     context.update(SAMPLES_CONTEXT_DEFAULTS)
150
151     return render_to_response(
152         'samples/library_detail.html',
153         context,
154         context_instance = RequestContext(request))
155
156 def lanes_for(request, username=None):
157     """
158     Generate a report of recent activity for a user
159     """
160     query = {}
161     if username is not None:
162         user = HTSUser.objects.get(username=username)
163         query.update({'library__affiliations__users__id':user.id})
164     fcl = ChangeList(request, Lane,
165         list_filter=[],
166         search_fields=['flowcell__flowcell_id', 'library__id', 'library__library_name'],
167         list_per_page=200,
168         queryset=Lane.objects.filter(**query)
169     )
170
171     context = { 'lanes': fcl, 'title': 'Lane Index'}
172
173     return render_to_response(
174         'samples/lanes_for.html',
175         context,
176         context_instance = RequestContext(request)
177     )
178           
179     
180 def summaryhtm_fc_cnm(request, flowcell_id, cnm):
181     """
182     returns a Summary.htm file if it exists.
183     """
184     fc_id, status = parse_flowcell_id(flowcell_id)
185     d = get_flowcell_result_dict(fc_id)
186     
187     if d is None:
188         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
189     
190     if cnm not in d:
191         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
192     
193     summary_filepath = d[cnm]['summary']
194     
195     if summary_filepath is None:
196         return HttpResponse('<b>Summary.htm for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
197     
198     f = open(summary_filepath, 'r')
199     
200     return HttpResponse(f)
201
202
203 def result_fc_cnm_eland_lane(request, flowcell_id, cnm, lane):
204     """
205     returns an eland_file upon calling.
206     """
207     fc_id, status = parse_flowcell_id(flowcell_id)
208     d = get_flowcell_result_dict(fc_id)
209     
210     if d is None:
211         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
212     
213     if cnm not in d:
214         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
215     
216     erd = d[cnm]['eland_results']
217     lane = int(lane)
218     
219     if lane not in erd:
220         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
221     
222     filepath = erd[lane]
223     
224     #f = opener.autoopen(filepath, 'r')
225     # return HttpResponse(f, mimetype="application/x-elandresult")
226
227     f = open(filepath, 'r')
228     return HttpResponse(f, mimetype='application/x-bzip2')
229     
230
231
232 def bedfile_fc_cnm_eland_lane_ucsc(request, fc_id, cnm, lane):
233     """
234     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane (ucsc compatible)
235     """
236     return bedfile_fc_cnm_eland_lane(request, fc_id, cnm, lane, ucsc_compatible=True)
237
238
239 def bedfile_fc_cnm_eland_lane(request, flowcell_id, cnm, lane, ucsc_compatible=False):
240     """
241     returns a bed file for a given flowcell, CN-M (i.e. C1-33), and lane
242     """
243     fc_id, status = parse_flowcell_id(flowcell_id)
244     d = get_flowcell_result_dict(fc_id)
245     
246     if d is None:
247         return HttpResponse('<b>Results for Flowcell %s not found.</b>' % (fc_id))
248     
249     if cnm not in d:
250         return HttpResponse('<b>Results for Flowcell %s; %s not found.</b>' % (fc_id, cnm))
251     
252     erd = d[cnm]['eland_results']
253     lane = int(lane)
254     
255     if lane not in erd:
256         return HttpResponse('<b>Results for Flowcell %s; %s; lane %s not found.</b>' % (fc_id, cnm, lane))
257     
258     filepath = erd[lane]
259     
260     # Eland result file
261     fi = opener.autoopen(filepath, 'r')
262     # output memory file
263     
264     name, description = makebed.make_description( fc_id, lane )
265     
266     bedgen = makebed.make_bed_from_eland_generator(fi, name, description)
267     
268     if ucsc_compatible:
269         return HttpResponse(bedgen)
270     else:
271         return HttpResponse(bedgen, mimetype="application/x-bedfile")
272
273
274 def _summary_stats(flowcell_id, lane_id):
275     """
276     Return the summary statistics for a given flowcell, lane, and end.
277     """
278     fc_id, status = parse_flowcell_id(flowcell_id)
279     fc_result_dict = get_flowcell_result_dict(fc_id)
280
281     summary_list = []
282     err_list = []
283     
284     if fc_result_dict is None:
285         err_list.append('Results for Flowcell %s not found.' % (fc_id))
286         return (summary_list, err_list)
287
288     for cycle_width in fc_result_dict:
289         xmlpath = fc_result_dict[cycle_width]['run_xml']
290         
291         if xmlpath is None:
292             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cycle_width))
293             continue
294         
295         run = load_pipeline_run_xml(xmlpath)
296         gerald_summary = run.gerald.summary.lane_results
297         for end in range(len(gerald_summary)):
298             end_summary = run.gerald.eland_results.results[end]
299             if end_summary.has_key(lane_id):
300                 eland_summary = run.gerald.eland_results.results[end][lane_id]
301             else:
302                 eland_summary = ResultLane(lane_id=lane_id, end=end)
303             # add information to lane_summary
304             eland_summary.flowcell_id = flowcell_id
305             if len(gerald_summary) > end and gerald_summary[end].has_key(lane_id):
306                 eland_summary.clusters = gerald_summary[end][lane_id].cluster
307             else:
308                 eland_summary.clusters = None
309             eland_summary.cycle_width = cycle_width
310             if hasattr(eland_summary, 'genome_map'):
311                 eland_summary.summarized_reads = runfolder.summarize_mapped_reads( 
312                                                    eland_summary.genome_map, 
313                                                    eland_summary.mapped_reads)
314
315             # grab some more information out of the flowcell db
316             flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
317             #pm_field = 'lane_%d_pM' % (lane_id)
318             lane_obj = flowcell.lane_set.get(lane_number=lane_id)
319             eland_summary.successful_pm = lane_obj.pM
320
321             summary_list.append(eland_summary)
322
323         #except Exception, e:
324         #    summary_list.append("Summary report needs to be updated.")
325         #    logging.error("Exception: " + str(e))
326     
327     return (summary_list, err_list)
328
329 def _summary_stats_old(flowcell_id, lane):
330     """
331     return a dictionary of summary stats for a given flowcell_id & lane.
332     """
333     fc_id, status = parse_flowcell_id(flowcell_id)
334     fc_result_dict = get_flowcell_result_dict(fc_id)
335     
336     dict_list = []
337     err_list = []
338     summary_list = []
339     
340     if fc_result_dict is None:
341         err_list.append('Results for Flowcell %s not found.' % (fc_id))
342         return (dict_list, err_list, summary_list)
343     
344     for cnm in fc_result_dict:
345     
346         xmlpath = fc_result_dict[cnm]['run_xml']
347         
348         if xmlpath is None:
349             err_list.append('Run xml for Flowcell %s(%s) not found.' % (fc_id, cnm))
350             continue
351         
352         tree = ElementTree.parse(xmlpath).getroot()
353         results = runfolder.PipelineRun(pathname='', xml=tree)
354         try:
355             lane_report = runfolder.summarize_lane(results.gerald, lane)
356             summary_list.append(os.linesep.join(lane_report))
357         except Exception, e:
358             summary_list.append("Summary report needs to be updated.")
359             logging.error("Exception: " + str(e))
360        
361         print >>sys.stderr, "----------------------------------"
362         print >>sys.stderr, "-- DOES NOT SUPPORT PAIRED END ---"
363         print >>sys.stderr, "----------------------------------"
364         lane_results = results.gerald.summary[0][lane]
365         lrs = lane_results
366         
367         d = {}
368         
369         d['average_alignment_score'] = lrs.average_alignment_score
370         d['average_first_cycle_intensity'] = lrs.average_first_cycle_intensity
371         d['cluster'] = lrs.cluster
372         d['lane'] = lrs.lane
373         d['flowcell'] = flowcell_id
374         d['cnm'] = cnm
375         d['percent_error_rate'] = lrs.percent_error_rate
376         d['percent_intensity_after_20_cycles'] = lrs.percent_intensity_after_20_cycles
377         d['percent_pass_filter_align'] = lrs.percent_pass_filter_align
378         d['percent_pass_filter_clusters'] = lrs.percent_pass_filter_clusters
379         
380         #FIXME: function finished, but need to take advantage of
381         #   may need to take in a list of lanes so we only have to
382         #   load the xml file once per flowcell rather than once
383         #   per lane.
384         dict_list.append(d)
385     
386     return (dict_list, err_list, summary_list)
387     
388     
389 def get_eland_result_type(pathname):
390     """
391     Guess the eland result file type from the filename
392     """
393     path, filename = os.path.split(pathname)
394     if 'extended' in filename:
395         return 'extended'
396     elif 'multi' in filename:
397         return 'multi'
398     elif 'result' in filename:
399         return 'result'
400     else:
401         return 'unknown'
402
403 def _make_eland_results(flowcell_id, lane, interesting_flowcells):
404     fc_id, status = parse_flowcell_id(flowcell_id)
405     cur_fc = interesting_flowcells.get(fc_id, None)
406     if cur_fc is None:
407       return []
408
409     flowcell = FlowCell.objects.get(flowcell_id=flowcell_id)
410     # Loop throw storage devices if a result has been archived
411     storage_id_list = []
412     if cur_fc is not None:
413         for lts in flowcell.longtermstorage_set.all():
414             for sd in lts.storage_devices.all():
415                 # Use barcode_id if it exists
416                 if sd.barcode_id is not None and sd.barcode_id != '':
417                     storage_id_list.append(sd.barcode_id)
418                 # Otherwise use UUID
419                 else:
420                     storage_id_list.append(sd.uuid)
421         
422     # Formatting for template use
423     if len(storage_id_list) == 0:
424         storage_ids = None
425     else:
426         storage_ids = ', '.join([ '<a href="/inventory/%s/">%s</a>' % (s,s) for s in storage_id_list ])
427
428     results = []
429     for cycle in cur_fc.keys():
430         result_path = cur_fc[cycle]['eland_results'].get(lane, None)
431         result_link = make_result_link(fc_id, cycle, lane, result_path)
432         results.append({'flowcell_id': fc_id,
433                         'run_date': flowcell.run_date,
434                         'cycle': cycle, 
435                         'lane': lane, 
436                         'summary_url': make_summary_url(flowcell_id, cycle),
437                         'result_url': result_link[0],
438                         'result_label': result_link[1],
439                         'bed_url': result_link[2],
440                         'storage_ids': storage_ids
441         })
442     return results
443
444 def make_summary_url(flowcell_id, cycle_name):
445     url = '/results/%s/%s/summary/' % (flowcell_id, cycle_name)
446     return url
447
448 def make_result_link(flowcell_id, cycle_name, lane, eland_result_path):
449     if eland_result_path is None:
450         return ("", "", "")
451
452     result_type = get_eland_result_type(eland_result_path)
453     result_url = '/results/%s/%s/eland_result/%s' % (flowcell_id, cycle_name, lane)
454     result_label = 'eland %s' % (result_type,)
455     bed_url = None
456     if result_type == 'result':
457        bed_url_pattern = '/results/%s/%s/bedfile/%s'
458        bed_url = bed_url_pattern % (flowcell_id, cycle_name, lane)
459     
460     return (result_url, result_label, bed_url)
461
462 def _files(flowcell_id, lane):
463     """
464     Sets up available files for download
465     """
466     lane = int(lane)
467
468     flowcell_id, id = parse_flowcell_id(flowcell_id)
469     d = get_flowcell_result_dict(flowcell_id)
470     
471     if d is None:
472         return ''
473     
474     output = []
475     
476     # c_name == 'CN-M' (i.e. C1-33)
477     for c_name in d:
478         
479         if d[c_name]['summary'] is not None:
480             output.append('<a href="/results/%s/%s/summary/">summary(%s)</a>' \
481                           % (flowcell_id, c_name, c_name))
482         
483         erd = d[c_name]['eland_results']
484         if lane in erd:
485             result_type = get_eland_result_type(erd[lane])
486             result_url_pattern = '<a href="/results/%s/%s/eland_result/%s">eland %s(%s)</a>'
487             output.append(result_url_pattern % (flowcell_id, c_name, lane, result_type, c_name))
488             if result_type == 'result':
489                 bed_url_pattern = '<a href="/results/%s/%s/bedfile/%s">bedfile(%s)</a>'
490                 output.append(bed_url_pattern % (flowcell_id, c_name, lane, c_name))
491     
492     if len(output) == 0:
493         return ''
494     
495     return '(' + '|'.join(output) + ')'
496
497 def library_id_to_admin_url(request, lib_id):
498     lib = Library.objects.get(id=lib_id)
499     return HttpResponseRedirect('/admin/samples/library/%s' % (lib.id,))
500
501 def library_dict(library_id):
502     """
503     Given a library id construct a dictionary containing important information
504     return None if nothing was found
505     """
506     try:
507         lib = Library.objects.get(id = library_id)
508     except Library.DoesNotExist, e:
509         return None
510
511     #lane_info = lane_information(lib.lane_set)
512     lane_info = []
513     for lane in lib.lane_set.all():
514         lane_info.append( {'flowcell':lane.flowcell.flowcell_id,
515                            'lane_number': lane.lane_number} )
516         
517     info = {
518         # 'affiliations'?
519         # 'aligned_reads': lib.aligned_reads,
520         #'amplified_into_sample': lib.amplified_into_sample, # into is a colleciton...
521         #'amplified_from_sample_id': lib.amplified_from_sample, 
522         #'antibody_name': lib.antibody_name(), # we have no antibodies.
523         'antibody_id': lib.antibody_id,
524         'avg_lib_size': lib.avg_lib_size,
525         'cell_line_id': lib.cell_line_id,
526         'cell_line': unicode_or_none(lib.cell_line),
527         'experiment_type': lib.experiment_type.name,
528         'experiment_type_id': lib.experiment_type_id,
529         'id': lib.id,
530         'lane_set': lane_info,
531         'library_id': lib.id,
532         'library_name': lib.library_name,
533         'library_species': lib.library_species.scientific_name,
534         'library_species_id': lib.library_species_id,
535         #'library_type': lib.library_type.name,
536         'library_type_id': lib.library_type_id,
537         'made_for': lib.made_for,
538         'made_by': lib.made_by,
539         'notes': lib.notes,
540         'replicate': lib.replicate,
541         'stopping_point': lib.stopping_point,
542         'successful_pM': unicode_or_none(lib.successful_pM),
543         'undiluted_concentration': unicode_or_none(lib.undiluted_concentration)
544         }
545     if lib.library_type_id is None:
546         info['library_type'] = None
547     else:
548         info['library_type'] = lib.library_type.name
549     return info
550
551 def library_json(request, library_id):
552     """
553     Return a json formatted library dictionary
554     """
555     require_api_key(request)
556     # what validation should we do on library_id?
557     
558     lib = library_dict(library_id)
559     if lib is None:
560         raise Http404
561
562     lib_json = json.dumps(lib)
563     return HttpResponse(lib_json, mimetype='application/json')
564
565 def species_json(request, species_id):
566     """
567     Return information about a species.
568     """
569     raise Http404
570
571 @login_required
572 def user_profile(request):
573     """
574     Information about the user
575     """
576     context = {
577                 'page_name': 'User Profile',
578                 'media': '',
579                 #'bcmagic': BarcodeMagicForm(),
580                 #'select': 'settings',
581             }
582     context.update(SAMPLES_CONTEXT_DEFAULTS)
583     return render_to_response('registration/profile.html', context,
584                               context_instance=RequestContext(request))
585